A polyether sulfone nanoscale porous filter membrane, a preparation method and application thereof

Abstract

The application relates to a polyether sulfone nanoscale porous filter membrane and a preparation method and application thereof, the filter membrane comprising a main body, the main body having non-directional winding passages, the main body comprising a pre-filter layer and a separation layer for intercepting impurities, the thickness of the separation layer being 0.4-13 mu m; the separation layer comprising separation fibers forming a porous structure, the SEM average diameter of the separation fibers being 15-55 nm; the PMI average pore diameter of the filter membrane being 10-80 nm, and the porosity being 40%-80%; the fiber density of the filter membrane being 1.1-1.5 g / cm 3 On the basis of having nanoscale pores, the filter membrane has excellent mechanical strength and high flux under the synergistic action of a thin separation layer, high fiber density, suitable fiber thickness and high porosity, and is suitable for nanoscale ultra-pure filtration in a moderate system in a semiconductor process.

Description

Technical Field

[0001] This application relates to the field of membrane materials technology, and in particular to a polyethersulfone nanoscale porous filter membrane, its preparation method and application. Background Technology

[0002] Semiconductor wet processing is a core step in integrated circuit manufacturing, widely used in critical process steps such as wafer surface cleaning, precision etching, photoresist removal, and doping. Filtration technology plays an indispensable role in this process, its core function being to ensure the ultra-high purity of the process medium and prevent potential damage to the wafer from contaminants at the source.

[0003] In wet manufacturing processes, wafer surfaces are extremely sensitive to contaminants. Even nanoscale particles, organic residues, or metal ions can cause device defects and abnormal electrical performance, directly impacting the final chip quality and yield. To mitigate these risks, the process must strictly utilize high-purity chemical reagents (such as hydrofluoric acid HF), ultrapure water, and deionized water as the base medium. The filtration system further ensures the stability and consistency of the process environment by precisely intercepting suspended particles, soluble organic matter, colloidal impurities, and trace metal ions in the liquid medium, ultimately providing crucial support for improved chip yield and performance optimization.

[0004] In actual semiconductor wet manufacturing processes, a large amount of chemical reagents are used for wafer cleaning, etching, and other operations. These reagents may contain nanoscale particulate impurities (e.g., tiny impurity particles with a diameter of 1-10 nm). Polyethersulfone (PES) filter membranes often possess good thermal and chemical stability, and through their appropriately sized pores, they can effectively remove these particles, preventing their deposition on the wafer surface and impacting the performance and yield of semiconductor devices. Besides particulate matter, chemical reagents may also contain soluble impurities such as metal ions and organic matter. PES filter membranes also have a certain interception effect on some soluble impurities, further improving the purity of chemical reagents and ensuring the stability and reliability of the semiconductor manufacturing process.

[0005] In practical applications, besides considering the filter membrane's efficiency in removing impurity particles, the mechanical strength of the filter membrane is particularly critical (especially in the semiconductor field). This is because in the semiconductor field, the cleanliness of the membrane material itself is crucial, requiring repeated cleaning processes to ensure its cleanliness. When cleaning the filter membrane, if the membrane strength is insufficient, the pore size is easily deformed (especially in filter membranes that trap nanoscale impurities), severely affecting the removal efficiency and flux. To improve the mechanical strength of the membrane, those skilled in the art generally control the fiber thickness of the filter membrane, hoping that the fibers forming the porous structure are thicker, or even as thicker as possible. However, further research has found that when the fibers are very thick, the filter membrane flux decreases, especially in nanoscale filter membranes, where the membrane flux decreases significantly and cannot meet the needs of practical applications. Therefore, it is generally believed in the industry that filter membrane flux and strength are often mutually exclusive.

[0006] Therefore, there is an urgent need for a nanoscale porous filter membrane that has both excellent mechanical strength and high flux. Summary of the Invention

[0007] This application provides a polyethersulfone nanoscale porous filter membrane, its preparation method, and its application. Based on nanoscale pores, the filter membrane exhibits excellent mechanical strength and high flux through the synergistic effect of a thin separation layer, high fiber density, appropriate fiber thickness, and high porosity. It is suitable for nanoscale ultrapure filtration of mild systems in semiconductor manufacturing processes, such as the preparation of ultrapure HF diluents and deionized water (DIW).

[0008] To achieve the above objectives, the main technical solution adopted in this application includes: a polyethersulfone nanoscale porous filter membrane, comprising a main body having non-directional tortuous pathways within the main body; the main body comprising a pre-filtration layer and a separation layer for impurity interception; the thickness of the separation layer being 0.4-13 μm; the separation layer comprising separation fibers forming a porous structure; the SEM average diameter of the separation fibers being 15-55 nm; the PMI average pore size of the filter membrane being 10-80 nm, and the porosity being 40%-80%; the fiber density of the filter membrane being 1.1-1.5 g / cm³. 3 .

[0009] The filter membrane of this invention is prepared from polyethersulfone. This type of polymer has excellent mechanical properties, thermal stability, biocompatibility, and ease of phase separation and pore formation. The filter membrane exhibits excellent and stable filtration performance in mild solution environments commonly found in semiconductor wet electronic chemicals, such as weak acids, weak alkalis, and aqueous solutions. The average pore size (PMI) of the filter membrane can characterize the overall pore size of the membrane, which to some extent relates to the filtration accuracy, applicable scenarios, and service life of the filter membrane. In this invention, the average PMI of the filter membrane is 10-80 nm, indicating that the filter membrane is mainly used to retain nanoscale impurity particles, and more specifically, impurity particles with a particle size of 1-10 nm. This is completely different from microfiltration membranes (which retain micron-sized impurity particles) and cannot be compared; they are fundamentally different.

[0010] In the membrane body structure of the filter membrane provided by this invention, it can be found that the entire filter membrane body is mainly divided into two regions in the thickness direction. One region is a pre-filtration layer, whose internal pores have relatively large pore sizes and are mainly used to intercept large particulate impurities in the fluid. The pre-filtration layer has a large dirt holding capacity and a fast flow rate, which plays a pre-filtration role for the fluid and at the same time plays a protective role for the separation layer, ensuring that the membrane as a whole has high mechanical strength. The other region is a separation layer, whose internal pores have relatively small pore sizes (nanoscale) and are mainly used to intercept fine particulate impurities, such as various nanoscale fine impurity particles in the fluid, which plays a separation role and ensures that the filter membrane has a high capture capacity for various fine impurities.

[0011] Given that the filter membrane of this invention is a nanoscale filter membrane, it is primarily aimed at high-precision filtration scenarios in the semiconductor field. Its core advantage lies in its adaptation to the low-impurity characteristics of fluids in this field: in actual filtration, the initial impurity content in the fluid to be treated is already low. Furthermore, the pre-filter layer, acting as a contaminant-holding core, can efficiently intercept slightly larger impurity particles in advance, further reducing the impurity load entering the separation layer and significantly reducing the number of nanoscale impurity particles in the fluid reaching the separation layer. Based on this characteristic, the separation layer can be designed with a relatively thin thickness. By precisely controlling the separation layer thickness to 0.4-13 μm, it can meet the practical requirements for retention efficiency by leveraging a suitable PMI (average pore size) and tortuous pathway, while matching the impurity load of semiconductor applications and avoiding the risk of leakage due to excessive thickness. This thickness control is deeply integrated with the low-impurity fluid characteristics and overall membrane structure design in the semiconductor field, distinguishing it from other fields with high nanoscale impurity loads (where the separation layer needs to be thickened to prevent leakage). Ultimately, through the synergistic design of "a slightly larger overall PMI average pore size, a higher porosity of 40%-80%, and a thin separation layer structure," the fluid flow resistance is significantly reduced while ensuring nanoscale retention accuracy, effectively improving the overall flux of the filter membrane and achieving rapid and efficient filtration of low-impurity fluids in the semiconductor field.

[0012] As is well known, the mechanical strength of filter membranes is particularly critical (especially in the semiconductor field). This is because the cleanliness of the membrane material itself is crucial in the semiconductor industry, requiring repeated cleaning processes to ensure its cleanliness. However, during filter membrane cleaning, if the membrane strength is insufficient, the pore size is easily deformed, especially for filter membranes that trap nanoscale impurities (nanoscale filter membranes). Even slight deformation of the pores within the separation layer can severely affect the filter membrane's flux, causing a rapid decrease in flux. In existing technologies, for porous filter membranes trapping nanoscale impurities, those skilled in the art typically control the fiber thickness of the filter membrane (especially the fibers forming the pores within the separation layer) to ensure the membrane's mechanical strength. However, this often leads to… The fibers that form the porous structure are too coarse, leading to a rapid decrease in flux. However, extensive research has shown that excessively coarse fibers are not necessary to ensure high mechanical strength of the filter membrane. In this invention, the fibers forming the porous structure of the filter membrane are not (nor do they need to be) too coarse. Instead, based on a certain average pore size (PMI), high porosity, and thin separation layer, the mechanical strength of the membrane is ensured through the synergistic effect of controlling appropriate fiber thickness and high fiber density (fiber density refers to the compactness of the fibers forming the porous structure; for porous filter membranes, high fiber density may mean denser fibers and a more compact overall membrane structure). This truly achieves a balance between flux and strength.

[0013] This is because fiber thickness determines the basic skeletal structure of the membrane. Precise control of fiber thickness provides the filter membrane with a uniform and stable skeletal structure, enabling it to maintain structural integrity under high fluid pressure. This avoids the problem of uneven pore structure caused by excessively thick fibers, while also preventing insufficient mechanical strength caused by excessively thin fibers. Furthermore, in this invention, the thickness of the separating fibers is used to represent the overall fiber thickness of the membrane. This is because the separation layer is the region with the smallest pore size inside the membrane and is also a key region affecting the membrane's filtration accuracy and flux. The separating fibers form the porous structure of the separation layer, and compared to fibers in other regions... The fibers are relatively thin; therefore, the thickness of the separating fibers has a significant impact on the strength of the membrane separation layer and even the overall membrane. Our research shows that the average SEM diameter of the separating fibers in this invention is controlled at 15-55 nm, coupled with a high fiber density of 1.1-1.5 g / cm³ (high fiber density often means that the material itself has a denser molecular structure, which enhances the internal bonding and skeletal support of the membrane). This ensures tight internal bonding and a stable skeletal structure while making the pore distribution more uniform and of appropriate size. This satisfies the requirement for trapping nanoscale impurities while providing sufficient and unobstructed pathways for fluid, thus achieving a balance between flux and strength. Furthermore, our research found that fiber thickness determines the basic skeletal structure of the membrane. Although thicker fibers can provide some mechanical support, if the fiber density is low, the bonding between fibers may not be tight enough, and the overall mechanical strength will still be limited. This is why excessively thick fibers are needed to ensure the overall mechanical strength of the membrane.

[0014] Furthermore, compared to existing technologies, the present invention achieves unexpected benefits by controlling the filter membrane to have a thinner separation layer, appropriate fiber thickness, and higher fiber density. This results in a filter membrane with higher cleanliness, lower dissolution, and easier cleaning, making it particularly suitable for the semiconductor field. This is likely because the dense structure with high fiber density restricts the migration of free chain segments, inhibits the release of small molecules, and reduces dissolution. In addition, the dense and uniform pore structure, and the thinner separation layer (where impurities are relatively difficult to clean), make it easier to remove impurities during cleaning, avoiding areas that are difficult to clean due to uneven pore size or a thicker separation layer. Furthermore, the relatively fine separation fibers form a relatively smooth nanoscale surface, reducing contaminant anchoring points and improving backwashing efficiency. Simultaneously, the polyethersulfone surface has no active groups, reducing the risk of chemical bonding of contaminants. Therefore, it better meets the stringent requirements for filter membrane cleanliness and stability in semiconductor wet processes.

[0015] The average pore size of the filter membrane (PMI) can be measured using a PMI pore size meter.

[0016] The thickness of the separation layer of the filter membrane can be determined by first tearing the filter membrane into a separation layer and a pre-filtration layer, and then testing the corresponding parameters of the separation layer; or by characterizing the morphology of the membrane cross-section using a scanning electron microscope, and then calculating it using computer software (such as Matlab, NIS-Elements, etc.) or by manual measurement; in addition, the thickness of the separation layer can also be determined by using colloidal gold of the corresponding particle size as impurity particles for retention testing (for example, the colloidal gold particle size corresponding to a filter membrane with a filtration accuracy of 2nm is 2nm, and the colloidal gold particle size corresponding to a filter membrane with a filtration accuracy of 5nm is 5nm), and the length of the colloidal gold retention area of ​​the corresponding particle size in the filter membrane is the thickness of the separation layer (multiple measurements are taken, and the average value is taken, preferably more than 5 times); of course, those skilled in the art can also obtain the above parameters through other measurement methods, and the above measurement methods are for reference only.

[0017] The average diameter of the separated fibers can be measured using a scanning electron microscope (SEM) to characterize the morphology of the separation layer region of the membrane cross-section. Then, the measurement can be performed using computer software (such as Matlab, NIS-Elements, etc.) or manually, followed by corresponding calculations. For example, the separation layer region of the membrane cross-section can be characterized using an electron microscope to obtain the corresponding SEM image. Since the separation fibers within the separation layer are roughly uniform in thickness, a certain area, such as 100 μm, can be selected. 2 (10μm x 10μm) or 25μm 2 (1μm multiplied by 1μm) or 1μm 2 (1μm x 1μm), the specific area size depends on the actual situation, and then use the corresponding computer software or manual to measure the diameter of the separated fibers on this area (preferably more than 10 times, the specific number of times depends on the situation), and then calculate and take the average value to obtain the average diameter of the separated fibers. Of course, those skilled in the art can also obtain the above parameters through other measurement methods. The above measurement methods are for reference only.

[0018] When the membrane porosity is too high, it leads to low mechanical strength, resulting in low industrial practical value and inability to meet market demands. Conversely, when the membrane porosity is too low, it affects the flow rate, leading to slower filtration speed, longer filtration time, and higher time costs. Furthermore, it results in low dirt-holding capacity, a short service life, and the need for frequent membrane replacement, significantly increasing economic costs. The filter membrane in this invention has a porosity of 40%-80%, giving it not only good mechanical strength but also a fast filtration speed, high flow rate, and high dirt-holding capacity, enabling it to trap a large number of impurity particles, resulting in a long service life and lower economic costs. Common methods for testing porosity include mercury intrusion porosimetry, density methods, and wet / dry membrane weighing methods. Of course, those skilled in the art can obtain the above parameters through other testing methods; the aforementioned testing methods are for reference only.

[0019] The fiber density of the filter membrane in this application was measured using a gas displacement method with an AccuPyc-II-1340 fully automatic density tester from the United States. It is preferable to perform multiple measurements (more than 5 times) and then take the average value. The test method is as follows: the sample is placed in the density tester, and helium gas is used as the medium. The pressure in the measuring chamber is gradually increased to a specified value, and then the helium gas expands into the expansion chamber. The equilibrium pressure of the two processes is automatically recorded by the instrument. According to the law of conservation of mass, the volumes of the measuring chamber and the expansion chamber are calibrated using a standard sphere, and then the volume of the sample is determined to calculate the corresponding density. Fiber density refers to the mass per unit volume of the material (filter membrane) in a fully compacted state, i.e., the density after removing the internal pores of the material (in this invention, the density of the fiber skeleton forming a porous structure). Fiber density is affected by factors such as the chemical structure of the polymer molecular chains and intermolecular forces. The filter membrane in this application has a high fiber density, indicating that the fiber skeleton of the filter membrane has strong supporting force.

[0020] The filter membrane in this application has a single-layer asymmetric membrane structure, meaning it is integrally molded without any "composite" or other processes. The entire membrane is made of polyethersulfone material, with uniform material throughout and no material variation. The only variation in the membrane's structure is its overall structure. In contrast, composite membranes have multiple layers, and the pore size changes abruptly when transitioning from one layer to another. In this application, non-directional tortuous pathways refer to randomly oriented groove structures and / or discretely distributed pore structures, and these non-directional tortuous pathways are interconnected.

[0021] As a further improvement of the present invention, the separation layer includes a liquid outlet surface located on the side of the separation layer opposite to the pre-filtration layer; the liquid outlet surface includes a plurality of first pores, the liquid outlet surface includes surface fibers for forming the first pores, the SEM average diameter of the surface fibers is 3-35 nm; the ratio between the SEM average diameter of the surface fibers and the SEM average pore size of the first pores is 0.6-0.95; and / or, the thickness of the filter membrane is 60-100 μm.

[0022] In this application, the liquid outlet surface, as the "final barrier" (outermost side of the separation layer) of the filter membrane, has its surface fiber size directly determining the structural stability of the filter membrane, and also affecting the flow rate and cleanliness (ease of backwashing). Research has shown that the average SEM diameter of the surface fibers is controlled to be 3-35 nm. This is primarily related to the fact that the filter membrane of this invention is mainly used for semiconductor filtration, requiring the retention of impurities of 1-10 nm (such as metal ion complexes). Furthermore, the nanoscale surface fibers form a tightly interwoven network structure, which, even with a relatively small surface diameter, can still disperse filtration pressure through "multi-point contact," giving the liquid outlet surface excellent tensile and tear resistance. Especially during high-pressure filtration, the finer surface fiber network is less prone to breakage, maintaining structural integrity. Simultaneously, the relatively fine surface fibers reduce mechanical obstruction to the fluid, and the gaps between the surface fibers (first pores) more easily form continuous mass transfer channels, reducing fluid resistance.

[0023] Furthermore, by adjusting the ratio of the average SEM diameter of the surface fibers to the average SEM pore size of the first pore, this ratio signifies that "the pore size of the first pore is slightly larger than the diameter of the surface fibers," forming a stable structure where "the fibers flexibly wrap around the pores." The surface fibers act as a framework, wrapping around the slightly larger pores. When filtration pressure is applied, the surface fibers provide sufficient support for the first pores, ensuring membrane pore stability (anti-collapse resistance); they are also less prone to breakage under fluid pressure impact, ensuring stable filtration over a long period. Simultaneously, it ensures that the pore size is large enough to allow fluid to pass through smoothly (reducing resistance), and because the membrane pore size is only slightly larger than the fiber diameter, the "constraint effect" of the fibers on the pores is sufficient to stabilize the pore shape, preventing temporary increases in pore size (leaking impurities) or decreases in pore size (reducing flux) due to pressure fluctuations, thus achieving flux stability. If the ratio is too small, the first pore lacks sufficient surface fiber support, and the membrane pores are prone to deformation and collapse under high pressure; if the ratio is too large (e.g., greater than 1), the fibers are too coarse, and excessive accumulation will lead to narrow pores, increasing resistance. This design is particularly well-suited for semiconductor filtration scenarios: it improves the filtration efficiency for low-impurity fluids (such as DIW and DHF) through "slightly larger pores," while ensuring the reliability of 1-10nm impurity retention thanks to "appropriate fiber constraint." At the same time, it enhances the structural toughness of the filter membrane under repeated pressure fluctuations, extending its service life.

[0024] Furthermore, by further adjusting the thickness of the filter membrane to 60-100μm, a "macroscopic support" is provided for the entire membrane. The pre-filtration layer, as a "buffer layer," can disperse the pressure transmitted to the separation layer, preventing the separation layer from breaking due to excessive local stress. The synergistic effect of these features ensures that the filter membrane, especially the liquid outlet surface, does not collapse (stable retention accuracy) or break (structural integrity) when subjected to filtration pressure, thus meeting the strength requirements of long-term semiconductor filtration. Simultaneously, it facilitates the creation of a permeable network characterized by "low-resistance channels and efficient mass transfer." The fine surface fibers, matched with a specific ratio of surface fiber diameter to the diameter of the first pore, ensure the pressure resistance of the liquid outlet surface, making the channels at the liquid outlet both unobstructed and stable, resulting in low local resistance during fluid flow. Furthermore, the separation layer itself is relatively thin (0.4-13μm), combined with an overall filter membrane thickness of 60-100μm (due to the high porosity of the pre-filtration layer), significantly shortening the mass transfer path of the fluid from entering the membrane to passing through it. In addition, the applicant found that this membrane structure design is very suitable for semiconductor filtration because semiconductor filtration has extremely high requirements for "flux stability" and "retention reliability" (impurity leakage or a sudden drop in flux can lead to wafer defects): the surface fiber to pore ratio design ensures that "pore diameter does not fluctuate with pressure," avoiding accidental impurity leakage; the matching of the overall thickness with the separation layer thickness ensures "stable mass transfer efficiency," meeting the stringent control of fluid supply speed in processes such as photolithography and etching (e.g., DIW cleaning requires a continuous and stable flow rate).

[0025] Individually, the nanoscale surface fibers provide microscopic strength and low resistance, the fiber-to-pore ratio ensures pore stability, and the overall thickness provides macroscopic support. When combined, the three form a filter membrane structure that is "microscopically robust, macroscopically flexible, with stable pores and high mass transfer efficiency." This structure not only meets the high-precision interception of semiconductor nanoscale impurities but also adapts to the needs of continuous industrial production through high throughput and damage resistance, further achieving a balance between strength and throughput when used as a semiconductor filter membrane.

[0026] The size of pores and the thickness of fibers on the membrane surface can be measured by characterizing the membrane structure using a scanning electron microscope, followed by measurement using computer software (such as Matlab, NIS-Elements, and NanoMeasurer) or by manual measurement, and then performing the corresponding calculations.

[0027] During the membrane preparation process, in the direction perpendicular to the membrane thickness (if the membrane is a flat sheet, this direction is planar; if the membrane is a hollow fiber membrane, this direction is perpendicular to the radius), its various characteristics, such as pore size, are roughly uniform and basically consistent. Therefore, the overall pore size and fiber thickness on the corresponding plane can be reflected by the pore size and fiber thickness in a certain area on that plane.

[0028] In actual measurement, the membrane outlet surface can be characterized using an electron microscope to obtain the corresponding SEM image. Since the pore size and fiber thickness on the membrane surface are roughly uniform, a certain area, such as 1 μm, can be selected. 2 (1μm x 1μm) or 25μm 2 (5μm x 5μm), or 100μm 2 (1μm x 1μm), the specific area size depends on the actual situation. Then, use appropriate computer software or manual measurement to measure the diameter of the surface fibers and the pore size of the first hole on this area (preferably more than 10 times, the specific number of times depends on the situation). Then calculate and take the average value to obtain the average diameter of the surface fibers and the average pore size of the first hole. Of course, those skilled in the art can also obtain the above parameters through other measurement methods. The above measurement methods are for reference only. The SEM average pore size of the second hole on the liquid inlet surface can also be obtained by referring to this method.

[0029] The thickness of the filter membrane can be determined by characterizing the membrane structure using a scanning electron microscope, followed by calculation using computer software (such as Matlab, NIS-Elements, etc.) or manual measurement. Of course, those skilled in the art can also obtain the above parameters using other measurement methods (such as direct measurement with measuring tools). The above measurement methods are for reference only. When the membrane thickness is too small, its mechanical strength will be low; when the membrane thickness is too large, its filtration time will be too long, resulting in excessive time costs. The thickness of the filter membrane in this application is 60-100 μm, ensuring that the filter membrane not only has high mechanical strength but also high filtration efficiency, short filtration time, and low time costs.

[0030] As a further improvement of the present invention, the ratio between the SEM average diameter of the separated fibers and the SEM average diameter of the surface fibers is 1.3-4; the pre-filtration layer includes a liquid inlet surface located on the side of the pre-filtration layer opposite to the separation layer; the liquid inlet surface includes a plurality of second pores, the ratio between the SEM average pore diameter of the second pores and the SEM average pore diameter of the first pores is not less than 10; the thickness of the separation layer is 1-9 μm.

[0031] In this application, the separating fibers are not excessively thick, but research has shown that the separating fibers are preferably slightly thicker than the surface fibers (the diameter ratio between the two is adjusted to 1.3-4), thereby creating a gradient structure of "fine surface and supportive inner layer" in the separating layer: the surface fibers have a smaller diameter, which can construct high-precision first pores (ensuring interception accuracy and preventing leakage); the separating fibers have a larger diameter, serving as the "skeleton support" for the surface fibers, improving the overall mechanical strength of the separating layer, preventing the surface fibers from breaking under high pressure due to excessive thinness, and providing a smoother transmission channel for the fluid (reducing deep resistance).

[0032] Meanwhile, the filter membrane of this application is further controlled to be a highly asymmetric structure, with the pore size of the second pore on the inlet surface being much larger than that of the first pore on the outlet surface (the ratio of the two is not less than 10), forming a "step-by-step interception" filtration gradient: the pre-filtration layer first intercepts larger impurity particles (such as those above 100 nm) in the fluid, avoiding blockage or wear of the separation layer by large particles of impurities, and reducing the dirt-holding load of the separation layer; at the same time, the large-diameter second pore can reduce the flow resistance of the pre-filtration layer, ensuring that the fluid enters the membrane quickly.

[0033] Furthermore, the thickness of the separation layer is preferably controlled within 1-9 μm. Combined with the above structural design, a balance of "low resistance + high strength" is achieved: The thin separation layer shortens the mass transfer path: Compared to a thick separation layer, the 1-9 μm thickness significantly reduces fluid resistance. Combined with the "coarse" fiber channels inside the separation layer (the separation fibers have appropriate fiber thickness), fluid retention within the separation layer is reduced, significantly improving filtration speed. The gradient structure avoids "thin and inefficient": The pre-filtration layer intercepts large impurities first, ensuring the fluid reaching the separation layer is already relatively clean. The thin separation layer does not need to bear excessively high contaminant pressure, thus ensuring retention efficiency (meeting semiconductor requirements) without rapid clogging due to excessive thinness, solving the problem of "thin separation layers being prone to failure."

[0034] Furthermore, the combined effect of the above features further enhances the overall structural strength and stability. The separation layer is reinforced with coarser separation fibers (1.3-4 times larger than surface fibers), providing a rigid framework for the thin separation layer (1-9 μm), resisting filtration pressure and preventing the thin structure from cracking or deforming due to stress. The pre-filtration layer provides buffering: the large pore size design at the inlet surface disperses fluid impact pressure, reducing direct stress on the separation layer. Simultaneously, the high overall fiber density of the membrane further protects the integrity and stability of the thin separation layer.

[0035] In summary, these features, through the synergy of "gradient interception, thin-layer high efficiency, and structural support," overcome the technical contradictions of "high precision inevitably leads to low throughput" and "thin separation layers inevitably lead to short lifespan," further enabling filter membranes to possess both high flow rate and high strength in the semiconductor field.

[0036] As a further improvement of the present invention, the porosity of the liquid inlet surface is 8%-40%, and the ratio between the porosity of the liquid inlet surface and the porosity of the liquid outlet surface is 1.1-2.

[0037] And / or, the average SEM pore size of the second hole is not less than 0.7 μm.

[0038] Surface porosity refers to the proportion of the total area of ​​pores to the total surface area of ​​a given surface. The inlet surface is the area where the fluid first contacts the filter membrane. Research has shown that adjusting the porosity of the inlet surface to 8%-40% is beneficial in two ways: firstly, it provides sufficient open area for the pre-filtration layer to allow the fluid to pass through quickly (avoiding excessive inlet resistance), and secondly, the non-porous fiber skeleton provides basic mechanical support (preventing damage to the inlet surface due to excessive porosity), meaning the inlet surface has sufficient pressure resistance. Furthermore, adjusting the porosity of the inlet surface to be slightly larger than that of the outlet surface (the ratio of the two is 1.1-2 times) helps to achieve a "loose inlet and precise outlet" effect. The gradient distribution is as follows: the inlet surface rapidly receives and initially diverts fluid through a larger pore ratio, reducing fluid retention on the membrane surface and ensuring uniform fluid flow to the separation layer, thus improving overall filtration efficiency. The outlet surface has a similar (slightly smaller) pore area ratio to the inlet surface, but with smaller pore sizes, enabling dense interception of nanoscale impurities. This avoids impurity leakage due to excessive pores, while also minimizing the impact on overall membrane flux, maintaining a relatively fast overall flux and allowing fluid to pass through the filter membrane in a shorter time. It also ensures uniform stress on the entire membrane, reducing the risk of localized tearing.

[0039] The second pore's SEM average pore size is no less than 0.7 μm, further indicating that the second pore size is much larger than the first pore. This allows the separation layer to preferentially intercept micron-sized large particulate impurities in the fluid (such as pipe corrosion debris and microbial aggregates), preventing them from entering the separation layer and clogging the nano-sized pores, thus reducing the separation layer's contaminant load. Furthermore, the large pore size of the second pore (≥0.7 μm) combined with a pore area ratio of 8%-40% on the inlet surface provides the fluid with a "wide channel + high open area ratio" inlet, reducing the resistance of the fluid entering the membrane and allowing a large amount of fluid to quickly penetrate into the pre-filtration layer.

[0040] The specific method for testing the pore area ratio of the inlet / outlet liquid surfaces is as follows: First, the inlet and outlet liquid surfaces of the filter membrane are characterized using an electron microscope to obtain corresponding ideal SEM images. Since the pore distribution on the membrane surface is roughly uniform, a certain suitable area region, such as 1 μm, is then selected. 2 (1μm x 1μm) or 25μm 2 (5μm x 5μm) or 100μm 2The specific area size (e.g., 1μm x 1μm) depends on the actual situation. Then, the diameter of all holes in the area is measured using appropriate computer software or manually. The area of ​​each hole in the area is then calculated, and the sum of the hole areas in the area is obtained. This sum is then divided by the area of ​​the selected area to obtain the hole area ratio of the corresponding area. Preferably, at least three statistically consistent areas are selected to measure the hole area ratio, and the average value is taken as the hole area ratio of the corresponding surface. Of course, those skilled in the art can also obtain the above parameters through other measurement methods. The above measurement methods are for reference only. In addition, the average diameter of the second hole is determined by referring to the test method of the first hole, so it will not be described again.

[0041] As a further improvement of the present invention, the pre-filtration layer includes a contaminant-holding zone and a transition zone; the transition zone is located on the side of the pre-filtration layer near the separation layer, the transition zone has a sponge-like structure, and the porosity of the transition zone is greater than that of the separation layer; the transition zone includes supporting fibers for forming a porous structure, and the supporting fibers on the side near the separation layer are connected to the separation fibers; the contaminant-holding zone is located on the side of the pre-filtration layer away from the separation layer, the contaminant-holding zone has a finger-like pore structure, and the ratio of the thickness of the contaminant-holding zone to the overall thickness of the membrane is not less than 0.3; the porosity of the filter membrane is 45%-75%.

[0042] In some polyethersulfone (PES) filter membranes, the pre-filtration layer includes a contaminant-holding zone and a transition zone. The contaminant-holding zone is located on the side of the pre-filtration layer away from the separation layer and has a finger-like pore structure. Its thickness is no less than 30% of the overall membrane thickness (high proportion thickness). This finger-like pore structure reduces fluid inlet resistance, ensuring that fluid can quickly enter the filter membrane and reduce the overall pressure drop at the inlet surface, thereby improving filtration efficiency. At the same time, the larger finger-like pores reduce the accumulation of impurities and the risk of clogging at the inlet surface, helping to maintain long-term stable filtration performance.

[0043] The high porosity of the transition zone (sponge-like structure) and the supporting fibers connected to the separation fibers, on the one hand, accept the fluid in the contamination zone and further intercept medium-sized impurities, and on the other hand, achieve a smooth transition from the large pore size of the contamination zone to the small pore size of the separation layer through the relatively dense sponge-like structure, avoiding damage to the separation layer by fluid impact, and providing structural support for the separation layer.

[0044] In addition, the overall porosity of the membrane is preferably adjusted to 45%-75% to match the membrane gradient structure. This ensures both dirt-holding capacity and transition stability while maintaining high overall flux of the filter membrane, thus meeting the stringent requirements of the semiconductor industry for fluid filtration efficiency and purity.

[0045] Inside the filter membrane, except for the separation layer, the other areas can be considered as a pre-filtration layer. The pre-filtration layer is divided into a dirt-holding zone and a transition zone along the thickness direction of the filter membrane (i.e., the fluid filtration direction). The dirt-holding zone, the transition zone, and the separation layer are arranged continuously along the "inlet liquid surface → outlet liquid surface" direction. The specific definitions and structural features are as follows:

[0046] The contaminant-holding zone is located on the side of the pre-filtration layer away from the separation layer (i.e., the side closer to the liquid inlet surface of the filter membrane). Its main structure is a finger-shaped pore structure, and the long axis of the finger-shaped pores extends basically parallel to the thickness direction of the filter membrane. The extent of the contaminant-holding zone is defined by the boundary of the finger-shaped pores or macropores (pore diameter significantly larger than the pores in the transition zone, and pore diameter ≥ 1 μm) closest to the liquid outlet surface. This boundary is the dividing line between the contaminant-holding zone and the transition zone. This dividing line can be clearly identified by scanning electron microscopy (SEM). There are continuously distributed finger-shaped pores on the side of the dividing line closer to the liquid inlet surface, and no finger-shaped pores or macropores on the side closer to the liquid outlet surface.

[0047] The transition zone is defined and structured as follows: the transition zone is located on the side of the pre-filtration layer closest to the separation layer (i.e., between the contaminant-holding zone and the separation layer). Its range is defined by the boundary of all finger-shaped pores or macropores in the contaminant-holding zone closest to the liquid outlet surface, and ends at the boundary between the pre-filtration layer and the separation layer. The transition zone has a sponge-like porous structure. SEM observation shows that its interior consists of uniformly connected fine pores, without finger-shaped pores or macropores with a diameter ≥1μm. Furthermore, on the side of the transition zone closest to the separation layer, the internal supporting fibers are directly connected to the separation fibers of the separation layer, forming a structurally continuous transition interface.

[0048] The above definition method can be achieved by combining SEM image observation with thickness measurement: select at least 3 different observation areas along the thickness direction of the filter membrane, and measure the "distance from the liquid inlet surface of the filter membrane to the boundary line of the contaminant-transition zone" (i.e., the average thickness of the contaminant-transition zone) and the "distance from the boundary line of the contaminant-transition zone to the interface of the transition zone-separation layer" (i.e., the average thickness of the transition zone). Take the arithmetic mean as the thickness parameter of the corresponding area. Of course, those skilled in the art can also obtain the above parameters through other measurement methods. The above measurement methods are for reference only.

[0049] As a further improvement of the present invention, the thickness of the transition zone is 5-40 μm; the SEM average diameter of the supporting fiber is 20-65 nm; and / or, the ratio between the SEM average diameter of the supporting fiber and the SEM average diameter of the separating fiber is 1.05-1.65.

[0050] By adjusting the thickness of the transition zone to 5-40μm, this thickness range ensures that the transition zone can fully accommodate the fluid in the contamination zone, achieving secondary interception of medium-sized impurities through a sponge-like structure, while avoiding increased fluid resistance due to excessive thickness, thus ensuring the overall flux of the filter membrane; at the same time, it provides appropriate structural buffering for the separation layer, reducing damage to the separation layer from direct fluid impact.

[0051] Further adjustments were made to the SEM average diameter of the support fibers, which was 20-65 nm. The nanoscale support fibers could construct relatively dense and interconnected sponge-like pores, maintaining sufficient porosity to ensure smooth fluid flow while enhancing the mechanical strength of the transition zone through the fine fiber network, preventing structural collapse under filtration pressure, and providing stable substrate support for the separation layer. In addition, the diameter ratio of the support fibers to the separation fibers was controlled at 1.05-1.65. This ratio ensured that the diameters of the support fibers and separation fibers were close, ensuring a tight connection between them (reducing interfacial gaps) and preventing fluid leakage from the connection points, which could lead to retention failure. At the same time, the support fibers were slightly thicker than the separation fibers, which could provide effective support for the separation fibers through their own rigidity, enhancing the deformation resistance of the separation layer.

[0052] The aforementioned features synergistically construct an integrated "buffer-support-mass transfer" transition structure. The 5-40μm thickness of the transition zone matches the 20-65nm thickness of the support fibers, ensuring secondary interception while forming a well-connected fluid channel, reducing mass transfer resistance between the transition zone and the separation layer. The diameter ratio of the support fibers to the separation fibers (1.05-1.65) ensures a tight connection between the two, preventing interface leakage and strengthening the structural stability of the separation layer through the rigidity of the support fibers, preventing deformation of the separation layer under pressure. The overall synergy enables the transition zone to effectively accommodate the contaminant zone and the separation layer, achieving graded interception of impurities, stable fluid transport, and stable structural support, ultimately improving the filter membrane's retention accuracy, flux stability, and service life, adapting to the flux and strength requirements of high-precision semiconductor filtration.

[0053] The average diameter of the supporting fibers can be obtained by characterizing the morphology of the membrane transition region using a scanning electron microscope (SEM), followed by measurement using computer software or manually, and then performing corresponding calculations. For example, first, the cross-section of the membrane transition region can be characterized using an SEM to obtain the corresponding SEM image. A suitable area can be selected, the specific size of which depends on the actual situation. Then, the diameter of the fibers in that area can be measured using appropriate computer software or manually (preferably more than 10 times, the specific number of times depending on the situation). Then, the average diameter of the fibers can be calculated. Subsequently, the average major axis, average minor axis, and average length of the first distance of the finger pores can be obtained using a similar method. Of course, those skilled in the art can also obtain the above parameters using other measurement methods, and the above measurement methods are for reference only.

[0054] As a further improvement of the present invention, the contamination-receiving area has a plurality of finger-shaped pores, the extension direction of the long axis of the finger-shaped pores is the thickness direction of the filter membrane; the SEM average long axis of the finger-shaped pores is 30-80μm; the ratio of the SEM average long axis of the finger-shaped pores to its SEM average short axis is not less than 3.

[0055] In the contaminant-holding zone, a non-sponge-like structure is preferred over a finger-like pore structure. This means the zone contains multiple finger-like pores (similar to finger-shaped holes with high porosity). The long axis of these finger-like pores extends primarily along the thickness direction of the filter membrane. This allows the finger-like pores to be oriented along the fluid filtration path (thickness direction), providing a direct channel for the fluid to reach the transition zone, reducing diffusion resistance, and increasing the fluid throughput in the contaminant-holding zone. Further, the average long axis of the finger-like pores is controlled to 30-80 μm. This size ensures that the finger-like pores have sufficient volume to accommodate larger impurities (such as micron-sized particles), improving the impurity carrying capacity of the contaminant-holding zone and preventing small pores from failing due to rapid filling. Furthermore, the ratio of the long axis to the short axis of the finger-like pores is not less than 3, giving the pore structure a "slender" shape. While ensuring volume, the narrow short axis reduces the space occupied by the pore walls, increasing pore connectivity. This enhances the overall porosity of the contaminant-holding zone while avoiding structural fragility caused by excessive compression between pores.

[0056] These features synergistically construct a "directional, efficient, high-capacity, and clogging-resistant" contaminant-holding structure: elongated, finger-like pores extending along the thickness direction (long diameter 30-80 μm, long diameter / short diameter ratio ≥3) form low-resistance channels penetrating the contaminant-holding zone, ensuring rapid fluid flow and trapping large impurities within the pores; the ratio of long to short diameters balances contaminant holding capacity and structural stability, avoiding insufficient support due to excessively large pores or clogging due to excessively small pores, while smoothly connecting with the sponge-like structure of the transition zone, providing clean fluid for subsequent fine filtration. Ultimately, this synergistic effect improves the contaminant holding efficiency and lifespan of the filter membrane, adapting to the graded interception requirements of impurities in semiconductor fluids.

[0057] The filter membrane cross-section was observed using a scanning electron microscope (SEM). Several regions (preferably at least 5 regions) were randomly selected within the contaminant-containing area, each region containing several complete finger-like pore structures. The major diameter was defined and measured: the major diameter of the finger-like pore is its maximum extension along the filter membrane thickness direction (i.e., the fluid filtration direction). Using the SEM's built-in measurement tools or gas image analysis software, the major diameter of each selected finger-like pore was measured (preferably more than 10 pores were measured), and the measured values ​​were recorded. Results were calculated: after removing obvious outliers (such as broken pores caused by sample preparation), the arithmetic mean of the remaining measured values ​​was taken as the SEM average major diameter of the finger-like pores. Of course, those skilled in the art can obtain the above parameters through other measurement methods; the above measurement methods are for reference only.

[0058] As a further improvement of the present invention, the distance between two adjacent finger-shaped holes perpendicular to the thickness direction is a first distance, the SEM average length of the first distance is 0.8-5μm, and the ratio between the first distance and the SEM average minor axis of the finger-shaped holes is not less than 0.1.

[0059] The combination of a first distance of 0.8-5μm and a ratio of not less than 0.1 ensures that the finger pores are moderately densely distributed laterally (perpendicular to the membrane thickness direction) to increase the total amount of contaminants, while the sufficiently wide interpore skeleton ensures that the filter membrane can withstand filtration pressure without deformation; thus achieving a balance between membrane pressure resistance and flux. In addition, under the synergistic effect of the above features, the contaminant-holding zone can efficiently retain larger impurities (such as particles and colloids of 1-50μm) in the raw liquid, and the fluid can flow smoothly through the finger pores to the transition zone, avoiding the decline in filtration performance caused by pore blockage or structural damage, providing stable preliminary support for the subsequent fine retention in the transition zone and the precise filtration of the separation layer.

[0060] As a further improvement of the present invention, the nearest SEM average distance between the finger-shaped pores and the liquid inlet surface is 2-10 μm; the PMI average pore size of the filter membrane is 15-60 nm; and the fiber density of the filter membrane is 1.2-1.45 g / cm³. 3 .

[0061] The nearest SEM average distance between the finger-shaped pores and the inlet surface is 2-10 μm: This distance limits the "initial buffer layer thickness" between the finger-shaped pores in the contaminant-holding zone and the inlet side of the filter membrane. This avoids the finger-shaped pores being directly exposed on the inlet surface due to the distance being too close (<2 μm), which can easily cause pore damage due to the impact of the raw liquid and direct blockage of the pores by impurities. It also prevents the transition layer below the inlet surface from being too thick due to the distance being too far (>10 μm), which can increase the initial permeation resistance of the fluid and reduce the filtration flux.

[0062] The synergistic enhancement of the filter membrane's "structural stability and anti-fouling ability" is achieved as follows: On the one hand, the buffer distance between the finger-shaped pores and the inlet liquid surface reduces the direct impact of the raw liquid on the finger-shaped pores. Combined with a fiber density of 1.2-1.45 g / cm³ (ensuring that the skeleton of the dirt-holding zone and transition zone is not easily deformed), this avoids the problems of pore collapse and abnormal pore size increase during filtration. On the other hand, the 15-60 nm PMI pore size ensures that the separation layer only retains target nanoscale impurities, while large particle impurities are retained by the finger-shaped pores in the dirt-holding zone and the inlet liquid buffer layer, reducing the contamination load on the separation layer. These three factors work together to extend the service life of the filter membrane, reduce the backwashing frequency, and improve the long-term filtration stability.

[0063] As a further improvement of the present invention, under the conditions of a positive pressure of 0.03 MPa and a temperature of 20°C, the time required for 50 ml of water to pass through a filter membrane with a diameter of 47 mm is 10 s to 160 s.

[0064] The tensile strength of the filter membrane is not less than 5 MPa, and the elongation at break is not less than 10%.

[0065] The TOC leaching amount of the filter membrane does not exceed 0.5 ppb, and the metal ion precipitation amount does not exceed 10 ppt.

[0066] Under a positive pressure of 0.03 MPa and a temperature of 20°C, it takes 10-160 seconds for 50 ml of water to pass through a filter membrane with a diameter of 47 mm. This indicates that the filter membrane has a high water flux, and the liquid can pass through the filter membrane quickly, meeting the high-flux filtration requirements in semiconductor wet processes.

[0067] The tensile strength of the filter membrane is not less than 5MPa and the elongation at break is not less than 10%, as measured by a universal tensile testing machine. This indicates that the filter membrane has good mechanical properties, can withstand certain mechanical stress during filtration, is not easily broken, and has a certain degree of flexibility, making it less likely to be damaged during stretching or bending operations, thus ensuring the stability and reliability of the filter membrane during long-term use.

[0068] The TOC leaching of the filter membrane does not exceed 0.5 ppb, and the metal ion precipitation does not exceed 10 ppt. This means that the filter membrane itself dissolves very little organic matter and metal ions and other impurities, which can effectively ensure the purity of the filtered liquid and avoid contamination of chemical reagents in the semiconductor manufacturing process by the leaching of the filter membrane material itself, thereby ensuring the performance and yield of semiconductor devices.

[0069] This invention also provides a method for preparing a polyethersulfone nanoscale porous filter membrane, comprising the following steps:

[0070] S1: Prepare a casting solution and cast it onto a carrier to form a liquid film; the casting solution comprises the following components by weight: 14-22 parts (preferably 16-20 parts) of polyethersulfone, 46-63 parts (preferably 50-60 parts) of a first solvent, 5-20 parts (preferably 8-16 parts) of a volatile second solvent, and 5-15 parts (preferably 8-12 parts) of a small molecule organic additive; the viscosity of the casting solution is 5000-15000 cps (preferably 8000-12000 cps).

[0071] S2: Pre-phase separation, the liquid film is placed in an environment with a relative humidity of 70%-95% (preferably 75-85%), and an airflow with a flow rate of 0.5-3m / s (preferably 1.5-2.5m / s) is blown onto the surface of the liquid film to allow the liquid film to undergo pre-phase separation, wherein the pre-phase separation time is 1-7s (preferably 2-5s).

[0072] S3: Coagulation and phase separation. The pre-separated liquid film is introduced into a coagulation bath for phase separation treatment to obtain a green film. The phase separation treatment time is 20-50s (preferably 30-40s). The coagulation bath includes water and a third solvent, and the mass content of the third solvent in the coagulation bath is 5%-20% (preferably 8%-15%).

[0073] Furthermore, it is preferable that the third solvent is the same as the first solvent;

[0074] S4: Cleaning and drying. The biofilm obtained in S3 is cleaned and dried to obtain a polyethersulfone nanoporous filter membrane.

[0075] As a further improvement of the present invention, the first solvent and the third solvent are both at least one selected from dimethyl sulfoxide, N-ethylpyrrolidone, N-methylpyrrolidone, and dimethylacetamide; preferably, the first solvent and the third solvent are the same.

[0076] The second solvent is acetone or tetrahydrofuran;

[0077] The small molecule organic additive is at least one of isopropanol, ethanol, ethylene glycol, and propylene glycol; the temperature of the casting solution is not lower than 30°C.

[0078] As a further improvement of the present invention, the temperature of the coagulation bath is 10-25°C, and is 10-45°C lower than the temperature of the casting solution; and / or, the mass ratio of the first solvent to the second solvent in the casting solution is 3-11.

[0079] By adopting the above technical solution, in preparing the polyethersulfone nanoporous filter membrane of this application, a casting solution is first prepared and then cast onto a carrier to form a liquid film. The casting solution includes polyethersulfone (PES) (a film-forming substance with good film-forming processing properties), an organic solvent (used to dissolve PES), and small molecule organic additives (used to adjust the pore structure and membrane fiber density).

[0080] One of the inventive aspects of this invention is the unique formulation of the casting solution. The organic solvent comprises a first solvent and a second solvent. The first solvent is preferably at least one of dimethyl sulfoxide, N-ethylpyrrolidone, N-methylpyrrolidone, and dimethylacetamide. The second solvent is preferably acetone or tetrahydrofuran. The second solvent is more volatile than the first solvent. Preferably, the mass ratio of the first solvent to the second solvent in the casting solution is 3-11. Optimizing this range helps balance the viscosity and pore-forming properties of the casting solution. The first solvent, as the main solvent component, provides sufficient dissolving power to ensure the uniformity of the casting solution and regulate the separation layer thickness (suppressing excessive phase separation). The second solvent, with its high volatility and pore-forming properties, triggers ideal pre-phase separation, laying the foundation for a thin separation layer. The synergistic effect of these two solvents balances the phase separation rate. Through the synergistic effect of a suitable pre-phase separation process, an ideal separation layer structure is achieved. Subsequent combination with a suitable re-phase separation process yields an ideal overall filter membrane structure.

[0081] As one of the inventive aspects of this invention, in order to achieve better fiber density, no polymeric pore-forming agents such as PVP and PEG were added to the casting solution (our research found that the presence of polymeric pore-forming agents would greatly reduce fiber density). At the same time, we added small molecule additives, which can also adjust the membrane pore structure to a certain extent. Combined with ideal casting solution viscosity and phase separation process, the final filter membrane can still have ideal pore size and porosity, while making the fiber structure formed by the filter membrane denser and the fiber density high. This enhances the mechanical strength of the filter membrane, enabling it to better withstand pressure during filtration, reducing the risk of membrane deformation and damage, thereby extending the service life of the filter membrane and improving the stability and reliability of filtration. Furthermore, the viscosity of the casting solution is controlled within the range of 5000-15000 cps (preferably 8000-12000 cps). This viscosity is beneficial for improving the diffusion performance of solvents and non-solvents, thus achieving a thin separation layer. Preferably, the casting solution temperature is not lower than 30℃, as a higher temperature helps increase the evaporation rate of the second solvent. Combined with the corresponding pre-phase separation process, this further accelerates the diffusion and escape of the second solvent from the liquid membrane surface, forming a synergistic evaporation effect of "internal heat + external drive," allowing the liquid membrane surface to complete preliminary densification in a short time (laying the foundation for the nanoscale structure of the separation layer), avoiding surface porosity due to insufficient evaporation, and preserving structural control space for subsequent phase separation. Simultaneously, appropriate temperature helps ensure the fluidity of the casting solution, enabling it to form a more uniform liquid film during casting, reducing defects caused by uneven liquid film and improving the various performance characteristics of the filter membrane.

[0082] The mass ratio of each component in the casting solution has a significant impact on the structure and performance of the filter membrane, such as the average pore size, porosity, and thickness of the filter membrane. This application achieves a suitable structure and ideal performance by controlling the mass ratio of each component in the casting solution and by combining the effects of pre-phase separation, coagulation bath, and other process conditions.

[0083] Then, a pre-phase separation treatment is performed. The liquid film is placed in an environment with a relative humidity of 70%-95%, and an airflow with a velocity of 0.5-3 m / s is blown onto the surface of the liquid film, while the pre-phase separation time is controlled to be 1-7 seconds. This pre-phase separation method using high-concentration water vapor and airflow has the following main effects:

[0084] ① Directional triggering of surface phase separation to construct a fine liquid surface structure: The high humidity environment of 70%-95% provides sufficient water vapor for the liquid film surface, while the airflow of 0.5-3m / s accelerates the escape of the volatile second solvent on the surface of the liquid film. Under the combined effect of the two, a local environment of "sudden drop in solvent concentration + water vapor penetration" is rapidly formed on the surface of the liquid film, triggering the rapid precipitation and rearrangement of polyethersulfone (PES) molecules to form a thin and relatively dense surface layer. This surface layer is the prototype of the subsequent liquid surface. The precipitated PES molecular chains lay the foundation for the formation of surface fibers and the first pore through orderly stacking, ensuring that the liquid surface has the fine pore structure required for nanoscale interception.

[0085] ② Controlling the initial morphology of the separation layer to avoid structural defects: The synergy of high humidity and airflow not only acts on the outermost layer of the liquid film, but also forms a gradient diffusion inward, which is conducive to the formation of separation fibers of ideal thickness. The short-time pre-phase separation of 1-7s precisely controls the phase separation depth, which is conducive to the formation of a sufficiently dense nanoscale structure of ideal thickness in the separation layer (to meet the retention accuracy), while avoiding the separation layer from being too dense and too thick (to prevent flux loss). At the same time, through the gradient exchange of solvent and water vapor, the fiber thickness inside the separation layer presents an orderly distribution of "fine on the surface and slightly coarser in the inner layer", which is suitable for the functional requirements of "precise retention + efficient mass transfer".

[0086] ③ It provides a stable substrate for subsequent solidification and phase separation. The initial surface structure formed by pre-phase separation has a certain rigidity, which can fix the basic shape of the separation layer before entering the solidification bath, avoiding structural collapse or disordered expansion of pores due to large-scale solvent escape during subsequent solidification and phase separation. This "shape-first, solidify-later" mode ensures that the separation layer eventually forms an ideal stable structure and effectively connects with the pre-filter layer.

[0087] Next, coagulation and phase separation are performed. The pre-separated liquid film is placed in a coagulation bath for phase separation treatment to obtain a green film. The phase separation treatment time is 20-50 seconds. The coagulation bath includes water and a third solvent, and the mass content of the third solvent in the coagulation bath is 5%-20%. The coagulation bath uses a combination of water and a third solvent (the third solvent is at least one of dimethyl sulfoxide, N-ethylpyrrolidone, N-methylpyrrolidone, and dimethylacetamide) to appropriately slow down the phase separation rate and avoid the formation of various disordered macropores, which would affect the overall strength of the membrane. The 20-50 second phase separation time ensures that the liquid film is completely solidified. Preferably, the coagulation bath temperature is 10-25℃ and 10-45℃ lower than that of the casting solution. The phase separation kinetics are further controlled by the temperature difference to promote the formation of a gradient pore structure (dense on the surface and loose with high porosity in the inner layer), which is conducive to the formation of an ideal transition zone and a contaminant-holding zone.

[0088] Finally, the membrane obtained in S3 is washed and dried to obtain a polyethersulfone nanoporous filter membrane. This filter membrane has excellent mechanical strength and high flux while meeting the requirements for nanoscale impurity rejection. It is suitable for nanoscale ultrapure filtration of mild systems in semiconductor manufacturing processes, such as the preparation of ultrapure HF diluent and deionized water.

[0089] The present invention also provides an application of a polyethersulfone nanoporous filter membrane, which is used for nanoscale ultrapure filtration of mild systems in semiconductor manufacturing processes.

[0090] Based on the physical and chemical properties of polyethersulfone (PES) material, the filter membrane prepared by this invention has good hydrophilicity and chemical compatibility, and is suitable for use in low-temperature weak acid, alkali and aqueous solutions. It exhibits excellent and stable filtration performance for nanoscale impurities (preferably 1-10 nm) in common WEC solutions such as DHF (hydrofluoric acid), DIW (deionized water), ST250 (photoresist remover), CO2 water (carbon dioxide aqueous solution) and BOE (buffered oxide etching solution, composed of hydrofluoric acid and ammonium fluoride).

[0091] The beneficial effects of this invention are as follows: The polyethersulfone nanoscale porous filter membrane of this invention comprises a main body, which includes a pre-filtration layer and a separation layer for retaining impurities. The thickness of the separation layer is 0.4-13 μm. The separation layer includes separation fibers forming a porous structure, and the SEM average diameter of the separation fibers is 15-55 nm. The average pore size (PMI) of the filter membrane is 10-80 nm, and the porosity is 40%-80%. The fiber density of the filter membrane is 1.1-1.5 g / cm³. 3This filter membrane, based on nanoscale pores, achieves high efficiency in retaining nanoscale impurities (especially 1-10 nanometer particles) while also exhibiting excellent mechanical strength and high flux through the synergistic effect of a thin separation layer, high fiber density, appropriate fiber thickness, and high porosity. Furthermore, the membrane boasts high cleanliness, low dissolution, and easy cleaning, making it suitable for ultrapure nanoscale filtration in mild systems during semiconductor manufacturing processes, such as preparing ultrapure HF diluents and deionized water. The preparation method provided by this invention allows for convenient, rapid, and effective preparation of the aforementioned polyethersulfone nanoscale porous membrane. Attached Figure Description

[0092] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0093] Figure 1 This is a scanning electron microscope (SEM) schematic diagram of the longitudinal section of the polyethersulfone nanoscale porous filter membrane prepared in Example 1, with a magnification of 700×.

[0094] Figure 2 This is a scanning electron microscope (SEM) schematic diagram of the longitudinal section of the polyethersulfone nanoscale porous filter membrane prepared in Example 1 near the liquid outlet surface, with a magnification of 10000×.

[0095] Figure 3 This is a scanning electron microscope (SEM) schematic diagram of the longitudinal section of the polyethersulfone nanoscale porous filter membrane prepared in Example 1, near the liquid outlet surface, with a magnification of 50000×.

[0096] Figure 4 This is a scanning electron microscope (SEM) image of the liquid outlet surface on the polyethersulfone nanoporous filter membrane prepared in Example 3, with a magnification of 50000×.

[0097] Figure 5 This is a scanning electron microscope (SEM) image of the liquid inlet surface of the polyethersulfone nanoporous filter membrane prepared in Example 3, with a magnification of 2000×.

[0098] Figure 6 This is a scanning electron microscope (SEM) schematic diagram of the longitudinal section of the polyethersulfone nanoporous filter membrane prepared in Example 3 near the liquid outlet surface, with a magnification of 10000×. Detailed Implementation

[0099] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0100] The following examples illustrate the polyethersulfone nanoscale porous filter membrane of the present invention. Example 1

[0101] A method for preparing a polyethersulfone nanoscale porous filter membrane includes the following steps:

[0102] S1: Prepare a casting solution and cast it onto a carrier to form a liquid film; the casting solution comprises the following components by weight: 20 parts polyethersulfone, 50 parts first solvent, 16 parts volatile second solvent and 8 parts small molecule organic additives; the viscosity of the casting solution is 11000 cps.

[0103] The first solvent is dimethyl sulfoxide; the second solvent is acetone; the small molecule organic additive is propylene glycol; and the temperature of the casting solution is 55℃.

[0104] S2: Pre-phase separation, the liquid film is placed in an environment with a relative humidity of 85%, and an airflow with a flow rate of 2.5 m / s is blown onto the surface of the liquid film to make the liquid film undergo pre-phase separation, wherein the pre-phase separation time is 5 s;

[0105] S3: Coagulation and phase separation. The pre-phase separated liquid film is introduced into a coagulation bath for phase separation treatment to obtain a green film. The phase separation treatment time is 30s. The coagulation bath includes water and a third solvent. The mass content of the third solvent in the coagulation bath is 8%. The third solvent is dimethyl sulfoxide. The temperature of the coagulation bath is 12℃, which is 43℃ ​​lower than the temperature of the casting solution.

[0106] S4: Cleaning and drying. The biofilm obtained in S3 is cleaned and dried to obtain a polyethersulfone nanoporous filter membrane. Example 2

[0107] A method for preparing a polyethersulfone nanoscale porous filter membrane includes the following steps:

[0108] S1: Prepare a casting solution and cast it onto a carrier to form a liquid film; the casting solution comprises the following components by weight: 18 parts polyethersulfone, 52 parts first solvent, 14 parts volatile second solvent, and 9 parts small molecule organic additives; the viscosity of the casting solution is 10000 cps.

[0109] The first solvent is N-ethylpyrrolidone; the second solvent is tetrahydrofuran; the small molecule organic additive is isopropanol; and the temperature of the casting solution is 50℃.

[0110] S2: Pre-phase separation, the liquid film is placed in an environment with a relative humidity of 83%, and an airflow with a flow rate of 2.1 m / s is blown onto the surface of the liquid film to cause pre-phase separation of the liquid film, wherein the pre-phase separation time is 4 s;

[0111] S3: Coagulation and phase separation. The pre-phase separated liquid film is introduced into a coagulation bath for phase separation treatment to obtain a green film. The phase separation treatment time is 33s. The coagulation bath includes water and a third solvent. The mass content of the third solvent in the coagulation bath is 10%. The third solvent is N-ethylpyrrolidone. The temperature of the coagulation bath is 15℃, which is 35℃ lower than the temperature of the casting solution.

[0112] S4: Cleaning and drying. The biofilm obtained in S3 is cleaned and dried to obtain a polyethersulfone nanoporous filter membrane. Example 3

[0113] A method for preparing a polyethersulfone nanoscale porous filter membrane includes the following steps:

[0114] S1: Prepare a casting solution and cast it onto a carrier to form a liquid film; the casting solution comprises the following components by weight: 17 parts polyethersulfone, 55 parts first solvent, 12 parts volatile second solvent and 10 parts small molecule organic additives; the viscosity of the casting solution is 9000 cps.

[0115] The first solvent is N-methylpyrrolidone; the second solvent is acetone; the small molecule organic additive is ethanol; and the temperature of the casting solution is 45℃.

[0116] S2: Pre-phase separation, the liquid film is placed in an environment with a relative humidity of 80%, and an airflow with a flow rate of 1.8 m / s is blown onto the surface of the liquid film to make the liquid film undergo pre-phase separation, wherein the pre-phase separation time is 3s;

[0117] S3: Coagulation and phase separation. The pre-phase separated liquid film is introduced into a coagulation bath for phase separation treatment to obtain a green film. The phase separation treatment time is 36s. The coagulation bath includes water and a third solvent. The mass content of the third solvent in the coagulation bath is 12%. The third solvent is N-methylpyrrolidone. The temperature of the coagulation bath is 20℃, which is 25℃ lower than the temperature of the casting solution.

[0118] S4: Cleaning and drying. The biofilm obtained in S3 is cleaned and dried to obtain a polyethersulfone nanoporous filter membrane. Example 4

[0119] A method for preparing a polyethersulfone nanoscale porous filter membrane includes the following steps:

[0120] S1: Prepare a casting solution and cast it onto a carrier to form a liquid film; the casting solution comprises the following components by weight: 16 parts polyethersulfone, 58 parts first solvent, 9 parts volatile second solvent and 12 parts small molecule organic additives; the viscosity of the casting solution is 8000 cps.

[0121] The first solvent is dimethylacetamide; the second solvent is tetrahydrofuran; the small molecule organic additive is isopropanol; and the temperature of the casting solution is 35℃.

[0122] S2: Pre-phase separation, the liquid film is placed in an environment with a relative humidity of 75%, and an airflow with a flow rate of 1.5 m / s is blown onto the surface of the liquid film to make the liquid film undergo pre-phase separation, wherein the pre-phase separation time is 2 s;

[0123] S3: Coagulation and phase separation. The pre-phase separated liquid film is introduced into a coagulation bath for phase separation treatment to obtain a green film. The phase separation treatment time is 40s. The coagulation bath includes water and a third solvent. The mass content of the third solvent in the coagulation bath is 14%. The third solvent is dimethylacetamide. The temperature of the coagulation bath is 24℃, which is 11℃ lower than the temperature of the casting solution.

[0124] S4: Cleaning and drying. The biofilm obtained in S3 is cleaned and dried to obtain a polyethersulfone nanoporous filter membrane.

[0125] Structural testing

[0126] The polyethersulfone porous filter membranes prepared in Examples 1-4 were characterized structurally. The morphology of the main membrane structure of each example sample was characterized by scanning electron microscopy (Hitachi S-5500), and the required data were obtained. Other tests were also conducted to obtain the corresponding data. The specific results are shown in Tables 1-4.

[0127] Table 1

[0128] Example 1 Example 2 Example 3 Example 4 Separation layer thickness / μm 8.5 6.4 4.2 2.1 SEM average diameter of separated fibers (nm) 25 30 36 48 PMI average pore size / nm 15 20 30 45 Porosity / % 45 51 60 72 Fiber density / g / cm3 1.45 1.38 1.32 1.25 SEM average diameter of surface fibers (nm) 10 16 25 35 SEM average pore size of the first hole (nm) 16 22 33 50 The ratio between the SEM average diameter of the surface fibers and the SEM average pore size of the first pore 0.625 0.727 0.758 0.700 Filter membrane thickness / μm 75 80 88 95

[0129] Table 2

[0130] Example 1 Example 2 Example 3 Example 4 The ratio between the SEM average diameter of the separated fibers and the SEM average diameter of the surface fibers. 2.500 1.875 1.440 1.371 The ratio between the SEM average aperture of the second hole and the SEM average aperture of the first hole. 55 49 41 32 Pore ​​area ratio of the liquid inlet surface / % 15 21 28 36 Pore ​​area ratio at the liquid outlet / % 12 16 21 25 The ratio between the pore area ratio of the inlet surface and the pore area ratio of the outlet surface 1.25 1.31 1.33 1.44 SEM average pore size of the second pore (μm) 876 1086 1342 1579

[0131] Table 3

[0132] Example 1 Example 2 Example 3 Example 4 The ratio of the thickness of the contamination zone to the overall membrane thickness 0.780 0.769 0.759 0.736 Contamination zone thickness / μm 58.5 61.5 66.8 69.9 Thickness of the transition region / μm 8 12 17 23 Pre-filter layer thickness / μm 66.5 73.5 83.8 92.9 SEM average diameter of supporting fibers (nm) 29 36 44 57 The ratio between the SEM average diameter of the supporting fibers and the SEM average diameter of the separated fibers. 1.16 1.20 1.22 1.19

[0133] Table 4

[0134] Example 1 Example 2 Example 3 Example 4 SEM average major diameter of finger-shaped apertures / μm 55.8 57.9 62 63 SEM average short diameter of finger-shaped apertures / μm 7.2 10.1 13.6 15.7 The ratio of the SEM mean major axis to the SEM mean minor axis of the finger-shaped apertures 7.75 5.73 4.56 4.01 SEM average length of the first distance / μm 1.4 2.1 2.6 3.4 The ratio between the first distance and the SEM mean short diameter of the finger aperture 0.194 0.208 0.191 0.217 The nearest SEM average distance between the finger-shaped orifice and the liquid inlet surface / μm; 2.7 3.6 4.8 6.9 Under a positive pressure of 0.03 MPa and a temperature of 20℃, the time (in seconds) required for 50 ml of water to pass through a filter membrane with a diameter of 47 mm. 137 102 46 19 Tensile strength / MPa 6.2 6.1 5.9 5.2 Elongation at break / % 20 25 32 33

[0135] The retention efficiency test (experimental equipment: Tianjin Logan KB-3 particle counter) showed that: the polyethersulfone porous filter membrane prepared in Example 1 had a retention efficiency of more than 90% for impurity particles with a particle size of 1 nm; the polyethersulfone porous filter membrane prepared in Example 2 had a retention efficiency of more than 90% for impurity particles with a particle size of 2 nm; the polyethersulfone porous filter membrane prepared in Example 3 had a retention efficiency of more than 90% for impurity particles with a particle size of 5 nm; and the polyethersulfone porous filter membrane prepared in Example 4 had a retention efficiency of more than 90% for impurity particles with a particle size of 10 nm. This indicates that the polyethersulfone porous filter membranes prepared in Examples 1-4 all have good retention efficiencies for impurity particles with a particle size of 1-10 nanometers, which can meet the needs of practical applications.

[0136] Each filter membrane (prepared in Examples 1-4) after rinsing and drying was made to have an effective filtration area of ​​0.2 m². 2 The filter element was then immersed in the 0K73 test reagent (a mixed solvent of 70% by mass propylene glycol monomethyl ether and 30% by mass propylene glycol monomethyl ether acetate) for 24 hours, and then the impurity content in the test reagent was measured (the impurities mainly include metal ions, which were measured by ICPMS). The test results are as follows: the amount of metal ion precipitation is not greater than 10 ppt.

[0137] Each filter membrane (prepared in Examples 1-4) after rinsing and drying was made to have an effective filtration area of ​​0.2 m². 2 The filter element was then rinsed with ultrapure water at a flow rate of 20L and a flow rate of 500ml / min. The downstream filtrate was then subjected to a total organic carbon (TOC) test (testing instrument: total organic carbon analyzer), and the TOC leaching amount was found to be no more than 0.5ppb.

[0138] This demonstrates that the polyethersulfone porous filter membranes prepared in Examples 1-4 all have good cleanliness and are suitable for nanoscale ultrapure filtration in mild systems during semiconductor manufacturing processes, such as preparing ultrapure HF diluents and deionized water.

[0139] In summary, the polyethersulfone porous filter membranes prepared in Examples 1-4 all have ideal membrane structures. They can efficiently retain nanoscale impurities (especially 1-10 nanometer impurity particles), while also exhibiting excellent mechanical strength, high flux, and fast filtration speed. Comparative Example 1

[0140] The process was carried out in accordance with Example 3, except that in the preparation of polyethersulfone nanoscale porous filter membrane, no small molecule organic additives were added to the casting solution in step S1, but 22 parts by mass of the polymer pore-forming agent polyvinylpyrrolidone were added; and the phase separation treatment time during solidification and phase separation in S3 was 15s; the other conditions remained unchanged.

[0141] The final polyethersulfone nanoporous filter membrane had a very low fiber density, less than 1.1 g / cm³. 3 It is only 1.05 g / cm³. 3 This results in the polyethersulfone filter membrane having excessively low tensile strength, only 2.3 MPa, making it impractical. Comparative Example 2

[0142] The procedure was carried out in accordance with Example 3, except that in the preparation of the polyethersulfone nanoporous filter membrane, the mass fraction of polyethersulfone (PES) in the casting solution in step S1 was 12 parts; and in the pre-phase separation in step S2, the relative humidity was 98%, the flow rate was 5 m / s, and the other conditions remained unchanged.

[0143] In the final polyethersulfone nanoporous filter membrane, the separating fibers are too fine, with an average SEM diameter of less than 15 nm, only 12 nm; this results in low tensile strength and elongation at break of the filter membrane, with a tensile strength of only 2.5 MPa and an elongation at break of 9%, making it impractical. Comparative Example 3

[0144] The procedure was carried out in accordance with Example 3, except that in the preparation of the polyethersulfone nanoporous filter membrane, the mass fraction of polyethersulfone (PES) in the casting solution in step S1 was 24 parts; and in the pre-phase separation in step S2, the relative humidity was 50% and the flow rate was 0.2 m / s; all other conditions remained unchanged.

[0145] In the final polyethersulfone nanoporous filter membrane, the separating fibers were too coarse, with an average SEM diameter greater than 55 nm and 60 nm. This resulted in a low flux of the polyethersulfone filter membrane. Under a positive pressure of 0.03 MPa and a temperature of 20 °C, it took 165 seconds for 50 ml of water to pass through the 47 mm diameter filter membrane. Furthermore, the cleanliness of the filter membrane was greatly reduced, which could not meet the requirements of semiconductor applications and was therefore impractical. Comparative Example 4

[0146] The procedure was carried out in accordance with Example 3, except that in the preparation of the polyethersulfone nanoscale porous filter membrane, the amount of the second solvent in the casting solution in step S1 was too large, and the amount added was 25 parts by mass, and the pre-phase separation time in step S2 was too long, which was 10s; the other conditions remained unchanged.

[0147] In the final polyethersulfone nanoporous filter membrane, the separation layer thickness was too large (15 μm) and the porosity was too low (30%). This resulted in a low flux of the polyethersulfone filter membrane. Under the conditions of a positive pressure of 0.03 MPa and a temperature of 20 °C, it took 168 seconds for 50 ml of water to pass through the 47 mm diameter filter membrane. Furthermore, the cleanliness of the filter membrane was greatly reduced, which could not meet the requirements of semiconductor applications and was not practical. Comparative Example 5

[0148] The procedure was carried out in accordance with Example 3, except that in the preparation of the polyethersulfone nanoscale porous filter membrane, the second solvent in the casting solution in step S1 was too little, with an addition amount of 3 parts by mass, and the pre-phase separation time in S2 was too short, at 0.5 s; the other conditions remained unchanged.

[0149] In the final polyethersulfone porous filter membrane, the separation layer was too small, almost non-existent, with a thickness of only 0.3 μm, and the porosity was too high, at 85%. As a result, the tensile strength and elongation at break of the polyethersulfone filter membrane were both too low, with a tensile strength of only 2.4 MPa and an elongation at break of 9.5%. Moreover, the retention efficiency was low, with a retention efficiency of less than 90% for 5 nm impurity particles, making it impractical.

[0150] The results of Comparative Examples 1-5 show that when the thickness of the separation layer in the prepared filter membrane is too small, the fiber density is too low, or the separation fibers are too fine, the mechanical strength of the filter membrane will be greatly reduced. During repeated rinsing, the membrane precision is easily affected, and the processing difficulty is too high, making it impractical. On the other hand, when the separation layer in the prepared filter membrane is too thick or the separation fibers are too coarse, the flux of the filter membrane is too small, and the cleanliness is too low, making it impractical as well.

[0151] Example 5 group

[0152] The examples in this group are prepared using the same method as in Example 3 to prepare polyethersulfone nanoscale filter membranes, except that:

[0153] In Example 5a, in step S1, the first solvent for preparing the casting solution was dimethyl sulfoxide, and the second solvent was ethyl acetate; in step S2, the relative humidity was 70%, the flow rate was 0.5 m / s, and the pre-phase separation time was 7 s; all other conditions remained unchanged.

[0154] Ultimately, the surface fibers of the resulting filter membrane had an excessively large SEM average diameter, greater than 35 nm, specifically 38 nm; and the separation layer thickness was 13 μm, resulting in a low filter membrane flux. Under positive pressure of 0.03 MPa and temperature of 20°C, it took 127 seconds for 50 ml of water to pass through a 47 mm diameter filter membrane; more ultrapure water was required for repeated rinsing to ensure membrane cleanliness, which was slightly worse than in Example 3; therefore, its practicality was lower than in Example 3.

[0155] In Example 5b, in step S1, the mass ratio of the first solvent to the second solvent in the casting solution is greater than 11, which is 12.6; wherein the first solvent is 63 parts and the easily volatile second solvent is 5 parts; and in step S2, the relative humidity is 95% and the flow rate is 3 m / s; the other conditions remain unchanged.

[0156] Ultimately, this resulted in an excessively small ratio between the average SEM diameter of the surface fibers and the average SEM pore size of the first pore in the prepared filter membrane, which was 0.5. At the same time, the filter membrane thickness was too small, less than 60 μm, only 50 μm; the separation layer thickness was 0.8 μm. As a result, the tensile strength of the filter membrane was low, with a tensile strength of 5 MPa and an elongation at break of 11%. Therefore, compared with Example 3, its practicality was lower.

[0157] In Example 5c, in step S1, the mass ratio of the first solvent to the second solvent in the casting solution is less than 3, which is 2.3; wherein the first solvent is 46 parts and the easily volatile second solvent is 20 parts; and in step S2, the relative humidity is 72% and the flow rate is 1 m / s; the other conditions remain unchanged.

[0158] Ultimately, this resulted in an excessively large ratio (1) between the average SEM diameter of the surface fibers and the average SEM pore size of the first pore in the prepared filter membrane. Simultaneously, the filter membrane thickness exceeded 100 μm, reaching 120 μm, while the separation layer thickness was only 11 μm. Consequently, the filter membrane flux was low. Under a positive pressure of 0.03 MPa and a temperature of 20°C, it took 110 seconds for 50 ml of water to pass through a 47 mm diameter filter membrane. Furthermore, more ultrapure water was required for repeated rinsing to maintain membrane cleanliness, which was slightly inferior to that of Example 3. Therefore, compared to Example 3, its practicality was lower.

[0159] Example 6 group

[0160] The examples in this group are prepared using the same method as in Example 3 to prepare polyethersulfone nanoscale filter membranes, except that:

[0161] In Example 6a, in step S1, the casting solution was prepared by adding 6 parts of small molecule organic additives and 18 parts of volatile second solvent; and the pre-phase separation time in S2 was 6s; the other conditions remained unchanged.

[0162] Ultimately, the ratio of the separating fibers to the surface fibers in the resulting filter membrane is less than 1.3, with a ratio of 1.2, and the fibers are basically the same thickness. The thickness of the filter membrane is 105 μm, resulting in a low filter membrane flux. Under the conditions of a positive pressure of 0.03 MPa and a temperature of 20°C, it takes 98 seconds for 50 ml of water to pass through a filter membrane with a diameter of 47 mm. Furthermore, more ultrapure water is needed for repeated rinsing to ensure the cleanliness of the membrane. The cleanliness is slightly worse than that of Example 3, so its practicality is lower than that of Example 3.

[0163] In Example 6b, in step S1, the casting solution was prepared by adding 14 parts of small molecule organic additives and 6 parts of volatile second solvent, and the pre-phase separation time was 1.5 s; all other conditions remained unchanged.

[0164] Ultimately, the ratio of the separating fibers to the surface fibers in the resulting filter membrane is greater than 4, specifically 4.5; and the filter membrane thickness is 58 μm. Consequently, the tensile strength of the resulting filter membrane is relatively low, with a tensile strength of 5.6 MPa and an elongation at break of 20%. Therefore, compared to Example 3, its practicality is lower.

[0165] Example 7 group

[0166] The examples in this group are prepared using the same method as in Example 3 to prepare polyethersulfone nanoscale filter membranes, except that:

[0167] In Example 7a, in step S1, 15 parts of small molecule organic additives were added to the casting solution; and in step S3, during coagulation and phase separation, the mass content of the third solvent in the coagulation bath was 5%; and the third solvent was different from the first solvent, being dimethylacetamide; the phase separation treatment time was 20s; and the other conditions remained unchanged.

[0168] Ultimately, the pore area ratio of the liquid inlet surface of the prepared filter membrane is less than 8%, reaching 6%; and the SEM average pore size of the second pore on the liquid inlet surface is less than 0.7 μm, reaching 650 μm; thus, the filter membrane flux is relatively low. Under the conditions of a positive pressure of 0.03 MPa and a temperature of 20°C, it takes 90 seconds for 50 ml of water to pass through a filter membrane with a diameter of 47 mm; and more ultrapure water is required for repeated rinsing to ensure membrane cleanliness. The cleanliness is slightly worse than that of Example 3, so its practicality is lower than that of Example 3.

[0169] In Example 7b: In step S1, 5 parts of small molecule organic additives were added to the casting solution, and the small molecule organic additives were glycerol; at the same time, in step S3, during coagulation and phase separation, the mass content of the third solvent in the coagulation bath was 15%; and the third solvent was different from the first solvent, being dimethyl sulfoxide; the phase separation treatment time was 50s; and the other conditions remained unchanged.

[0170] This ultimately resulted in a filter membrane with a pore area ratio of 45% at the liquid inlet surface, exceeding 40%, while the fiber density of the filter membrane was 1.15 g / cm³. 3 Ultimately, the tensile strength of the resulting filter membrane is relatively low, with a tensile strength of 5.5 MPa and an elongation at break of 19%. Therefore, its practicality is lower compared to Example 3.

[0171] In Example 7c: In step S1, the casting solution was prepared by adding 60 parts of the first solvent and 6 parts of the volatile second solvent, and the casting solution temperature was adjusted to 55°C; at the same time, in step S3, the solidification and phase separation was carried out by setting the temperature of the solidification bath to 5°C, which was 50°C lower than the temperature of the casting solution; the other conditions remained unchanged.

[0172] This results in a large difference in the pore area ratio between the inlet and outlet liquid surfaces of the prepared filter membrane, with a ratio of 2.1, and the separation layer thickness is 0.9 μm. Consequently, the tensile strength of the prepared filter membrane is relatively low, with a tensile strength of 5.9 MPa and an elongation at break of 25%. Therefore, compared with Example 3, its practicality is lower.

[0173] In Example 7d: In step S1, the casting solution was prepared by adding 48 parts of the first solvent and 19 parts of the volatile second solvent, and the casting solution temperature was adjusted to 33°C; at the same time, in step S3, the solidification and phase separation was carried out by setting the temperature of the solidification bath to 28°C, which was 5°C lower than the temperature of the casting solution; the other conditions remained unchanged.

[0174] This resulted in a smaller difference in pore area ratio between the inlet and outlet liquid surfaces of the prepared filter membrane, with a ratio of 1.05, and a separation layer thickness of 9.3 μm. Consequently, the filter membrane flux was lower. Under conditions of a positive pressure of 0.03 MPa and a temperature of 20°C, it took 79 seconds for 50 ml of water to pass through a 47 mm diameter filter membrane. Furthermore, more ultrapure water was required for repeated rinsing to ensure membrane cleanliness, which was slightly worse than that of Example 3. Therefore, compared to Example 3, its practicality was lower.

[0175] As can be seen from the above embodiments, the membrane structure, such as the thickness of the surface fibers, the ratio between the surface fibers and the pore size of the first pore, the thickness of the filter membrane, and the ratio between the separating fibers and the surface fibers, will have a certain impact on the membrane performance (such as flux, strength, and cleanliness).

[0176] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0177] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A polyethersulfone nanoscale porous filtration membrane, characterized in that: The main body has a non-directional tortuous path inside, The main body includes a pre-filter layer and a separation layer for trapping impurities, and the thickness of the separation layer is 0.4-13 μm; The separation layer includes separation fibers forming a porous structure, and the SEM average diameter of the separation fibers is 15-55 nm; The PMI average pore diameter of the filter membrane is 10-80nm, and the porosity is 40%-80%; the fiber density of the filter membrane is 1.1-1.5g / cm 3 .

2. The polyether sulfone nanoscale porous filter membrane according to claim 1, characterized in that: The separation layer includes a liquid outlet surface, which is located on the side of the separation layer away from the pre-filter layer; the liquid outlet surface contains a plurality of first holes, and the liquid outlet surface contains surface fibers for forming the first holes, and the SEM average diameter of the surface fibers is 3-35 nm; The ratio between the SEM average diameter of the surface fibers and the SEM average pore size of the first holes is 0.6-0.95; And / or, the thickness of the filter membrane is 60-100 μm.

3. The polyether sulfone nanoscale porous filter membrane according to claim 2, characterized in that: The ratio between the SEM average diameter of the separation fibers and the SEM average diameter of the surface fibers is 1.3-4; The pre-filter layer includes a liquid inlet surface, which is located on the side of the pre-filter layer away from the separation layer; the liquid inlet surface contains a plurality of second holes, and the ratio between the SEM average pore size of the second holes and The SEM average pore size of the first holes is not less than 10; The thickness of the separation layer is 1-9 μm.

4. The polyether sulfone nanoscale porous filter membrane according to claim 3, characterized in that: The hole area rate of the liquid inlet surface is 8%-40%, and the ratio between the hole area rate of the liquid inlet surface and the hole area rate of the liquid outlet surface is 1.1-2; And / or, the SEM average pore size of the second holes is not less than 0.7 μm.

5. The polyether sulfone nanoscale porous filter membrane according to claim 3 or 4, characterized in that: The pre-filter layer includes a pollution area and a transition area; the transition area is located on the side of the pre-filter layer close to the separation layer, the transition area is a sponge-like structure, and the porosity of the transition area is greater than that of the separation layer; the transition area includes support fibers for forming a porous structure, and the support fibers close to the separation layer are connected to the separation fibers; The pollution area is located on the side of the pre-filter layer away from the separation layer, and the pollution area is a finger-like hole structure, and the ratio between the thickness of the pollution area and the overall thickness of the membrane is not less than 0.3; The porosity of the filter membrane is 45%-75%.

6. The polyether sulfone nanoscale porous filter membrane according to claim 5, characterized in that: The thickness of the transition area is 5-40 μm; and the SEM average diameter of the support fibers is 20-65 nm; And / or, the ratio between the SEM average diameter of the support fibers and the SEM average diameter of the separation fibers is 1.05-1.

65.

7. The polyether sulfone nanoscale porous filter membrane according to claim 5, characterized in that: The pollution area has a plurality of finger holes, the extension direction of the long diameter of the finger hole is the thickness direction of the filter membrane; the SEM average long diameter of the finger hole is 30-80 μm; the ratio of the SEM average long diameter of the finger hole to the SEM average short diameter thereof is not less than 3.

8. The polyether sulfone nanoscale porous filter membrane according to claim 7, characterized in that: The distance between two adjacent finger holes in the thickness direction is a first distance, the SEM average length of the first distance is 0.8-5 μm, and the ratio between the first distance and the SEM average short diameter of the finger hole is not less than 0.

1.

9. The polyether sulfone nanoscale porous filter membrane according to claim 7, characterized in that: The nearest SEM average distance between the finger hole and the liquid inlet surface is 2-10 μm; The PMI average pore size of the filter membrane is 15-60 nm, and the fiber density of the filter membrane is 1.2-1.45 g / cm 3 .

10. The polyether sulfone nanoscale porous filter membrane according to claim 1, characterized in that: Under the conditions of a positive pressure of 0.03 MPa and a temperature of 20℃, the time required for 50 ml of water to pass through a filter membrane with a diameter of 47 mm is 10 s-160 s; The tensile strength of the filter membrane is not less than 5 MPa, and the elongation at break is not less than 10%; The TOC dissolution amount of the filter membrane is not more than 0.5 ppb, and the metal ion precipitation amount is not more than 10 ppt.

11. A method of making a polyethersulfone nanoscale porous filtration membrane according to any one of claims 1 to 10, wherein: Comprising the following steps: S1: preparing a casting solution and casting it onto a carrier to form a liquid film; the casting solution comprises the following substances in mass parts: 14-22 parts of polyether sulfone, 46-63 parts of a first solvent, 5-20 parts of a volatile second solvent, and 5-15 parts of a small-molecule organic additive; the viscosity of the casting solution is 5000-15000 cps; S2: pre-phase separation, placing the liquid film in an environment with a relative humidity of 70%-95%, and blowing an air flow with a flow rate of 0.5-3 m / s to the surface of the liquid film to make the liquid film undergo pre-phase separation, wherein the pre-phase separation time is 1-7 s; S3: solidification and phase separation, making the liquid film after pre-phase separation enter a coagulation bath for phase separation treatment to obtain a green membrane; wherein the phase separation treatment time is 20-50 s; the coagulation bath comprises water and a third solvent, and the mass content of the third solvent in the coagulation bath is 5%-20%; S4: cleaning and drying, cleaning and drying the green membrane obtained in S3 to prepare a polyether sulfone nanoscale porous filter membrane.

12. The preparation method of the polyether sulfone nanoscale porous filter membrane according to claim 11, characterized in that: The first solvent and the third solvent are at least one of dimethyl sulfoxide, N-ethyl pyrrolidone, N-methyl pyrrolidone, and dimethyl acetamide; The second solvent is acetone or tetrahydrofuran; The small-molecule organic additive is at least one of isopropyl alcohol, ethanol, ethylene glycol, and propylene glycol; The temperature of the casting solution is not less than 30℃.

13. The preparation method of the polyether sulfone nanoscale porous filter membrane according to claim 11, characterized in that: The temperature of the coagulation bath is 10-25℃, and is 10-45℃ lower than the temperature of the casting solution; and / or, the mass part ratio of the first solvent to the second solvent in the casting solution is 3-11.

14. Use of a polyethersulfone nanoscale porous filtration membrane according to any one of claims 1-10, characterized in that: The filter membrane is used for nanoscale ultra-pure filtration in a mild system in semiconductor processing.

Images