Sintered metal filter membranes and their preparation methods, filter elements, and methods for filtering supercritical carbon dioxide.

Abstract

This invention relates to a sintered metal filter membrane, its preparation method, filter element, and a method for filtering supercritical carbon dioxide. The sintered metal filter membrane is a hollow tubular structure comprising a coarse layer and a fine layer sintered together. The initial IPA (initial bubble point) of the sintered metal filter membrane is 50-400 kPa. The coarse layer includes a coarse porous body, the average width of the sintered neck of the coarse porous body is denoted as W, the average area equivalent diameter D of the pores in the coarse porous body is 0.8-6 μm, and W / D is not less than 0.04. The fine layer includes a fine porous body and attachment sites distributed within the fine porous body. The fine porous body is sintered and attached to the attachment sites, forming attached particles on the attachment sites. The average adhesion degree F of the attached particles on the attachment sites does not exceed 80%. The probability of interfacial delamination during the sintering process of this sintered metal filter membrane is very low.

Description

Technical Field

[0001] This invention relates to the field of filtration materials technology, and in particular to sintered metal filter membranes and their preparation methods, filter elements, and methods for filtering supercritical carbon dioxide. Background Technology

[0002] Sintered metal filter membranes are made from metal powder as raw material, formed through pressing / spraying / deposition filtration and further sintered at high temperatures. They can be used to separate impurities from fluids and achieve precision filtration. For example, they can be used to filter high-purity fluids required by the semiconductor and microelectronics manufacturing industries, as well as high-purity fluids required for etching, cleaning and other surface manufacturing processes.

[0003] The fluid can be in the form of gas or liquid, or in the form of supercritical fluid, such as supercritical carbon dioxide. Filtering supercritical carbon dioxide requires relatively harsh supercritical conditions, such as temperatures above 30°C and pressures exceeding 25MPa, 30MPa, 35MPa, or even 40MPa. This places extremely high demands on the pressure resistance of the filter element. Sintered metal filter membranes, due to their high mechanical strength, are suitable for filtering such high-pressure fluids.

[0004] USRE36249E1 discloses a high-porosity metal membrane filtration device for semiconductor industrial gas purification. The method for preparing the high-porosity metal membrane element includes: depositing a substantially uniform low-density sinterable dendritic material bed into a mold suitable for applying compressive force; compressing the low-density bed of sinterable dendritic material to form a green body; and finally sintering the green body at a temperature below the melting point of the metal material. The porous metal membrane element obtained by this method is a single-layer, sheet-like structure with relatively uniform pore size and porosity. However, the process gas to be treated may contain both large and small particulate contaminants. Therefore, the porous metal membrane element often experiences premature clogging of its pores by large-diameter particulate contaminants during use, resulting in a short service life for the filtration device. Furthermore, increasing the filtration area of ​​the sheet-like porous metal membrane element results in a very large overall size of the filtration device, which is unsuitable for semiconductor process installation and production line environments.

[0005] To address the shortcomings of existing single-layer, sheet-like porous metal membrane elements, such as premature clogging and difficulty in expanding the filtration area, hollow tubular double-layer or multi-layer sintered metal filter membranes with pore size gradients have been developed.

[0006] To obtain hollow tubular sintered metal filter membranes with pore size gradients, taking a double-layer sintered metal filter membrane as an example, a powder sintering process is generally adopted. First, coarse metal particles with larger particle sizes and fine metal particles with smaller particle sizes are loaded into an isostatic pressing mold and pressurized to form a hollow tubular thin compressed body. Then, the thin compressed body is sintered so that the parts of the particles in contact on their surfaces fuse together to form a sintered neck. During the filtration process, the coarse layer acts as a pre-filtration layer, retaining large-diameter particulate impurities, while the fine layer retains small-diameter particulate impurities. The combined effect of the coarse and fine layers achieves high retention efficiency for particulate impurities of various sizes. Simultaneously, this non-centralized retention structure also improves the overall dirt-holding capacity of the sintered metal filter membrane. Furthermore, the hollow tubular structure allows for the installation of sintered metal filter membranes with larger filtration areas within the same filter volume, making it more suitable for semiconductor manufacturing lines. In addition, the larger the size of the sintered neck formed by the coarse metal particles, the higher the bonding strength between the particles. Therefore, the coarse layer also supports the fine layer, improving the overall mechanical strength of the sintered metal filter membrane.

[0007] However, the double-layer hollow tubular sintered metal filter membrane produced by the above powder sintering process has a relatively high probability of delamination, that is, the probability of adjacent layers peeling off at the interface is very high, even exceeding 50%. Summary of the Invention

[0008] To address the high delamination probability issue in double-layered, hollow tubular sintered metal filter membranes, this application provides a sintered metal filter membrane and its preparation method. It also provides a filter element including the sintered metal filter membrane and a method for filtering supercritical carbon dioxide.

[0009] A sintered metal filter membrane, wherein the sintered metal filter membrane is in the shape of a hollow tube and includes a coarse layer and a fine layer sintered together. The fine layer is located radially outside the coarse layer. The surface of the fine layer facing away from the coarse layer is the outer surface, and the surface of the coarse layer facing away from the fine layer is the inner surface. The initial bubble point (IPA) of the sintered metal filter membrane is 50-400 kPa.

[0010] The coarse layer includes a coarse porous body, the average width of the sintering neck of the coarse porous body is W, the average area equivalent diameter D of the pores of the coarse porous body is 0.8-6μm, and W / D is not less than 0.04;

[0011] The fine layer includes a fine porous body and attachment sites distributed within the fine porous body. The fine porous body is sintered and attached to the attachment sites, and attachment particles are formed on the attachment sites. The average adhesion degree F of the attachment particles on the attachment sites does not exceed 80%.

[0012] It is generally believed that delamination at the interface of hollow tubular sintered metal filter membranes, comprising two layers, is usually caused by excessive sintering shrinkage of the inner layer and insufficient sintering shrinkage of the outer layer. This results in a greater radial inward shrinkage of the inner layer relative to the outer layer, and consequently, a significantly greater radial inward tensile force on the inner layer than on the outer layer, leading to delamination at the interface. Therefore, the delamination problem should be solved by reducing the sintering shrinkage of the inner layer and / or increasing the sintering shrinkage of the outer layer. However, the inventors of this application have discovered that the corresponding technical means not only fail to solve the delamination problem but actually exacerbate it.

[0013] The inventors of this application discovered that by controlling the average area equivalent diameter D of the pores in the coarse porous body to be 0.8-6 μm, and by ensuring that the ratio W / D of the average width W of the sintering neck of the coarse porous body to the area equivalent diameter D of the pores in the coarse porous body is not less than 0.04, and by ensuring that the average adhesion degree F of the adhering particles of the fine layer at the adhesion site is not greater than 80%, the probability of delamination at the interface between the coarse and fine layers can be significantly reduced. In other words, this application employs a method that controls the maximum sintering degree of the fine layer while simultaneously increasing the minimum sintering degree of the coarse layer. This method, which is generally considered to increase the probability of interface delamination, actually solves the problem of interface delamination between the coarse and fine layers, which is a very unexpected result.

[0014] When sintered metal filter membranes are prepared using metal powder as raw material through a specific sintering process, the metal powder always changes towards the direction of lower surface energy during sintering. Coarse metal particles fuse together to form sintering necks, and adjacent metal particles approach each other, causing macroscopic shrinkage in the coarse layer. A similar phenomenon occurs in the fine layer; that is, both coarse and fine layers exhibit a certain sintering shrinkage rate during sintering. Because smaller metal powder particles have a larger specific surface area and higher surface energy, they are more likely to spontaneously change to a lower energy state during sintering. Therefore, the original gas-solid interface tends to form a lower-energy solid-solid interface. Specifically, smaller particle sizes require less sintering driving force and are easier to sinter. Therefore, under the same sintering conditions, smaller metal particles are easier to sinter. Moreover, during sintering, smaller particles have a higher degree of sintering and better melt flow, making them more likely to approach each other. Therefore, the overall sintering shrinkage rate of the fine layer is usually higher than that of the coarse layer.

[0015] Further analysis reveals the reason for the delamination at the interface between the coarse and fine layers: In the sintered metal filter membrane provided in this application, the fine layer is located radially outward, and the coarse layer is located radially inward. The sintering shrinkage rate of the fine layer should be higher than that of the coarse layer. The fine layer, with its greater shrinkage, should be tightly bound to the outer periphery of the coarse layer, and delamination should not occur under these conditions. However, when the difference between the sintering shrinkage rate of the fine layer and the coarse layer (hereinafter referred to as the positive shrinkage rate difference) is too large, the fine layer, with its greater shrinkage, is tightly bound to the outer periphery of the coarse layer. Furthermore, due to the obstruction of the coarse layer, the fine layer cannot continue to shrink radially inward; that is, it can only shrink circumferentially. During the circumferential contraction of the fine layer relative to the coarse layer, any local area of ​​the fine layer at the interface between the fine and coarse layers will be simultaneously subjected to tangential tensile forces generated by the melting and contraction of the metal particles of the fine layer on both sides of the circumferential direction. When the tangential tensile force exceeds the interfacial bonding force between the fine and coarse layers, the local area of ​​the fine layer will slide along the direction tangential to the interface. This will result in the fine layer sliding circumferentially relative to the coarse layer at the macroscopic level, rather than moving radially inward. The final result is that the coarse and fine layers only form contact at the interface without forming an effective fusion, that is, the coarse and fine layers will delaminate at the interface.

[0016] This application controls the pore size D of the coarse layer to be 0.8-6 μm and the ratio W / D not less than 0.04, effectively controlling the coarse layer to have a relatively high degree of sintering, naturally resulting in a relatively high sintering shrinkage rate. This is because, during the sintering process, after the metal particles in the coarse layer are heated and melted, the surfaces of adjacent and contacting metal particles tend to move closer together and fuse into one. In this process, the distance between adjacent metal particles inevitably decreases, resulting in a reduction in macroscopic size. The fused joint between the fused adjacent metal particles solidifies to form a sintering neck, thereby firmly sintering the metal particles into a whole. Since the higher the degree of sintering, the greater the degree to which the molten metal particles move closer together and fuse into one, the smaller the distance between adjacent metal particles. This is reflected in the final sintered coarse layer, where the width of the sintering neck is larger and the size of the pores is smaller. Therefore, the larger the ratio W / D of the average width W of the sintering neck to the average area equivalent diameter D of the pores, the higher the degree of sintering of the coarse layer and the greater its macroscopic shrinkage rate.

[0017] Similarly, by controlling the F of the fine layer to be no greater than 80%, the fine layer effectively has a low degree of sintering, naturally resulting in a low sintering shrinkage rate. This is because, during the sintering process, the metal particles in the fine layer not only sinter together to form a fine porous matrix, but also inevitably fuse with the attachment sites, forming attached particles. Therefore, the higher the degree of sintering of the fine layer, the more metal particles sinter with the attachment sites, forming more attached particles. When the sintered metal filter membrane is damaged for microscopic observation of the cross-section, the attached particles still attached to the attachment sites can be observed. The adhesion degree of the attached particles at any attachment site is the ratio of the area of ​​the surface of the attachment site where the attached particles are attached to to the total surface area of ​​the attachment site. The average adhesion degree of multiple (e.g., more than 10) attachment sites is calculated as the average adhesion degree F of the attached particles of the fine porous matrix at the attachment sites. The higher the degree of sintering, the more particles adhere to the adhesion sites, resulting in a larger surface area where the particles are attached, and a higher degree of adhesion. Therefore, a higher average adhesion degree F of the particles at the adhesion sites in a fine porous matrix indicates a higher degree of sintering in the fine layers and a greater macroscopic shrinkage rate.

[0018] This application controls the average area equivalent diameter D of the pores in the coarse porous body within the range of 0.8-6 μm, and ensures that the ratio of the average width W of the sintering neck of the coarse porous body to the average area equivalent diameter D of the pores in the coarse porous body, W / D, is not less than 0.04. This corresponds to the minimum degree of sintering and the minimum sintering shrinkage rate of the coarse layer. Since D is 0.8-6 μm and the minimum value of W / D is 0.04, it indicates that the coarse layer, based on the average area equivalent diameter D of the pores being 0.8-6 μm, has at least a sintering neck with a width of not less than 0.04D. This means that the coarse layer has at least a certain degree of sintering and naturally has a certain degree of sintering shrinkage rate. Thus, the minimum degree of sintering of the coarse layer is relatively high, and its minimum sintering shrinkage rate is also relatively large, allowing the coarse layer to generate a sufficiently large radial inward shrinkage, thereby providing a sufficiently large accommodation space for the radial inward shrinkage of the fine layer.

[0019] Furthermore, in the fine layer, the average adhesion degree F of the attached particles at the attachment sites does not exceed 80%, which corresponds to the highest sintering degree and maximum sintering shrinkage rate of the fine layer. Since F does not exceed 80%, it indicates that at least 20% of the fine porous body is not sintered as a whole with the attachment sites. Therefore, the highest sintering degree of the fine layer will not be too high, and the maximum sintering shrinkage rate will not be too large. Thus, the shrinkage of the fine layer will not be too large. As mentioned earlier, since the radial inward shrinkage of the coarse layer is large when W / D takes the minimum value of 0.04, it provides a sufficiently large space for the radial inward shrinkage of the fine layer, which enables the fine layer to generate a sufficiently large radial inward shrinkage. Correspondingly, the circumferential shrinkage of the fine layer is greatly reduced. This greatly reduces the tangential tension caused by circumferential shrinkage in the fine layer. Thus, the tangential tension on the fine layer is not greater than the interfacial bonding force between the fine layer and the coarse layer, and the probability of delamination between the fine layer and the coarse layer is significantly reduced.

[0020] In summary, on the one hand, a W / D ratio of not less than 0.04 ensures that the radial inward shrinkage of the coarse layer is large enough to provide sufficient space for the radial inward shrinkage of the fine layer. On the other hand, a F ratio of not more than 80% ensures that the shrinkage of the fine layer itself is not too large. In addition, the coarse layer provides sufficient space for the radial inward shrinkage of the fine layer, and the circumferential shrinkage of the fine layer is greatly reduced. Therefore, the tangential tensile force on the fine layer will not be greater than the interfacial bonding force between the fine and coarse layers. This greatly reduces the probability of the fine layer sliding circumferentially relative to the coarse layer along the interface. The fine and coarse layers form a strong fusion at the interface, thereby greatly reducing the probability of interfacial delamination.

[0021] Meanwhile, the initial bubble point of the sintered metal filter membrane is 50-400 kPa, which means that the pore size of the sintered metal filter membrane is in the nanometer range. In other words, the pore size of the fine layer is in the nanometer range, which can effectively retain particulate pollutants of 20-100 nm. The high degree of sintering of the coarse layer, combined with a D of 0.8-6 μm, can ensure its ability to retain large particulate impurities and enhance the fine layer.

[0022] It should be noted that the sintering of the coarse and fine layers is carried out simultaneously. In fact, the minimum sintering degree of the coarse layer is positively correlated with that of the fine layer. When W / D is at its minimum value of 0.04, both the coarse and fine layers are at their minimum sintering degree. At this point, the sintering degree of the coarse layer is sufficient to meet the usage requirements. Moreover, since the fine layer is easier to sinter, its sintering degree is often higher. However, the shrinkage rate of the fine layer is lower than that when F equals 80%, which means that the circumferential shrinkage of the fine layer is also less than that when F is 80%. Therefore, at the sintering degree corresponding to a W / D value of 0.04, the radial inward shrinkage of the coarse layer is large enough to provide sufficient space for the radial inward shrinkage of the fine layer. Furthermore, the tangential tensile force on the fine layer is less than the interfacial bonding force between the fine and coarse layers. Thus, the probability of the fine layer sliding circumferentially relative to the coarse layer along the interface is even lower.

[0023] Correspondingly, the maximum sintering degree of the fine layer and the sintering degree of the coarse layer are also positively correlated. In the fine layer, when the average adhesion degree F of the attached particles at the adhesion site reaches the maximum value of 80%, both the fine layer and the coarse layer are at the maximum sintering degree. At this time, the sintering degree of the fine layer is controlled to be not too high. Although the coarse layer is relatively more difficult to sinter, its sintering degree is still higher than the sintering degree corresponding to W / D of 0.04, and the shrinkage rate is also higher than the shrinkage rate corresponding to W / D of 0.04. Therefore, the radial inward shrinkage generated by the coarse layer is also much greater than the radial inward shrinkage corresponding to W / D of 0.04. It provides a larger space for the radial inward shrinkage of the fine layer, so the fine layer can generate a larger radial inward shrinkage, and its circumferential shrinkage is further reduced. Thus, the tangential tensile force on the fine layer is less than the interfacial bonding force between the fine layer and the coarse layer. Therefore, the probability of the fine layer sliding circumferentially relative to the coarse layer along the interface is lower.

[0024] The average width W of the sintered neck of the coarse porous body can be obtained by the following method: import the SEM image of the inner surface of the sintered metal filter membrane (i.e. the surface of the coarse layer facing away from the fine layer) into measurement software such as NanoMeasure, measure the width of at least 10 sintered necks, and calculate the average value of the width of the 10 sintered necks as the average width W of the sintered neck.

[0025] Obtaining the equivalent diameter D of the pores in a coarse porous body: For example, import the SEM image of the inner surface of a sintered metal filter membrane into ImageJ measurement software, and use the software to directly measure the average area of ​​the pores in the coarse layer. The specific steps are as follows:

[0026] ① Open the image and adjust the pixels; Open the image in the image processing software ImageJ, and select Image>Type>8-bit to convert the electron microscope image into an 8-bit grayscale image;

[0027] ② Set the scale; Use the line tool from the toolbar to draw a straight line within the scale area, making the drawn line the same length as the scale on the electron microscope image. Select Analyze > Set Scale, enter the length of the scale in "Known Distance" (for example, enter 10), enter the unit in "Unit of length" (for example, enter um), and select OK.

[0028] ③ Image binarization; Select Image > Adjust > Threshold... to divide the electron microscope image into thresholds to obtain a binarized image;

[0029] ④ Set measurement parameters; In Analyze>Set Measurements, ensure that “Area” is checked, and then select OK;

[0030] ⑤ Data Measurement; In Analyze > Analayze Particles, set “Size (pixel^2)” to 0-Infinity, set “Show” to “Outlines”, and ensure that the boxes for “Display results”, “Exclude on edges”, and “Include holes” are checked. Select OK, then select Results > Summarize. The average area of ​​the holes will be measured in the Results screen.

[0031] The average area equivalent diameter D is further calculated from the average area S of the hole measured in the above steps, using the following formula: Of course, the average area of ​​the pores in the coarse layer can also be measured using other measurement software, and the equivalent diameter D of the average area can be calculated further.

[0032] The method for measuring the average adhesion degree F of adhering particles at the adhesion site in the fine layer is as follows:

[0033] The cross-sectional SEM image of the sintered metal filter membrane is imported into the NanoMeasurer software. The shape of the outer contour of the adhesion site may be circular, elliptical, rectangular, square, etc. The diameter, major / minor axis, length / width, or side length of the outer contour of the adhesion site are measured. The approximate value of the total area S of the outer contour of the adhesion site is then calculated. Similarly, the approximate value of the sum of the areas ΔS of the outer contours of all the adhesion particles attached to the surface of the adhesion site is calculated. The ratio of ΔS to S, ΔS / S*100%, is the adhesion degree of the adhesion particles on the surface of that adhesion site. The adhesion degree of the surfaces of more than 10 adhesion sites is calculated, and the average value is taken as the average adhesion degree F.

[0034] It should be noted that the aforementioned measurement methods are for reference only, and other methods that can obtain the corresponding results are also acceptable.

[0035] Preferably, the W / D ratio is not less than 0.1 and the F ratio is not more than 70%; and / or, the W / D ratio is not more than 9 and the F ratio is not less than 8%.

[0036] A W / D ratio of not less than 0.1 indicates a further improvement in the minimum sintering degree of the coarse layer, a further increase in the minimum sintering shrinkage rate, and a further increase in its radial inward shrinkage. This provides a larger space for the radial inward shrinkage of the fine layer. Combined with further control of F not exceeding 70%, this means a further decrease in the maximum sintering degree of the fine layer. The reduction in the maximum sintering shrinkage rate of the fine layer, combined with the larger radial inward shrinkage space of the coarse layer, inevitably leads to a reduction in the circumferential shrinkage of the fine layer. This means that the maximum tangential tensile force on the fine layer also decreases, i.e., the tangential tensile force is less than the interfacial bonding force between the coarse and fine layers. The probability of the fine layer sliding circumferentially relative to the coarse layer along the interface is further reduced, thus the probability of interface delamination is also lower.

[0037] With a W / D ratio not exceeding 9 and a F ratio not exceeding 70%, it ensures that neither the coarse nor fine layer is severely overburned, guaranteeing that each layer has a certain porosity and a sufficiently large pore size, thereby providing the necessary coarse and fine filtration performance and higher retention efficiency. With a F ratio not less than 8% and a W / D ratio not less than 0.1, it ensures that both the coarse and fine layers are more fully sintered, the overall filter membrane has sufficiently high strength, and on the basis that the coarse and fine layers each have high porosity and a large pore size, it can better adapt to the filtration conditions of supercritical fluids.

[0038] In addition, the W / D ratio should not exceed 9 and the F ratio should not be less than 8%. By controlling the maximum sintering degree of the coarse layer and the minimum sintering degree of the fine layer, the result of the maximum shrinkage rate of the coarse layer minus the minimum shrinkage rate of the fine layer (hereinafter referred to as the negative shrinkage rate difference) will not be too large. In this way, the radial inward shrinkage of the coarse layer will always match the radial inward shrinkage of the fine layer. As a result, the difference between the radial inward tension on the coarse layer and the radial inward tension on the fine layer will not be too large. On the basis that the difference between the two is less than the bonding force between the coarse and fine layers, the interface delamination problem caused by this will not occur.

[0039] Preferably, when F does not exceed 50%, W / D and F 2 The ratio is 15-50; when F exceeds 50%, the ratio of W / D to F is 8-18.

[0040] When the sintering shrinkage rate of the fine layer is large, preferably, the sintering shrinkage rate of the coarse layer should also be large to provide a larger space for radial inward shrinkage, thereby reducing the tangential tensile force caused by the circumferential shrinkage of the fine layer and thus reducing the probability of interface delamination. Conversely, when the sintering shrinkage rate of the fine layer is small, preferably, the sintering shrinkage rate of the coarse layer should not be too large. On the one hand, it is necessary to avoid excessive sintering shrinkage rate of the coarse layer, which would lead to excessive radial inward tensile force on the coarse layer. On the other hand, allowing the fine layer to be appropriately clamped to the outer periphery of the coarse layer can promote sintering bonding between the fine and coarse layers at the interface, thereby improving the interfacial bonding force between the two layers and helping to reduce the probability of interface delamination. However, the W / D value of the coarse layer and the F value of the fine layer are not linearly related.

[0041] In the initial stage of sintering, the coarse particles in the coarse layer have a large specific surface area and tend to sinter towards the direction with lower surface energy, meaning the sintering driving force is relatively large. The width of the sintering neck of the coarse layer increases rapidly and the equivalent area diameter of the pores decreases rapidly simultaneously. Therefore, as the degree of sintering of the coarse layer increases, the rapid changes of these two factors cause the W / D value of the coarse layer to increase rapidly. As the degree of sintering further increases, the surface energy becomes lower, and the rate of increase in the width of the sintering neck of the coarse layer decreases accordingly, as does the rate of decrease in the equivalent area diameter of the pores. This manifests as a slower rate of increase in the W / D value of the coarse layer. Therefore, for the coarse layer, the W / D value initially increases rapidly with the increase in the degree of sintering, and although it continues to increase thereafter, the rate of increase slows down.

[0042] Correspondingly, in the initial stage of sintering, although the fine particles in the fine layer also exhibit rapid sintering due to their strong tendency to sinter towards lower surface energy, the rate of change of F (fine particles) is relatively fast under a single influencing factor, as only the adhesion degree of the attached particles at the adhesion sites needs to be considered. However, inevitably, the rate of increase of F is lower than that of increase of W / D (fine surface energy) in the coarse layer. As the sintering degree of the fine layer further increases, the surface energy of the fine layer also decreases to a lower level. At this point, the sintering speed of the fine layer slows down, and the rate of increase of F also slows down. However, as mentioned above, the sintering difficulty of the fine layer is lower than that of the coarse layer. Therefore, the sintering rate of the fine layer should be faster than that of the coarse layer, and when the sintering degree is high, the rate of change of F in the fine layer is closer to that in the coarse layer.

[0043] Specifically in this application, when F is no greater than 50%, the rate of change of W / D relative to F is greater, and W / D and F... 2The ratio of W / D to F is 15-50. When F exceeds 50%, the rate of change of W / D relative to F is relatively small, with the ratio of W / D to F being 8-18. In other words, W / D has a corresponding value for F that is not greater than 50% and is greater than 50%. Thus, for a certain shrinkage rate of the fine layer, the coarse layer will produce a matching shrinkage rate. That is, for a certain shrinkage amount of the fine layer, the coarse layer will form a matching shrinkage amount. Therefore, during the shrinkage process of the fine and coarse layers, the fine layer is always tightly bound to the outer periphery of the coarse layer, which can increase the interfacial bonding force between the coarse and fine layers to a certain extent. Moreover, due to the matching shrinkage rates of the two, the coarse layer forms a sufficient radial inward shrinkage amount to provide a large enough space to accommodate the radial shrinkage of the fine layer. Thus, the fine layer can produce a larger radial inward shrinkage amount, thereby significantly reducing its circumferential shrinkage amount. Correspondingly, the tangential tensile force on the fine layer will also be significantly reduced, making the tangential tensile force much smaller than the interfacial bonding force between the coarse and fine layers. The probability of the fine layer circumferentially slipping relative to the coarse layer is greatly reduced.

[0044] Preferably, the ratio of the average maximum Fret diameter to the average minimum Fret diameter of the hole is 1.3-5, and the W / D ratio is not greater than 8.

[0045] The size and shape of the pores in the coarse porous matrix are related to the size and shape of the metal particles forming the coarse porous matrix through sintering. The ratio of the average maximum Fret diameter to the average minimum Fret diameter of the pores characterizes the irregularity of the pore shape, indirectly reflecting the irregularity of the metal particle shape. The smaller the ratio of the average maximum Fret diameter to the average minimum Fret diameter of the pores, the lower the irregularity of the pore shape, and the more spherical the pore shape tends to be. Correspondingly, the irregularity of the metal particle shape is also lower, that is, the more uniform the anisotropic dimensions of the metal particles are, and the closer the shape of the metal particles is to a sphere.

[0046] In this application, the ratio of the average maximum Fret diameter to the average minimum Fret diameter of the controlled pores is not less than 1.3, meaning that the shape of the pores differs somewhat from that of a sphere. This results in sufficiently large differences in the anisotropic dimensions of the metal particles forming the coarse porous body. The difference in particle shape from a sphere is beneficial for increasing the sintering degree of the coarse layer (because non-spherical particles have a higher specific surface area and surface energy, making them easier to sinter). It also increases the radial inward shrinkage of the coarse layer, thereby increasing the radial inward shrinkage of the fine layer and reducing the circumferential shrinkage of the fine layer, further reducing the tangential tensile force on the fine layer. Moreover, the difference in particle shape from a sphere often makes the surface of the coarse layer near the fine layer more uneven, increasing the contact and bonding area between the two layers, thus increasing the interfacial bonding force and further increasing the difficulty of circumferential slippage of the fine layer relative to the coarse layer, significantly reducing the probability of circumferential slippage of the fine layer relative to the coarse layer.

[0047] Simultaneously, the ratio of the average maximum Fret diameter to the average minimum Fret diameter of the controlled pores is not greater than 5, and the W / D ratio is not greater than 8. This controls the maximum difference in the dimensions of the controlled pores in all directions, which in turn controls the maximum difference in the dimensions of the metal particles in all directions. It also controls the degree to which the shape of the metal particles deviates from a spherical shape, which facilitates better pressing of the metal particles and better control of the sintering degree of the coarse layer. This ensures that the sintering degree does not exceed the sintering degree corresponding to a W / D value of 10, thus preventing the coarse layer from being over-burned and ensuring that the coarse layer has a high porosity after sintering, thereby providing better pre-filtration performance. It also prevents the difference in negative shrinkage rates between the coarse and fine layers from being too large, so that the radial inward shrinkage of the coarse and fine layers always remains matched. This ensures that the difference between the radial inward tension on the coarse layer and the radial inward tension on the fine layer is less than the interfacial bonding force between the coarse and fine layers, thereby reducing the probability of interfacial delamination caused by this.

[0048] The maximum and minimum Freret diameters of the pores in the coarse layer are obtained using a similar method to the method described above for measuring the average area of ​​the pores. The difference lies in setting the measurement parameters in step ④; in Analyze>Set Measurements, ensure that "Feret's diameter" is checked. In the final result, "Feret" and "MinFeret" represent the maximum and minimum Freret diameters, respectively. Measure the maximum and minimum Freret diameters of the pores in at least 10 sintered metal filter membrane products. Further calculate the average of these 10 maximum Freret diameters to obtain the average maximum Freret diameter of the pores, and the average of these 10 minimum Freret diameters to obtain the average minimum Freret diameter of the pores. Then, further calculate the ratio of the average maximum Freret diameter to the average minimum Freret diameter of the pores.

[0049] Preferably, the average maximum Fret diameter of the pores in the coarse layer is 1.5-8 μm, and the average minimum Fret diameter is 0.6-3 μm.

[0050] Based on the ratio of the average maximum Fret diameter to the average minimum Fret diameter of the pores being 1.3-5, the average maximum Fret diameter of the pores is further limited to 1.5-8 μm and the average minimum Fret diameter to 0.6-3 μm. This indicates that the size of the coarse particles is also approximately 1-8 μm, meaning that the coarse particles are neither too large nor too small, and the irregularity of their shape is neither too high nor too low. This allows for a moderate increase in the sintering degree of the coarse particles, thereby improving the interfacial bonding force between the coarse and fine layers and reducing the probability of interfacial delamination caused by the circumferential slippage of the fine layer relative to the coarse layer. However, this also prevents the coarse particles from being severely overburned, thus avoiding interfacial delamination problems caused by excessive negative shrinkage.

[0051] Preferably, the interface between the coarse layer and the fine layer is wavy, with a peak near the outer surface and a trough near the inner surface. The distance between the trough and the outer surface is defined as H1, and the distance between the peak and the outer surface is defined as H2. The average value of H2 is not less than 50 μm, and H1 > H2 with an average difference of 2-5 μm.

[0052] The interface between the coarse and fine layers cannot be a smooth surface; instead, it extends in a wavy, tortuous manner. Generally, to improve the filtration stability of the fine layer, it is desirable to increase the uniformity of its thickness. This is because the thinnest region of the fine layer always has lower fluid resistance during fluid filtration, and the fluid naturally tends to flow towards the region with lower fluid resistance. Therefore, the thinnest region of the fine layer often has a higher fluid filtration load. However, if the smaller channels in this region become clogged with impurities due to the higher load, the fluid can only pass through the larger channels in that region for filtration. Since the larger channels have slightly lower retention capacity, the probability of small impurities leaking out increases.

[0053] However, for the metal filter membrane in this application, the wavy interface between the fine and coarse layers can increase the area of ​​the interface, which is beneficial to increasing the interfacial bonding force between the coarse and fine layers. Moreover, when the fine layer tends to slip circumferentially, the mechanical interlocking generated by the uneven interface can block this slippage tendency, thereby further reducing the probability of interface delamination.

[0054] Specifically, in this application, the average difference between the distance H1 between the trough and the outer surface and the distance H2 between the peak and the outer surface at the interface between the coarse and fine layers is not less than 2 μm, that is, the average height difference between the peak and the trough is not less than 2 μm. This makes the interface area larger, the interfacial bonding force on the fine layer greater, and the resistance to the circumferential slip of the fine layer relative to the coarse layer stronger. This is beneficial to further reduce the probability of delamination caused by the circumferential slip of the fine layer relative to the coarse layer.

[0055] Furthermore, the average difference between the distance H1 between the trough and the outer surface at the interface between the coarse and fine layers and the distance H2 between the trough and the outer surface is controlled to be no greater than 5 μm, that is, the average height difference between the trough and the trough is no greater than 5 μm, and the average value of H2 is no less than 50 μm, that is, the average thickness of the thinnest region of the fine layer is no less than 50 μm. This avoids excessive unevenness at the interface, ensures that the thinnest region of the fine layer is also thick enough, and the degree of unevenness is small (the average thickness fluctuation is no higher than 10%). The overall flow resistance of the fine layer with its meandering internal pores is still relatively uniform. The fluid filtered by the coarse layer will be more evenly distributed to all parts of the interface, so that the pores in the thinnest region of the fine layer will not be blocked prematurely, and the resulting decrease in retention efficiency will not occur.

[0056] In short, by controlling the interface between the coarse and fine layers to have an uneven, wavy shape, and specifically controlling the height difference between the peaks and troughs, the ability of the sintered metal filter membrane of this application to resist interfacial delamination can be significantly improved, while having little impact on filtration and retention capabilities.

[0057] The measurement methods for the distance H1 between the trough and the outer surface, and the distance H2 between the peak and the outer surface are as follows: import the cross-sectional SEM image of the sintered metal filter membrane into the NanoMeasurer measurement software, measure the distance H1 between the trough and the outer surface at more than 3 locations within the field of view, measure the distance H2 between the peak and the outer surface at more than 3 locations, calculate the average value of the difference between H1 and H2, and obtain the average value of H2.

[0058] Preferably, the difference between the maximum distance between the trough and the outer surface and the minimum distance between the crest and the outer surface does not exceed 8 μm.

[0059] By controlling the difference between the maximum distance between the trough and the outer surface and the minimum distance between the crest and the outer surface to not exceed 8μm, the maximum value of the height difference between the crest and the trough (i.e., the difference between the maximum value of H1 and the minimum value of H2) is not greater than 8μm. Thus, the difference between the maximum and minimum thickness of the fine layer is not greater than 8μm. In this way, the thinnest area of ​​the fine layer is thick enough, the degree of unevenness is small (the maximum value of the thickness fluctuation is not higher than 16%), and the overall flow resistance of the fine layer with its meandering internal pores is small. The fluid filtered by the coarse layer can be more evenly dispersed to all parts of the interface. The probability of premature blockage of the pores in the thinnest area of ​​the fine layer is extremely low, thus providing higher retention efficiency.

[0060] Preferably, the average angle A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers at the crest or trough is 2-30°, and F is not greater than 60%.

[0061] The interface between the coarse and fine layers extends in a wavy, tortuous manner, giving it a certain angle of inclination relative to the tangent at the crest or trough. This increases the interface area, thereby increasing interfacial bonding and reducing the probability of interfacial delamination. However, due to the difference in shrinkage rates between the coarse and fine layers on both sides of the interface (the fine layer shrinks more than the coarse layer), if the angle of inclination is too large (i.e., the interface is too steep), the degree of intersection between the coarse and fine layers is too high. In the intersection area, the fine and coarse layers restrict each other's flow and shrinkage, leading to local stress concentration. This results in local delamination of the coarse and fine layers in the intersection area, which in turn reduces the interfacial bonding.

[0062] In this application, the average angle A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers is controlled to be no less than 2°. Combined with an average height difference of 2-5 μm between crests and troughs, this ensures a sufficiently high inclination at the interface, meaning the distance between adjacent crests and troughs is not excessive, resulting in a larger interface area and greater interfacial bonding force on the fine layer. This strengthens the resistance to circumferential slippage of the fine layer relative to the coarse layer, further increasing the difficulty of circumferential slippage and reducing the probability of interface delamination. Furthermore, the phase... The average angle A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers at the crest or trough is no greater than 30°, and F is no greater than 60%. At the same time, the maximum inclination of the interface and the maximum sintering degree of the fine layer are controlled, that is, the degree of mutual intersection between the coarse and fine layers are controlled. This prevents the fine and coarse layers from restricting each other's flow and shrinkage in the area where they intersect, avoids local stress concentration, and thus avoids the phenomenon of local delamination of the coarse and fine layers in the area where they intersect. This prevents the bonding force at the interface from decreasing, thereby reducing the probability of interface delamination.

[0063] The method for measuring the average angle A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers at the crest or trough is as follows: Import the cross-sectional SEM image of the sintered metal filter membrane into the NanoMeasurer measurement software. Within the field of view, measure at least three sets of angles A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers at the crest or trough. Specifically, draw the line connecting adjacent crests and troughs, the tangent at the interface at the crest or trough, and the perpendicular line segments intersecting the aforementioned line and tangent at both ends. The line, tangent, and perpendicular line segments form a right triangle. The length of the perpendicular line segment is equal to the difference between H1 and H2. The value of angle A can be obtained by using the arcsine. Calculate at least three values ​​of A in this way, and then calculate the average value. Alternatively, import the cross-sectional SEM image of the sintered metal filter membrane into CAD drawing software, and after drawing the connecting lines between adjacent peaks and troughs and the tangents at the interface at the peaks or troughs, directly mark the angle A between the connecting lines and the tangents.

[0064] Preferably, the porosity of the sintered metal filter membrane is 12-40%, the pore area ratio of the inner surface is 10-40%, and the thickness of the coarse layer is 300-1000 μm.

[0065] The porosity of the sintered metal filter membrane was measured by the water displacement method. The specific operating steps are as follows:

[0066] First, the apparent volume V0 of the sintered metal filter membrane is calculated from the inner and outer diameters and axial height of the hollow tubular sintered metal filter membrane.

[0067] Then, the hollow tubular sintered metal filter membrane is completely immersed in water of volume V1, and a vacuum is applied for 30 minutes to ensure that the water completely fills the pores of the hollow tubular sintered metal filter membrane. The total volume V2 of water immersing the hollow tubular sintered metal filter membrane is then recorded. Therefore, the porosity of the sintered metal filter membrane is... .

[0068] The porosity of the inner surface can be measured as follows: import the SEM image of the inner surface of the sintered metal filter membrane into ImageJ image processing software, and use the software to calculate the sum of the areas of all pores on the inner surface and the total area of ​​the inner surface. The ratio of the sum of the areas of all pores to the total area of ​​the inner surface is the porosity.

[0069] In this application, the porosity of the sintered metal filter membrane is 12-40%, which enables the sintered metal filter membrane to provide high retention efficiency and sufficiently high flux, meeting certain filtration performance requirements. At the same time, the pore area ratio of the inner surface is 10-40%, and the overall porosity of the coarse layer is close to 10-40%, so the coarse layer has a suitable degree of sintering, avoiding over-sintering that would cause pore blockage, and also avoiding under-sintering. Combined with the thickness of the coarse layer being 300-1000μm, the radial inward shrinkage that the coarse layer can generate is large enough to provide a sufficiently large space for the radial inward shrinkage of the fine layer, which facilitates the increase of the radial inward shrinkage of the fine layer, thereby reducing its circumferential shrinkage, reducing the tangential tensile force on the fine layer, and reducing the probability of interfacial delamination.

[0070] Moreover, the overall porosity of the coarse layer is not much different from that of the overall porosity of the filter membrane, and thus the overall porosity of the fine layer is also not much different from that of the overall porosity of the filter membrane. This means that the fine layer also has a relatively high porosity, which also indicates that the sintering degree of the fine layer is not too high. As a result, the shrinkage rate and shrinkage amount of the fine layer are not too large, the circumferential shrinkage amount of the fine layer is not too large, and the tangential tensile force on the fine layer is not too large, which helps to reduce the probability of the fine layer slipping relative to the coarse layer in the circumferential direction.

[0071] A method for preparing the sintered metal filter membrane includes the following steps:

[0072] S1, the coarse particles with a particle size of 1-20μm are compressed for the first time to obtain a hollow tubular first embryo, wherein the pressure of the first compression is P1 and the time is t1;

[0073] S2, the fine-layer particles are placed radially outside the first green embryo and simultaneously compressed a second time to obtain a hollow tubular second green embryo. The fine-layer particles include fine particles and shrinkage particles. The particle size of the fine particles is 50-300 nm, and the particle size of the shrinkage particles is 0.5-2.5 μm. The mass of the shrinkage particles is 20-50% of the total mass of the fine particles and shrinkage particles. The pressure of the second compression is P2, the time is t2, 0.1≤P2 / P1≤0.4, 1.5≤(t1+t2) / t1≤3;

[0074] S3, the second green embryo is sintered under vacuum conditions. The holding temperature during sintering is T0 and the holding time is t0. After sintering, the sintered metal filter membrane is obtained by cooling.

[0075] In this application, coarse particles with a diameter of 1-20 μm undergo two compressions, S1 and S2, while fine particles with a diameter of 50-300 nm and shrinkage particles with a diameter of 0.5-2.5 μm undergo only one compression in step S2. Furthermore, P2 / P1 ≥ 0.1 and (t1+t2) / t1 ≥ 1.5, ensuring that the coarse particles undergo sufficiently high pressure and a sufficiently long compression time before sintering. This results in a sufficiently high density and a larger contact area between the coarse particles in the final second green compact. Since the contact points between metal particles are more easily sintered and fused together during sintering, the higher compression density and larger contact area of ​​the coarse particles make the coarse layer easier to sinter, thus facilitating a greater degree of sintering and a higher sintering shrinkage rate. Furthermore, ensuring that the compression pressure and compression time in step S2 are not too low or too short ensures that the fine layer particles are fully compressed to a certain density, so that all the fine layer particles can be sintered, resulting in a fine layer with sufficiently high strength and providing effective fine filtration performance.

[0076] Conversely, it is also necessary to control P2 / P1≤0.4 and (t1+t2) / t1≤3, that is, to control the compression pressure and compression time in step S2 to be not too high and not too long. On the one hand, after the coarse particles are compressed to a certain density, the effect of further increasing the compression pressure and compression time in step S2 on the compression density of the coarse layer gradually decreases, and the final increase in the density of the coarse layer is limited, and the sintering degree and sintering shrinkage rate of the coarse layer tend to be stable. On the other hand, the density of the fine layer particles increases continuously with the compression, which can easily lead to the fine layer obtaining an excessively high degree of sintering and an excessively large sintering shrinkage rate during the sintering process. Since the sintering shrinkage rate of the coarse layer tends to be stable, while the sintering shrinkage rate of the fine layer continues to increase, the fine layer is naturally subjected to higher tangential tensile force, and the probability of delamination at the interlayer interface increases accordingly.

[0077] Simultaneously, 20-50% by mass of shrinkage-reducing particles are mixed into the fine particles. The shrinkage-reducing particles have a particle size of 0.5-2.5 μm, while the fine particles have a particle size of 50-300 nm. In other words, a suitable amount of shrinkage-reducing particles with a larger particle size are mixed into the fine particles. Compared with the fine particles, the shrinkage-reducing particles with a larger particle size have poorer fluidity and are more difficult to compress and densify. Therefore, the addition of shrinkage-reducing particles can control the overall density of the fine layer particles in step S2, thereby reducing the amount of contact between fine particles and between fine particles and shrinkage-reducing particles before sintering. Moreover, the shrinkage-reducing particles with a larger particle size are more difficult to sinter. That is, the addition of shrinkage-reducing particles leads to lower density and higher sintering difficulty, which can moderately reduce the overall sintering degree of the fine layer, thereby reducing the shrinkage of the fine layer.

[0078] In other words, by controlling the number of compressions, compression pressure, and compression time, and by adding an appropriate amount of shrinkage-reducing particles, on the one hand, the coarse layer can provide a larger space for the radial inward shrinkage of the fine layer, facilitating a larger radial inward shrinkage amount in the fine layer; on the other hand, the total shrinkage rate and total shrinkage amount of the fine layer are controlled. Under this dual effect, the circumferential shrinkage amount of the fine layer is even smaller, ensuring that the tangential tensile force on the fine layer is less than the interfacial bonding force. As a result, in the sintered metal filter membrane obtained by sintering, the average area equivalent diameter D of the pores in the coarse layer is 0.8-6 μm, and the ratio of the average width W of the sintered neck of the coarse layer to the average area equivalent diameter D of the pores in the coarse layer is not less than 0.04. The average adhesion degree F of the attached particles in the fine layer at the adhesion site does not exceed 80%, the probability of the fine layer forming circumferential slip relative to the coarse layer is sufficiently low, and the probability of interfacial delamination is significantly reduced. At the same time, the initial bubble point of the IPA of the sintered metal filter membrane is 50-400 kPa, which can efficiently intercept particulate pollutants of 20-100 nm.

[0079] Preferably, the coarse particles are made of nickel or nickel carbonyl, and the fine particles and the shrinkage particles are made of stainless steel; in step S3, P3 does not exceed 0.01 Pa, the heat preservation temperature T0 is 780-820℃, and the heat preservation time t0 is 20-40 min.

[0080] The materials of the coarse, fine, and shrinkage-reducing particles are matched with the holding temperature and time to ensure that both the coarse and fine layers are fully sintered. In particular, it ensures that the coarse and fine layers form an effective fusion at the interface, increasing the interfacial bonding force and reducing the probability of delamination. The absolute pressure P3 during the sintering process does not exceed 0.01 Pa, resulting in a very low oxygen content in the sintering furnace. This prevents the metal particles from being oxidized, which would lead to a decrease in the strength of the final sintered metal filter membrane (because metal oxide particles and metal particles cannot be sintered together).

[0081] Preferably, in step S1, P1 is 70-120 MPa and t1 is 60-120 s.

[0082] Based on 0.1≤P2 / P1≤0.4 and 1.5≤(t1+t2) / t1≤3, further limits are made in step S1, where P1 is 70-120MPa and t1 is 60-120s. This ensures that the coarse particles have already been compressed to a large extent in step S1 and thus have high density. Combined with the pressing process in step S2, this ensures that there is a suitable amount of contact between fine particles and between fine particles and reduced-density particles. This allows the fine layer to have sufficient strength and provide effective fine filtration performance after sintering. At the same time, the coarse layer forms a sufficiently large radial inward contraction to provide a large enough space to accommodate the radial inward contraction of the fine layer, thereby increasing the radial contraction of the fine layer and reducing its circumferential contraction. This reduces the tangential tensile force on the fine layer, ensuring that the tangential tensile force is less than the interfacial bonding force and reducing the probability of interfacial delamination.

[0083] Preferably, in step S3, a buffer process is added after the heat preservation is completed and before the cooling process. The buffer process involves maintaining the temperature at the buffer temperature T1 for a duration of t3, where T0-10≤T1≤T0-5, and t3 is 5-10 minutes. Since the coarse layer is located radially inward and the fine layer is located radially outward, during the actual cooling process, the temperature of the coarse layer will be slightly higher than that of the fine layer, meaning the entire sintered body has a temperature gradient from the inside out. For typical cooling processes, due to the rapid temperature drop and the good thermal conductivity of the metal itself, although an unavoidable temperature gradient exists, the internal and external temperatures of the sintered body often drop below the sintering temperature within a very short time. Therefore, the actual time difference between the coarse and fine layers stopping sintering is relatively short.

[0084] This application adds a buffer process after the heat preservation is completed and before the cooling process, and controls the temperature of the buffer process to be only 5-10°C lower than the heat preservation temperature. Therefore, based on the inevitable temperature gradient between the inner and outer sides of the sintered body, since the outer surface temperature of the sintered body is only 10°C lower than the heat preservation temperature, the temperature difference between the inner radial side of the sintered body, such as the area where the coarse layer is located, and the outer surface is small, the cooling driving force is low, and the temperature drops slowly. This causes the area where the fine layer is located on the outer radial side of the sintered body to stop sintering because the temperature has dropped below the sintering temperature, while the area where the coarse layer is located on the inner radial side of the sintered body is still in a slow sintering state because the temperature drops slowly. Therefore, the actual time when the coarse layer stops sintering is slightly later than the actual time when the fine layer stops sintering. The specific temperature control during the buffering process allows the coarse layer to have an additional 5-10 minutes of slow sintering compared to the fine layer. This further improves the sintering degree of the coarse layer, enabling it to form a larger radial inward shrinkage. This provides a larger space for the radial inward shrinkage of the fine layer, facilitating a larger radial inward shrinkage of the fine layer and reducing its circumferential shrinkage. The probability of interface delamination is further reduced.

[0085] Preferably, the cooling process includes a first cooling stage and a second cooling stage. In the first cooling stage, the temperature decreases from the buffer temperature T1 to 480-600℃ at a rate of 0.8-1.5℃ / min, and in the second cooling stage, the temperature decreases from the final temperature of the first cooling stage to 25-80℃ at a rate of 1.8-4.5℃ / min.

[0086] After the buffering process is over, the temperature of the sintered body is still relatively high. If it is cooled directly and rapidly to room temperature, local stress concentration is likely to occur at the interface between the coarse and fine layers, which can lead to local delamination at the interface.

[0087] Therefore, in this application, after the buffering process, the sintered body first undergoes a first cooling stage with a relatively slow cooling rate, and then undergoes a second cooling stage with a faster cooling rate. In the first cooling stage, the sintered body decreases from the buffer temperature T1 to 480-600℃ at a rate of 0.8-1.5℃ / min. In the second cooling stage, the sintered body decreases from the final temperature of the first cooling stage to 25-80℃ at a rate of 1.8-4.5℃ / min. That is, the temperature of the sintered body is first reduced from the buffer temperature to 480-600℃ at a slower cooling rate. Because the cooling rate is slower, the problem of stress concentration at the interface can be avoided, thereby avoiding the problem of local stress concentration. This ensures that the fine layer is always tightly bound to the outer periphery of the coarse layer, increasing the interfacial bonding force and reducing the probability of interfacial delamination. Then, the temperature is reduced to 25-80℃ at a faster cooling rate so that the sintered body can be cooled at a faster rate to obtain a sintered metal filter membrane and improve the cooling efficiency.

[0088] Furthermore, this application also provides a filter element comprising the sintered metal filter membrane described above.

[0089] Furthermore, this application also provides a method for filtering supercritical carbon dioxide, wherein supercritical carbon dioxide is passed through the sintered metal filter membrane during filtration. Attached Figure Description

[0090] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0091] Figure 1 This is a schematic diagram of the inner surface SEM image of the sintered metal filter membrane product of this application;

[0092] Figure 2 This is a schematic cross-sectional SEM image of the sintered metal filter membrane product of this application;

[0093] Figure 3 This is a schematic diagram of the filter element structure of this application;

[0094] Figure 4 This is a schematic diagram of the structure of the filter in this application.

[0095] Figure 3-4 In the middle, 100-filter element, 10-sintered metal filter membrane, 20-end plate, 30-connector component, 31-hollow channel, 200-outer shell, 210-connector. Detailed Implementation

[0096] The technical solution of the present invention will be further described below through specific embodiments. However, those skilled in the art will understand that the following embodiments are only for illustrating the present invention and should not be regarded as limiting the scope of the present invention. Unless otherwise specified, specific conditions in the embodiments are performed under conventional conditions or conditions recommended by the manufacturer. Reagents or instruments used without specified manufacturers are all commercially available conventional products. Example 1

[0097] Step S1: Provide coarse particles and a compression mold. The compression mold includes a compression mandrel, an annular sleeve on its outer periphery, and two plugs at both ends of the mandrel's axial direction. An annular cavity is formed between the compression mandrel, the annular sleeve, and the two plugs. The coarse particles are filled in the annular cavity and located on the outer periphery of the compression mandrel. The filling thickness of the coarse particles is 345.0 μm. The coarse particles are subjected to a first isostatic compression to obtain a hollow tubular first green embryo. The coarse particles are carbonyl nickel powder with a particle size of 6-8 μm and an average particle size of 7.1 μm. The aspect ratio of the coarse particles is 2.72. The compression pressure P1 is 70 MPa and the compression time t1 is 60 s.

[0098] Step S2: After the first isostatic compression, the annular sleeve is ejected radially outward. If necessary, a larger annular sleeve is replaced, and a gap is formed between the outer periphery of the first green preform and the inner wall of the annular sleeve. The fine layer particles are placed in this gap, i.e., filled radially on the outer side of the first green preform obtained in step S1. The filling thickness of the fine layer particles is 60.8 μm. The fine layer particles and the first green preform are simultaneously subjected to a second isostatic compression to obtain a hollow tubular second green preform. The fine layer particles include uniformly mixed fine particles and shrinkage-reducing particles. Both the fine particles and shrinkage-reducing particles are made of 316 stainless steel. The particle size of the fine particles is 200-300 nm with an average particle size of 250 nm, and the particle size of the shrinkage-reducing particles is 1.5-2.5 μm with an average particle size of 2.1 μm. The mass of the shrinkage-reducing particles is 49.81% of the total mass of the fine particles and shrinkage-reducing particles. The compression pressure P2 is 8 MPa, and the time t2 is 50 s.

[0099] Step S3: The second green embryo obtained in step S2 is sintered under a pressure P3 of 0.002 Pa. The holding temperature T0 during the sintering process is 780℃ and the holding time t0 is 40 min.

[0100] After the heat preservation process, the sintered body undergoes a cooling treatment, which includes a first cooling stage and a second cooling stage. In the first cooling stage, the temperature decreases from the heat preservation temperature T0 of 780℃ to 480℃ at a rate of 1.5℃ / min. In the second cooling stage, the temperature decreases from the final temperature of the first cooling stage (480℃) to 50℃ at a rate of 1.8℃ / min, thereby obtaining the sintered metal filter membrane. Specific raw material selection and process parameters are detailed in Tables 1, 2-1, and 2-2.

[0101] To clarify, the ratio of the major and minor diameters of the coarse particles is obtained as follows: Select any 10 coarse particles from the SEM image of the raw material. Draw four intersecting line segments on the surface of each particle. The intersection point of the line segments represents the approximate geometric center of the coarse particle. The start and end points of the line segments represent the intersection points between the line segments and the edges of the coarse particles. The angle between any two adjacent line segments is approximately 45 degrees. Measure the length of each line segment using software such as NanoMeasure. Calculate the ratio of the maximum to the minimum length of these four line segments; this is the ratio of the major and minor diameters of the coarse particle. Alternatively, the maximum and minimum Frett diameters of the coarse particles in the SEM image can be measured using ImageJ software. The maximum Frett diameter is the major diameter of the coarse particle, and the minimum Frett diameter is the minor diameter. Furthermore, the particle size is the average of the major and minor diameters. Example 2

[0102] The main difference between Example 2 and Example 1 is that in step S1, the coarse particles are carbonyl nickel powder with a particle size of 10-14 μm and an average particle size of 12.6 μm, and the aspect ratio of the coarse particles is 2.27. The compression pressure P1 is higher and the compression time t1 is longer. In step S2, the fine particles have a particle size of 50-150 nm and an average particle size of 100 nm, and the shrinkage particles have a particle size of 0.5-1.5 μm and an average particle size of 1.32 μm. The amount of shrinkage particles added is less, accounting for 24.21% of the total mass of the fine particles and shrinkage particles. The compression pressure P2 is higher and the compression time t2 is longer. In step S3, the holding temperature T0 is higher. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 3

[0103] The main difference between Example 3 and Example 2 is that in step S1, the aspect ratio of the coarse particles is 1.38, the compression pressure P1 is lower, and the compression time t1 is shorter; in step S2, the amount of shrinkage-reducing particles added is slightly higher, accounting for 26.25% of the total mass of the fine particles and shrinkage-reducing particles, and the compression time t2 is shorter; in step S3, the heat preservation temperature T0 is lower. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 4

[0104] The main difference between Example 4 and Example 2 is that in step S1, the coarse particles are nickel powder with a particle size of 15-18 μm and an average particle size of 16.2 μm, and the aspect ratio of the coarse particles is 5.11, with a higher compression pressure P1; in step S2, the amount of shrinkage-reducing particles added is slightly more, with a mass of 25.37% of the total mass of fine particles and shrinkage-reducing particles; in step S3, a buffering process is added after the heat preservation process and before the cooling process. This buffering process involves maintaining the temperature at the buffer temperature T1 for a time of t3, where the buffer temperature T1 is 805℃ and the buffering time t3 is 5 min. Correspondingly, the first cooling stage of the cooling process is a temperature drop from the buffer temperature T1 of 805℃ to 600℃ at a rate of 0.8℃ / min, and the second cooling stage is a temperature drop from the final temperature of the first cooling stage of 600℃ to 80℃ at a rate of 4.5℃ / min, thereby obtaining a sintered metal filter membrane. For details on the selection of raw materials and process parameters, please refer to Tables 1, 2-1 and 2-2. Example 5

[0105] The main difference between Example 5 and Example 1 is that in step S1, the coarse particles are carbonyl nickel powder with a particle size of 7-9 μm and an average particle size of 8.3 μm, and the aspect ratio of the coarse particles is 2.79; in step S2, the amount of shrinkage-reducing particles added is slightly less, and the mass is 49.33% of the total mass of fine particles and shrinkage-reducing particles. For specific raw material selection and process parameters, please refer to Tables 1, 2-1 and 2-2. Example 6

[0106] The main difference between Example 6 and Example 2 is that in step S1, the coarse particles are carbonyl nickel powder with a particle size of 9-12 μm and an average particle size of 10.9 μm, and the aspect ratio of the coarse particles is 2.46; in step S2, the amount of shrinkage particles added is slightly less, and the mass is 20.41% of the total mass of fine particles and shrinkage particles, the compression pressure P2 is higher, and the compression time t2 is longer; in step S3, a buffer process is added after the heat preservation process and before the cooling process. This buffer process is to continue to hold at the buffer temperature T1 for a time of t3, where the buffer temperature T1 is 805℃ and the buffer time t3 is 10 min. Correspondingly, the first cooling stage of the cooling process is that the temperature drops from the buffer temperature T1 of 805℃ to 530℃ at a rate of 1.0℃ / min, and the second cooling stage is that the temperature drops from the final temperature of the first cooling stage of 530℃ to 50℃ at a rate of 3.5℃ / min, thereby obtaining a sintered metal filter membrane. For details on the selection of raw materials and process parameters, please refer to Tables 1, 2-1 and 2-2. Example 7

[0107] The main difference between Example 7 and Example 4 is that in step S1, the coarse particles are nickel powder with a particle size of 16-20 μm and an average particle size of 18.2 μm, and the aspect ratio of the coarse particles is 6.22, resulting in a higher compression pressure P1; in step S2, the amount of shrinkage-reducing particles added is slightly less, accounting for 24.85% of the total mass of fine particles and shrinkage-reducing particles, resulting in a higher compression pressure P2; and in step S3, the holding temperature T0 is higher. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 8

[0108] The main difference between Example 8 and Example 4 is that in step S1, the coarse particles are nickel powder with a particle size of 16-20 μm and an average particle size of 18.2 μm, and the aspect ratio of the coarse particles is 5.02, with a longer compression time t1; in step S2, the fine particles have a particle size of 200-300 nm and an average particle size of 250 nm, and the shrinkage particles have a particle size of 1.5-2.5 μm and an average particle size of 2.1 μm, with a slightly smaller amount of shrinkage particles added, accounting for 21.98% of the total mass of the fine and shrinkage particles, resulting in a higher compression pressure P2 and a longer compression time t2; in step S3, the holding temperature T0 is higher, the holding time t0 is longer, the buffer temperature T1 is higher, and the buffer time t3 is longer. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 9

[0109] The main difference between Example 9 and Example 1 is that in step S1, the aspect ratio of the coarse particles is 2.79, the compression pressure P1 is higher, and the compression time t1 is longer; in step S2, the amount of shrinkage-reducing particles added is less, accounting for 39.72% of the total mass of fine particles and shrinkage-reducing particles, the compression pressure P2 is higher, and the compression time t2 is longer; in step S3, the heat preservation temperature T0 is higher. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 10

[0110] The main difference between Example 10 and Example 6 is that in step S1, the aspect ratio of the coarse particles is 4.85, and the compression pressure P1 is lower; in step S2, the amount of shrinkage-reducing particles added is greater, accounting for 31.63% of the total mass of the fine particles and shrinkage-reducing particles, resulting in a lower compression pressure P2 and a shorter compression time t2; in step S3, the heat preservation time t0 is shorter. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 11

[0111] The main difference between Example 11 and Example 1 is that in step S1, the coarse particles are carbonyl nickel powder with a particle size of 1-6 μm and an average particle size of 3.3 μm, and the aspect ratio of the coarse particles is 1.87; in step S2, the amount of shrinkage-reducing particles added is less, accounting for 48.51% of the total mass of fine particles and shrinkage-reducing particles, and the compression pressure P2 is higher and the compression time t2 is longer; in step S3, the holding temperature T0 is higher. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Example 12

[0112] The main difference between Example 12 and Example 5 is that in step S1, the coarse particles are nickel powder with a particle size of 16-20 μm and an average particle size of 18.2 μm, and the aspect ratio of the coarse particles is 2.91; in step S2, the fine particles have a particle size of 50-150 nm and an average particle size of 100 nm, the amount of shrinkage-reducing particles added is less, and the mass is 21.02% of the total mass of fine particles and shrinkage-reducing particles, the compression pressure P2 is higher, and the compression time t2 is longer; in step S3, the holding temperature T0 is higher. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Comparative Example 1

[0113] The main difference between Comparative Example 1 and Example 5 is that in step S1, the coarse particles are carbonyl nickel powder with a particle size of 20-25 μm and an average particle size of 23.4 μm, and the aspect ratio of the coarse particles is 2.85. The compression pressure P1 is too low and the compression time t1 is too short. In step S2, the compression pressure P2 is higher and the compression time t2 is longer. In step S3, the holding temperature T0 is higher. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2. Comparative Example 2

[0114] The main difference between Comparative Example 2 and Example 5 is that, in step S2, no shrinkage particles are mixed into the fine particles; the compression pressure P2 is too high and the compression time t2 is too long. For specific raw material selection and process parameters, please refer to Tables 1, 2-1, and 2-2.

[0115] Table 1. Raw material parameters for Examples 1-12 and Comparative Examples 1-2

[0116]

[0117] Table 2-1 Process parameters for Examples 1-12 and Comparative Examples 1-2

[0118] <![CDATA[P1 / MPa]]> <![CDATA[t1 / s]]> <![CDATA[P2 / MPa]]> <![CDATA[t2 / s]]> <![CDATA[P3 / Pa]]> <![CDATA[T0 / ℃]]> <![CDATA[t0 / min]]> <![CDATA[T1 / ℃]]> <![CDATA[t3 / min]]> Example 1 70 60 8 50 0.002 780 40 / / Example 2 110 100 18 90 0.006 810 30 / / Example 3 100 90 18 80 0.006 800 30 / / Example 4 120 100 18 90 0.002 810 30 805 5 Example 5 70 60 8 50 0.002 780 40 / / Example 6 110 100 30 100 0.006 810 30 805 10 Example 7 120 90 25 90 0.002 820 40 / / Example 8 120 120 30 100 0.002 820 40 815 10 Example 9 80 70 12 90 0.002 800 40 / / Example 10 90 100 15 90 0.008 810 20 / / Example 11 70 60 10 60 0.002 790 30 / / Example 12 70 60 30 100 0.002 810 30 / / Comparative Example 1 40 30 30 100 0.002 810 30 / / Comparative Example 2 70 60 45 250 0.006 780 40 / /

[0119] Table 2-2 Process parameters for Examples 1-12 and Comparative Examples 1-2 (continued)

[0120]

[0121] Performance testing methods and results

[0122] I. Morphological parameters of sintered metal filter membranes

[0123] The SEM images of the inner surface and cross-section of the sintered metal filter membranes prepared in Examples 1-12 and Comparative Examples 1-2, combined with manual measurement or NanoMeasure / ImageJ software, can be used to measure parameters such as the thickness of the coarse layer, the average width W of the sintered neck of the coarse layer, the average area equivalent diameter D of the pores, the average maximum and average minimum Frett diameter of the pores, the thickness of the fine layer, the average adhesion degree F of the adhering particles of the fine layer at the adhesion sites, the distances H1 and H2 between the troughs and crests of the interface between the coarse and fine layers and the outer surface of the filter membrane, and the angle A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers at the crest or trough. The SEM images of the inner surface and cross-section of the sintered metal filter membranes prepared in the embodiments of this application are shown below. Figure 1 and Figure 2 As shown.

[0124] For detailed testing methods, please refer to the invention description section.

[0125] II. Probability of Interfacial Delamination in Sintered Metal Filter Membranes

[0126] The testing method was as follows: The preparation methods of the sintered metal filter membranes in Examples 1-12 and Comparative Examples 1-2 were repeated five times each. Specifically, five sintered metal filter membrane products were prepared for each example and comparative example. The end of each sintered metal filter membrane was observed, and the length L of the interfacial delamination region of each sintered metal filter membrane was measured. The ratio of L to the perimeter L0 of the interface was calculated as L / L0*100%. The average value of L / L0 for the five filter membrane products in each example and comparative example was further calculated as the interfacial delamination probability of the corresponding product in that example or comparative example. It should be noted that because the end is more unique than the interior, the uniformity of stress during pressing and sintering is worse, resulting in a higher probability of interfacial delamination compared to the interior. Therefore, the interfacial delamination probability was calculated by observing the end.

[0127] III. Performance of Sintered Metal Filter Membranes

[0128] 3.1 Pressure Holding Performance of Sintered Metal Filter Membranes

[0129] The testing method is as follows: First, a filter element made from the sintered metal filter membrane prepared in Examples 1-12 is impregnated with molten paraffin and cooled to seal the membrane pores. Then, high-pressure water is injected into the hollow interior of the filter element. The pressure of the high-pressure water is gradually increased from 1 MPa, increasing by 1 MPa each time, and each pressure is maintained for 1 hour until the pressure of the high-pressure water suddenly drops (the filter element ruptures), thus measuring the highest pressure resistance value of the corresponding filter element. For example, when the pressure of the high-pressure water is 5 MPa, the pressure holding is passed; however, when the pressure of the high-pressure water is increased to 6 MPa, the pressure of the high-pressure water drops suddenly. It can be preliminarily determined that the pressure holding pressure of the filter element is not less than 5 MPa and less than 6 MPa. Further, the pressure of the high-pressure water is increased in increments of 0.1 MPa until the accurate pressure holding pressure is measured.

[0130] To clarify, during the pressure holding test, the downstream pressure of the filter element is atmospheric pressure, and the upstream pressure is the pressure of high-pressure water. At this time, the pressure difference between the upstream and downstream sides of the filter element is almost equal to the pressure of the high-pressure water. In fact, under the filtration conditions of supercritical carbon dioxide, the pressure difference between the upstream and downstream sides of the filter element generally does not exceed 2 MPa. Therefore, when the pressure holding pressure of the filter element reaches more than 10 MPa, it will almost not be damaged by the pressure during use (even if pressure fluctuations cause a sudden increase in pressure). Therefore, tests for higher pressure holding pressures are no longer conducted.

[0131] 3.2 Initial bubble point of IPA in sintered metal filter membranes

[0132] Test Procedure: First, completely impregnate the sintered metal filter membrane prepared in Examples 1-12 with IPA; then, slowly introduce gas (such as nitrogen) and gradually increase the pressure. The pressure at which the first continuous bubble appears at the outlet is the initial bubble point of the IPA. Other operational details can be found in ASTM F316-70 or ISO 2942:2018.

[0133] 3.3 Porosity of sintered metal filter membranes

[0134] For detailed testing methods, please refer to the invention description section.

[0135] The test results are recorded in Tables 3 and 4.

[0136] Table 3 Characterization results of fine and coarse layers

[0137]

[0138] Table 4 Performance test results of sintered metal filter membranes

[0139]

[0140] Data Analysis and Conclusions

[0141] The performance test results of Examples 5-6 and 12, and Comparative Examples 1-2 show that in Comparative Example 1, when the sintering degree of the coarse layer is too low, the average width W of the sintering neck of the coarse porous body is relatively small, only 0.13 μm, while the size of the pores in the coarse porous body is very large, with the average area equivalent diameter D of the pores exceeding 6 μm. This results in a very small ratio W / D of the average width W of the sintering neck to the average area equivalent diameter D of the pores, only 0.02. However, the sintering degree of the fine layer is higher, and the average adhesion degree F of the attached particles in the fine layer at their attachment sites is higher. The percentage was 78.90%, close to 80%. Due to the low degree of sintering of the coarse layer, its sintering shrinkage rate was too low, resulting in a small radial shrinkage and a small radial accommodation space for the fine layer. The interfacial bonding force between the coarse and fine layers was also very small. However, the degree of sintering of the fine layer was very high, and its sintering shrinkage rate was also very high. Because its radial shrinkage amplitude was very low, its circumferential shrinkage was very large, which generated a large tangential tensile force on the fine layer. Moreover, this tangential tensile force was much greater than the interfacial bonding force, so its delamination probability reached 56.8%, which exceeded 50%. In Comparative Example 2, when the sintering degree of the fine layer is too high, the average adhesion degree F of the adhering particles at their adhesion sites reaches 85.90%, exceeding 80%. Although the sintering degree of the coarse layer is also high, the ratio of the average width W of the sintered neck to the average area equivalent diameter D of the pores (W / D) reaches 0.04, providing a larger radial accommodation space for the layer. However, due to the excessively high sintering degree and sintering shrinkage rate of the fine layer, the circumferential shrinkage of the fine layer is too large, and the tangential tensile force on the fine layer is too large, far exceeding the interfacial bonding force. Therefore, its delamination probability even exceeds 60%. Thus, regardless of whether W / D is too low or F is too high, once either exceeds the range, the probability of interfacial delamination in the sintered metal filter membrane product exceeds 50% or even greater than 60%. In summary, both excessively low sintering degree of the coarse layer and excessively high sintering degree of the fine layer will cause the tangential tensile force on the fine layer to exceed the interfacial bonding force, greatly increasing the probability of interfacial delamination.

[0142] The performance test results of Examples 1-2 and Examples 5-6 show that when W / D and F are appropriately selected, i.e. W / D is not less than 0.1 and F is not more than 70%; or W / D is not more than 9 and F is not less than 8%, the probability of interfacial delamination of the corresponding sintered metal filter membrane is lower, all below 5%, and the filter membrane product also has a suitable porosity, which can provide reliable filtration performance, and its overall effect is more preferred.

[0143] The performance test results of Examples 3-4 and Examples 7-8 show that when the shape of the pores in the coarse layer deviates from the spherical shape to a suitable degree, that is, when the shape of the coarse particles deviates from the spherical shape to a suitable degree, the sintering degree of the coarse layer can be better controlled, the interfacial bonding force between the coarse and fine layers can be improved, and the over-burning of the coarse layer can be prevented. Specifically, the ratio of the average maximum Fret diameter to the average minimum Fret diameter of the pores is 1.3-5, and the W / D is not greater than 8. The physical interlocking between the interfaces makes the bonding force between the interfaces greater and the probability of interface delamination is lower.

[0144] The performance test results of Examples 3-4 and Examples 7 and 11 show that when the interface between the coarse and fine layers has an uneven, wavy shape, such that the average difference between the distance H1 between the trough and the outer surface and the distance H2 between the trough and the outer surface is 2-5 μm, and the average value of H2 is not less than 50 μm and H1>H2, the sintered metal filter membrane has a higher resistance to interface delamination, that is, a lower probability of interface delamination, and can provide better filtration and retention performance.

[0145] The above performance test results show that when the average area equivalent diameter D of the pores in the coarse porous matrix of the sintered metal filter membrane is 0.8-6 μm, the ratio of the average width W of the sintered neck to the average area equivalent diameter D of the pores (W / D) is not less than 0.04, and the average adhesion degree F of the attached particles at the adhesion sites in the fine porous matrix does not exceed 80%, the coarse layer can provide a space for the radial inward shrinkage of the fine layer that matches the sintering shrinkage of the fine layer. This can significantly reduce the circumferential shrinkage of the fine layer, so that the tangential tensile force on the fine layer is not greater than the interfacial bonding force between the fine and coarse layers. This greatly reduces the probability of the fine layer sliding circumferentially relative to the coarse layer along the interface. The fine and coarse layers form a firm fusion at the interface, significantly reducing the probability of interfacial delamination. At the same time, the initial bubble point of the IPA of the sintered metal filter membrane is 50-400 kPa, which can effectively trap particulate pollutants of 20-100 nm.

[0146] The general steps for fabricating filter element 100 from the sintered metal filter membranes obtained in Examples 1-12 are as follows: The closed end plate 20 and the connector component 30 with a hollow channel 31 are welded to both ends of the hollow tubular sintered metal filter membrane 10, respectively. The hollow channel 31 of the connector component 30 communicates with the surface of the coarse layer of the sintered metal filter membrane 10 away from the fine layer. A schematic diagram of the filter element 100 is shown below. Figure 3 The filter element 100 and the housing 200 are assembled together to form a filter, wherein the housing 200 has a connector 210, such as... Figure 4As shown. During filtration, the supercritical carbon dioxide to be filtered is introduced into the hollow interior of the filter element 100 through the hollow channel 31 of the connector component 30, and passes through the coarse layer and the fine layer in sequence. Large particulate pollutants and small particulate pollutants are intercepted in sequence, and the clean supercritical carbon dioxide flows out from the surface of the fine layer away from the coarse layer (the arrow indicates the direction of fluid flow), and finally is discharged from the connector 210 of the outer shell 200.

[0147] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0148] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A sintered metal filter membrane, characterized in that, The sintered metal filter membrane is a hollow tubular structure, comprising a coarse layer and a fine layer sintered together. The fine layer is located radially outside the coarse layer. The surface of the fine layer facing away from the coarse layer is the outer surface, and the surface of the coarse layer facing away from the fine layer is the inner surface. The initial bubble point of the IPA of the sintered metal filter membrane is 50-400 kPa. The coarse layer includes a coarse porous body, the average width of the sintering neck of the coarse porous body is denoted as W, the average area equivalent diameter D of the pores of the coarse porous body is 0.8-6μm, and W / D is not less than 0.04; The fine layer includes a fine porous body and attachment sites distributed within the fine porous body. The fine porous body is sintered and attached to the attachment sites, and attachment particles are formed on the attachment sites. The average adhesion degree F of the attachment particles on the attachment sites does not exceed 80%.

2. The sintered metal filter membrane according to claim 1, characterized in that, W / D not less than 0.1, F not more than 70%; and / or, W / D not more than 9, F not less than 8%.

3. The sintered metal filter membrane according to claim 1, characterized in that, When F does not exceed 50%, W / D and F 2 The ratio is 15-50; when F exceeds 50%, the ratio of W / D to F is 8-18.

4. The sintered metal filter membrane according to claim 1, characterized in that, The ratio of the average maximum Fret diameter to the average minimum Fret diameter of the hole is 1.3-5, and W / D is not greater than 8.

5. The sintered metal filter membrane according to claim 4, characterized in that, The average maximum Fret diameter of the pores in the coarse porous body is 1.5-8 μm, and the average minimum Fret diameter is 0.6-3 μm.

6. The sintered metal filter membrane according to claim 1, characterized in that, The interface between the coarse layer and the fine layer is wavy, with a peak near the outer surface and a trough near the inner surface. The distance between the trough and the outer surface is defined as H1, and the distance between the peak and the outer surface is defined as H2. The average value of H2 is not less than 50 μm, and H1 > H2 with an average difference of 2-5 μm.

7. The sintered metal filter membrane according to claim 6, characterized in that, The difference between the maximum distance between the trough and the outer surface and the minimum distance between the crest and the outer surface shall not exceed 8 μm.

8. The sintered metal filter membrane according to claim 6, characterized in that, The average angle A between the line connecting adjacent crests and troughs and the tangent at the interface between the coarse and fine layers at the crest or trough is 2-30°, and F is not greater than 60%.

9. The sintered metal filter membrane according to claim 1, characterized in that, The sintered metal filter membrane has a porosity of 12-40%, the pore area ratio of the inner surface is 10-40%, and the thickness of the coarse layer is 300-1000 μm.

10. A method for preparing a sintered metal filter membrane as described in any one of claims 1-9, characterized in that, Includes the following steps: S1: The coarse particles with a particle size of 1-20μm are compressed for the first time to obtain a hollow tubular first embryo, wherein the pressure of the first compression is P1 and the time is t1; S2: Place the fine-layer particles on the radial outer side of the first green embryo obtained in step S1, and simultaneously perform a second compression with the first green embryo to obtain a hollow tubular second green embryo. The fine-layer particles include fine particles and shrinkage particles. The particle size of the fine particles is 50-300 nm, and the particle size of the shrinkage particles is 0.5-2.5 μm. The mass of the shrinkage particles is 20-50% of the total mass of the fine particles and shrinkage particles. The pressure of the second compression is P2, the time is t2, 0.1≤P2 / P1≤0.4, 1.5≤(t1+t2) / t1≤3; S3: The second green embryo obtained in step S2 is sintered under a pressure of P3. The holding temperature during the sintering process is T0, and the holding time is t0. After the holding is completed, the sintered metal filter membrane is obtained by cooling.

11. The method for preparing a sintered metal filter membrane according to claim 10, characterized in that, The coarse particles are made of nickel or nickel carbonyl, and the fine particles and the shrinkage particles are made of stainless steel; in step S3, P3 does not exceed 0.01 Pa, the heat preservation temperature T0 is 780℃-820℃, and the heat preservation time t0 is 20-40 min.

12. The method for preparing a sintered metal filter membrane according to claim 10, characterized in that, In step S1, P1 is 70-120 MPa and t1 is 60-120 s.

13. The method for preparing a sintered metal filter membrane according to any one of claims 10-12, characterized in that, In step S3, a buffering process is added after the heat preservation is completed and before the cooling process. The buffering process is to continue to maintain the temperature at the buffer temperature T1 for a time of t3, where T0-10≤T1≤T0-5 and t3 is 5-10min.

14. The method for preparing a sintered metal filter membrane according to claim 13, characterized in that, The cooling process includes a first cooling stage and a second cooling stage. In the first cooling stage, the temperature decreases from the buffer temperature T1 to 480-600℃ at a rate of 0.8-1.5℃ / min. In the second cooling stage, the temperature decreases from the final temperature of the first cooling stage to 25-80℃ at a rate of 1.8-4.5℃ / min.

15. A filter element, characterized in that, The filter element comprises a sintered metal filter membrane as described in any one of claims 1-9.

16. A method for filtering supercritical carbon dioxide, characterized in that, During filtration, supercritical carbon dioxide is passed through the sintered metal filter membrane as described in any one of claims 1-9.

Images