Method for analyzing membrane damage of composite semipermeable membrane

Abstract

A method for analyzing a membrane damage of a composite semipermeable membrane for analyzing an influence of the membrane damage generated in a separation function layer of the composite semipermeable membrane in a separation membrane element including: a step of acquiring a dyed image obtained by dyeing the composite semipermeable membrane or a fragment of the composite semipermeable membrane with a dyeing agent; and a step of collating the dyed image with a shape image of a supply side flow channel material, and, on the basis of the collation result, extracting from the dyed image a position where the membrane damage is generated on the separation function layer of the composite semipermeable membrane by the supply side flow channel material provided on the separation function layer side of the composite semipermeable membrane. The influence of the supply side flow channel material on the membrane damage of the separation function layer is calculated.

Description

Method for Analyzing Film Defects of Composite Semipermeable Membrane

[0001] The present invention relates to a method for analyzing defects of a composite semipermeable membrane used for the selective separation of liquid mixtures.

[0002] Regarding the separation of liquid mixtures, there are various techniques for separating a solvent (e.g., water) and a substance dissolved therein (e.g., salts). Among them, the use of membrane separation methods, which have the characteristics of energy saving, space saving, and high separation performance, is expanding. Separation membranes used in membrane separation methods for filtering a liquid mixture and separating it into concentrated water and filtered water include microfiltration membranes, ultrafiltration membranes, nanofiltration membranes, reverse osmosis membranes, etc. Separation membrane elements using these separation membranes are applied to various water treatment applications such as desalination of seawater and brackish water, production of ultrapure water, reuse of wastewater, and recovery of valuable substances.

[0003] Most of the commercially available reverse osmosis membranes and nanofiltration membranes are composite semipermeable membranes, and there are two types: those having a gel layer and a crosslinked separation functional layer of a polymer on a support membrane, and those having a separation functional layer obtained by polycondensing monomers on a support membrane. Among them, a composite semipermeable membrane obtained by coating a support membrane with a separation functional layer made of a crosslinked polyamide obtained by a polycondensation reaction between a polyfunctional amine and a polyfunctional acid halide is widely used as a high-performance separation membrane with excellent solvent permeability and selective separation properties. The composite semipermeable membrane is laminated together with a supply-side channel material such as a plastic net and a permeate-side channel material such as a tricot. This laminate is wound around a cylindrical water collection pipe with a large number of holes, together with a film for enhancing pressure resistance as needed, and is preferably used as a spiral-type separation membrane element.

[0004] Physical damage can occur to the surface of the composite semipermeable membrane during operation of the separation membrane element. For example, prolonged operation at excessively high pressure, operations involving rapid pressure fluctuations, foreign matter mixed into the treated water, scale generated during operation, foulant, etc., can cause physical damage to the membrane surface. If damage to the surface of the composite semipermeable membrane leads to defects in the separation functional layer, it can lead to a decrease in the performance of the separation membrane element. Therefore, if a decrease in performance due to the composite semipermeable membrane is suspected, it is important to inspect the surface of the composite semipermeable membrane for physical damage. Even if this is not the case, checking for physical damage to the composite semipermeable membrane removed from used separation membrane elements during periodic separation membrane element replacement can lead to stable operation thereafter.

[0005] A method for inspecting for physical damage to a composite semipermeable membrane has been disclosed, which involves cross-flowing a staining solution onto the flat surface of the composite semipermeable membrane in question, passing pressurized water through it, and observing the stained area (Patent Document 1). Furthermore, a method for diagnosing abnormalities such as physical degradation in a separation membrane module has been disclosed, which involves determining the abnormality based on the separation performance of at least two types of solutes (Patent Document 2).

[0006] International Publication No. 2015 / 063975, International Publication No. 2023 / 127810

[0007] However, according to the inventors' findings, the method described in Patent Document 1 had room for improvement in that, in the composite semipermeable membrane to be evaluated obtained by disassembling the separation membrane element, not all of the stained areas with the staining solution correspond to damaged areas that affect the performance degradation of the separation membrane element. Furthermore, while the method described in Patent Document 2 is effective for analyzing the separation membrane element as is, it cannot evaluate a composite semipermeable membrane having membrane defects caused by the supply-side flow channel material, which is obtained by disassembling the separation membrane element. The present invention has been made in view of the above, and aims to provide a method for extracting and analyzing physical membrane defects that lead to a performance degradation during the operation of a separation membrane element.

[0008] The present invention aims to solve the above problems, and according to the present invention, the following invention is provided: (1) A method for analyzing membrane defects in a composite semipermeable membrane, comprising a laminate in which a composite semipermeable membrane having a separation functional layer on a support membrane, a supply-side channel material, and a permeable-side channel material are laminated, wherein the method analyzes the effect of membrane defects occurring in the separation functional layer, comprising: step A, obtaining a stained image by staining the composite semipermeable membrane or a fragment of the composite semipermeable membrane with a dye; step B, comparing the stained image with a shape image of the supply-side channel material, and based on the comparison result, extracting from the stained image the location of the membrane defect that occurred in the separation functional layer of the composite semipermeable membrane due to the supply-side channel material provided on the separation functional layer side of the laminate, and calculating the effect of the supply-side channel material on the membrane defect in the separation functional layer. (2) The method for analyzing defects in a composite semipermeable membrane according to (1), wherein in step B, the stained image is converted into stained image data that is binarized into black and white, the shape image of the supply-side channel material is converted into shape image data of the supply-side channel material that is binarized into black and white, and when comparing the binarized stained image data and the binarized shape image data of the supply-side channel material, the total number of pixels that are black in both image data at the position of comparison is taken as X, and either the binarized shape image data of the supply-side channel material or the binarized stained image data is arbitrarily translated and / or rotated with respect to the other image data to be compared to calculate max(X), which is the maximum value of X, and when the total number of pixels defined as black in the binarized stained image data is taken as Y, the influence of the supply-side channel material on defects in the composite semipermeable membrane is estimated using max(X) and Y. (3) The method for analyzing defects in a composite semipermeable membrane according to (2), wherein, before step A, a step is made to determine the removal rate for at least two types of solutes with respect to the composite semipermeable membrane or a fragment of the composite semipermeable membrane by a salt permeability performance measurement method, and in step B, the influence of physical defects other than the influence of the supply-side flow channel material on the removal rate for at least two types of solutes is quantified using the estimated max(X) and Y.(4) The membrane defect analysis method for a composite semipermeable membrane according to (3), wherein the at least two types of solutes are ionic substances with different valencies or substances with different molecular weights. (5) The membrane defect analysis method for a composite semipermeable membrane according to (4), wherein the ionic substances with different valencies are at least a substance composed of monovalent ions and a substance composed of divalent ions. (6) The membrane defect analysis method for a composite semipermeable membrane according to any one of (2) to (5), wherein a calibration curve is prepared in advance relating the removal rate for the at least two types of solutes of a plurality of comparison membranes in which physical scratches are made on the separation functional layer side of an unused composite semipermeable membrane, and the stained area calculated from stained images obtained by staining the plurality of comparison membranes with the staining agent, and the effect of physical scratches other than the supply side flow channel material on the removal rate for the two types of solutes in the composite semipermeable membrane or the fragment of the composite semipermeable membrane to be analyzed is quantified using max(X) and Y of the composite semipermeable membrane or fragment of the composite semipermeable membrane to be analyzed and the calibration curve. (7) A method for analyzing membrane defects of a composite semipermeable membrane according to any one of (2) to (6), wherein, before staining the composite semipermeable membrane or a fragment of the composite semipermeable membrane with the dye, any deposits adhering to the surface of the composite semipermeable membrane or a fragment of the composite semipermeable membrane are removed. (8) A method for analyzing membrane defects of a composite semipermeable membrane according to any one of (2) to (7), wherein, after staining the composite semipermeable membrane with the dye, the stained image is captured from the support membrane side of the stained composite semipermeable membrane. (9) A method for analyzing membrane defects of a composite semipermeable membrane according to any one of (2) to (8), wherein, as a preliminary image processing, either the binarized shape image data of the supply-side flow channel material or the binarized stained image data is rotated arbitrarily with respect to the other image data of the comparison target, so that the regular membrane defects in the stained image are aligned parallel to the horizontal axis or vertical axis of the image. (10) A method for analyzing membrane defects in a composite semipermeable membrane according to (9), wherein the orientation of membrane defects is calculated based on the result of Hough transform of the stained image data, and the rotation angle in the prior image processing is determined in which the binarized stained image data is arbitrarily rotated with respect to the other image data to be compared.

[0009] This invention enables more reliable analysis of physical membrane defects that affect the performance degradation during operation of separation membrane elements.

[0010] This is a diagram of the binarized stained image in Example 1. This is a diagram showing the detected lines after performing a Hough transform on the binarized stained image in Example 1 and detecting lines within the image. This is a diagram of the binarized stained image in Example 1 after rotating it so that the detected lines are vertical. This is a diagram showing the image of the supply-side flow channel material in Example 1 and line segments parallel to the strands constituting the flow channel material. This is a diagram of Example 1 with the stained image and the supply-side flow channel material superimposed so that they overlap as much as possible. This is a graph showing the change in separation performance and the contribution rates of chemical and physical degradation in Example 1. This is a graph showing the results of calculating the contribution rates of chemical and physical degradation after correction using the technology of this disclosure in Example 1. This is a graph showing the results of calculating the contribution rates of chemical and physical degradation after correction using the technology of this disclosure in Example 4. This is a conceptual diagram of how to determine the contribution rates of chemical and physical degradation from the change in separation performance for two types of solutes in a salt permeation performance measurement method.

[0011] The present invention will be described in detail below, but these are merely examples of preferred embodiments, and the present invention is not limited to these. The "~" indicating a numerical range is used to mean that the numbers written before and after it are included as the lower and upper limits.

[0012] The separation membrane element targeted by the membrane defect analysis method of the present invention comprises a laminate comprising a composite semipermeable membrane having a separation functional layer on a support membrane, a supply-side channel material, and a permeable-side channel material.

[0013] The support membrane of the above-mentioned composite semipermeable membrane does not substantially exhibit separation performance of ionic substances, etc., and is intended to provide strength to the separation function layer that is responsible for the separation performance. The material and shape of the above-mentioned support membrane are not particularly limited, but a structure consisting of a substrate and a porous support layer is an example.

[0014] The material and shape of the base material are not particularly limited, but examples include fabrics or nonwovens mainly composed of at least one of polyester, polyamide, or polyolefin. Of these, polyester is preferably used as the base material due to its high mechanical and thermal stability. The thickness of the base material is generally in the range of 10 to 200 μm to ensure dimensional stability.

[0015] The separation functional layer is placed on a support film. The material and shape of the separation functional layer are not particularly limited, but cross-linked polyamide is preferably used from the viewpoint of selective separation and water permeability. The separation functional layer is preferably a thin film in order to improve water permeability. The separation functional layer usually has a thickness of about 0.01 to 1 μm.

[0016] Examples of supply-side flow channel materials include plastic nets containing components such as polyethylene, polypropylene, polyvinyl chloride, polyester, and nylon.

[0017] Examples of permeable channel materials include tricot containing components such as polyester, nylon, and polyurethane.

[0018] In the laminate, the supply-side channel material is provided on the separation functional layer side of the composite semipermeable membrane, and the permeate-side channel material is provided on the support membrane side of the composite semipermeable membrane. The laminate is wound around a cylindrical water collection pipe with numerous holes and is suitably used as a spiral-type separation membrane element. In this case, a film or tape may be used in combination as needed to increase pressure resistance.

[0019] The separation membrane elements targeted by the membrane defect analysis method of the present invention are generally operated by applying high pressure exceeding the osmotic pressure of the raw water using a high-pressure pump, but in such cases, physical damage (defects) may occur on the surface of the separation membrane. Causes of this include, for example, prolonged operation at excessive high pressure, operations involving rapid pressure fluctuations, foreign matter mixed into the water being treated, scale generated during operation, foulants, etc.

[0020] Physical damage to the separation functional layer can lead to a decrease in the performance of the separation membrane element. Conventionally, a method for inspecting for defects in composite semipermeable membranes is known, which involves cross-flowing a staining solution onto the surface of the composite semipermeable membrane to be evaluated, passing pressurized water through it, and observing whether there are any stained areas. However, according to the inventors' findings, the conventional technique had room for improvement in that, when a staining solution is applied to a composite semipermeable membrane obtained by disassembling a separation membrane element, not all of the stained areas correspond to damaged areas that affect the performance of the separation membrane element. This is because the surface of the composite semipermeable membrane is wound around the separation membrane element in contact with the supply-side flow channel material. For example, if the supply-side flow channel material is a plastic net, even if a defect occurs in the composite semipermeable membrane directly below the entanglement point of the plastic net, as long as the separation membrane element is in operation, water will not pass through that defect. As a result, solute and water leakage will not occur, and no performance degradation will occur. Therefore, in the conventional method described above, if the composite semipermeable membrane is physically damaged by being surrounded by the separation membrane element or by being compressed by the net during operation, the composite semipermeable membrane obtained by disassembling the separation membrane element often shows staining directly beneath the supply-side flow channel material adjacent to the separation membrane.

[0021] One embodiment of the present invention includes a step A to obtain a stained image obtained by staining a composite semipermeable membrane or a fragment of a composite semipermeable membrane of a separation membrane element with a dye, and a step B to compare the stained image with a shape image of the supply-side channel material and, based on the comparison result, extract from the stained image the location of membrane defects that occurred in the separation functional layer of the composite semipermeable membrane due to the supply-side channel material provided on the separation functional layer side of the composite semipermeable membrane in the laminate of the separation membrane element, thereby making it possible to calculate the influence of the supply-side channel material on the membrane defects of the separation functional layer. Therefore, stained areas caused by the supply-side channel material can be excluded from the stained image obtained by staining the composite semipermeable membrane to be evaluated, and physical membrane defects that led to a decrease in performance during operation of the separation membrane element can be extracted and analyzed.

[0022] The composite semipermeable membrane of the separation membrane element can be removed by disassembling the separation membrane element. Furthermore, multiple fragments of the composite semipermeable membrane can be cut from the removed composite semipermeable membrane. While the method for removing the composite semipermeable membrane and cutting the fragments is not particularly limited, it is preferable to carry out the process carefully to avoid creating new scratches on the surface of the composite semipermeable membrane, especially by avoiding stress in the shear direction.

[0023] If the surface of the composite semipermeable membrane is contaminated with deposits, it is preferable to remove the deposits from the composite semipermeable membrane or composite semipermeable membrane fragment before staining with a staining agent. The method of removing the deposits is not particularly limited, but it is preferable to remove them by cleaning with chemicals. Preferred chemicals include acidic chemicals such as hydrochloric acid, sulfuric acid, and citric acid; alkaline chemicals such as sodium hydroxide and potassium hydroxide; surfactants such as sodium dodecyl sulfate and sodium polyoxyethylene lauryl sulfate; and chelating agents such as sodium ethylenediaminetetraacetate and sodium hexametaphosphate. Methods for cleaning composite semipermeable membrane fragments with chemicals include immersing the composite semipermeable membrane in an aqueous solution of the chemical or passing an aqueous solution of the chemical through the composite semipermeable membrane, but are not particularly limited. However, it is preferable to use a cleaning method that does not cause a chemical change to the composite semipermeable membrane. When diluting the above chemicals to prepare an aqueous solution, it is preferable to use pure water or ultrapure water, and it is even more preferable to use ultrapure water. The above cleaning with chemicals may be performed before disassembling the separation membrane element.

[0024] In step A above, a stained image is obtained by staining the composite semipermeable membrane or a fragment of the composite semipermeable membrane of the separation membrane element with a dye. Staining may be performed before disassembling the element, or it may be performed using the composite semipermeable membrane or a fragment after disassembly. If performed before disassembly, an aqueous solution containing a predetermined concentration of dye is pressurized and cross-flow filtered onto the separation membrane element to be inspected, and after standing for a predetermined time, it is washed with water or an aqueous solution that does not contain the dye. If performed after disassembly, for example, the object to be inspected may be set in a reverse osmosis membrane evaluation cell and stained by similarly pressurizing an aqueous solution containing the dye (dye solution) and cross-flow filtering. Alternatively, if performed after disassembly, the stain solution may be supplied under atmospheric pressure, and after standing for a predetermined time, it may be washed with water or an aqueous solution that does not contain the dye. Alternatively, it may be washed with water containing a surfactant as a cleaning agent. When staining under atmospheric pressure, in order to uniformly supply the dye solution to the membrane surface, a resin film may be placed on the membrane surface after supplying the dye solution to the membrane surface.

[0025] There are no particular restrictions on the dyes added to the staining solution when staining the membrane to be inspected, but for example, dyes and pigments may be added. There are no particular restrictions on the dyes to be added, but examples include triphenylmethane-based basic dyes such as crystal violet, gentian violet, pyoctanin blue, and methyl violet, which stain purple, as well as rhodamine B, which stains reddish-purple, basic black, which stains black, malachite green, which stains green, methylene blue, which stains blue, and bismarkbron B, which stains reddish-brown. There are also no particular restrictions on pigments, but examples include inorganic pigments and organic pigments. There are no particular restrictions on inorganic pigments, but examples include white pigments containing zinc oxide, lead white, lithopone, titanium dioxide, precipitated barium sulfate and barite powder, red pigments containing red lead and iron oxide red, yellow pigments containing lead yellow and zinc yellow, blue pigments containing ultramarine blue, Prussian blue (ferrocyanide potassium), YInMn blue, and black pigments containing carbon black. Organic pigments are not particularly limited, but examples include polycyclic pigments such as isoindolinone, isoindoline, azomethine, anthraquinone, anthrone, xanthene, diketopyrrolopyrrole, perylene, perinone, quinacridone, indigoid, dioxazine, and phthalocyanine; azo pigments such as monoazo pigments, disazo pigments, and condensed disazo pigments; lake pigments; and fluorescent pigments.

[0026] Furthermore, there are no particular restrictions on the molecular weight of the staining agent added to the staining solution in reverse osmosis membrane staining. However, if the molecular weight is too small, the staining agent will permeate through the pores of the undamaged part of the reverse osmosis membrane, making it impossible to properly inspect for the presence and / or degree of physical damage to the reverse osmosis membrane based on the presence or absence of a stained area on the reverse osmosis membrane permeation side. Therefore, it is preferable that the molecular weight of the staining agent be 300 or more. When the composite semipermeable membrane is a polyamide-based reverse osmosis membrane, methyl violet is particularly preferred from the viewpoint of ease of staining, stability, molecular weight, and availability.

[0027] Examples of solvents used in the dyeing solution include pure water, distilled water, tap water, seawater, rainwater, industrial water, and well water. However, pure water, which does not contain any components other than the dyeing agent, is more preferable in order to avoid causing physical and chemical damage to the film during dyeing.

[0028] The concentration of the staining solution is preferably 1.0 mg / L to 3000 mg / L, more preferably 300 mg / L or more, and even more preferably 1000 mg / L or less, from the viewpoint of the intensity of the color of the stained area.

[0029] When staining water under pressure, the supply pressure of the water to be treated is preferably 0.2 to 8.0 MPa, more preferably 0.4 MPa or higher, and even more preferably 5.5 MPa or lower, from the viewpoint of efficiently separating the water into concentrated water and permeate. It is also preferable to supply the water at a pressure exceeding the osmotic pressure of the water to be treated.

[0030] The flow rate during cross-flow filtration is preferably 0.1 to 10 L / min, more preferably 1 L / min or more, and even more preferably 5 L / min or less, from the viewpoint of efficient staining.

[0031] The contact time between the membrane and the water to be treated is preferably 1 to 180 minutes, more preferably 5 minutes or more, even more preferably 10 minutes or more, even more preferably 120 minutes or less, and even more preferably 60 minutes or less, from the viewpoint of the intensity of the color in the stained area.

[0032] The stained composite semipermeable membrane is imaged electronically and used for subsequent analysis. Since physical scratches allow dyes and pigments to pass through easily, it is preferable that the surface being imaged when capturing stained images is the support membrane side of the composite semipermeable membrane, i.e., the permeable side (the back side of the separation functional layer).

[0033] There are no particular restrictions on the means of image acquisition, but for example, images can be acquired using imaging equipment such as a digital camera, a CIS scanner, or a CCD scanner. There are no particular restrictions on the image quality or storage format when acquiring images. The resolution is preferably 200 dpi or higher, preferably 400 dpi or higher, and more preferably 600 dpi or higher. The color may be color or grayscale. In either case, it is preferable that each channel has 256 gradations or more. There are no particular restrictions on the file storage format, but it is preferable to use a lossless compressed format or an uncompressed format so that no degradation occurs before and after the image processing described later.

[0034] In step B described above, the stained image and the shape image of the supply-side channel material are compared, and based on the comparison result, the location of membrane defects that occurred in the separation functional layer of the composite semipermeable membrane due to the supply-side channel material provided on the separation functional layer side of the composite semipermeable membrane in the laminate is extracted from the stained image. This makes it possible to identify the portion of the membrane defects caused by the supply-side channel material from the stained image.

[0035] The shape image of the supply-side channel material may be created using the imaging method described above to create an image of the actual channel material, or a regular image matching the actual pattern may be artificially drawn. When imaging the actual channel material, it is preferable to image the channel material in a state where it is not distorted or stretched. In addition, the shape image or stained image of the supply-side channel material may be appropriately enlarged or reduced depending on the scale of the comparison target.

[0036] In one embodiment of the present invention, in step B above, the dyed image is converted into dyed image data that is binarized into black and white, the shape image of the supply-side channel material is converted into shape image data of the supply-side channel material that is binarized into black and white, and when the binarized dyed image data and the binarized shape image data of the supply-side channel material are compared, the total number of pixels that are black in both image data is taken as X, and either the binarized shape image data of the supply-side channel material or the binarized dyed image data is arbitrarily translated and / or rotated with respect to the other image data to be compared to calculate the maximum value of X, max(X) is calculated as max(X), and the total number of pixels defined as black in the binarized dyed image data is taken as Y, then the influence of the supply-side channel material on the defects in the composite semipermeable membrane can be estimated using max(X) and Y. Here, estimating the influence of the supply-side channel material on defects in the composite semipermeable membrane using max(X) and Y includes, for example, calculating the ratio obtained by dividing max(X) by Y, and then calculating the influence of the supply-side channel material on defects in the composite semipermeable membrane based on that ratio. Furthermore, pixels defined as black include, for example, pixels whose brightness is below an arbitrary threshold.

[0037] The stained image and the shape image of the supply-side channel material are converted to grayscale images as needed, and then binarized into black and white by setting an arbitrary threshold. In the stained image, the threshold required to distinguish between stained and unstained areas (for example, one point between 0 and 255 levels in the case of 8 bits) is set to enable differentiation between them. Here, binarization into black and white means that pixels with a brightness of above the above threshold are made white, and all others are made black. The threshold is set to an appropriate value according to the color of the film and the characteristics of the imaging equipment so as to minimize excess or deficiency in distinguishing the smallest stained area contained within the inspection area of ​​the stained image, thereby enabling high-precision differentiation between stained and unstained areas contained within the inspection area of ​​the stained image.

[0038] When comparing the binarized image of the stained image with the binarized image of the supply channel material, the two images are superimposed while arbitrarily changing their relative position and rotation angle, and the portion where the stained area and the portion where the supply channel material is present is extracted. When one image is superimposed on the other image to be compared, the total number of pixels where the stained area and the portion where the supply channel material is present is denoted as X. It is preferable to find a superimposition method in which the images are arbitrarily translated and / or rotated so that X takes its maximum value max(X), as this allows for the appropriate extraction of the portion where the two images overlap.

[0039] Calculating the maximum value max(X) by arbitrarily translating and / or rotating an image can be time-consuming if performed by brute force, depending on the number of pixels in the image used. Therefore, as a pre-processing step before arbitrarily rotating either the binarized shape image data of the supply channel material or the binarized stained image data relative to the other image data to be compared, the computational load can be reduced by rotating the stained image so that the regular film scratches are aligned parallel to the horizontal or vertical axis of the image. In other words, by devising a way to make overlapping easier beforehand, the computational load can be reduced and the time required for calculation can be shortened. For this purpose, we focus on the regularity of the pattern of the binarized image. If the supply channel material is a net, the stained image often shows scratches originating from the net's intersection points arranged in a diamond shape. By rotating the image beforehand so that the lines connecting these regular scratches point in a specific direction, the candidate angles for rotation in subsequent calculations can be narrowed down to a few. The specific direction is not particularly limited, but it is preferable for the rotation to be parallel to the vertical or horizontal axis of the image data, as this facilitates analysis.

[0040] The rotation angle can be determined by a human visually inspecting the actual image to find regularity and measure it, or it can be determined by computer-aided image analysis. In cases where there are very few scratches or scratches that do not originate from the flow channel material are arranged in a linear pattern, visual inspection may yield more accurate results.

[0041] Image analysis using a computer is preferable in that a series of analyses can be fully automated. The method of image analysis using a computer is not particularly limited, but it is preferable to calculate the orientation of film defects based on the result of the Hough transform and determine the rotation angle in this process. The Hough transform is mainly an image processing used for detecting straight lines, and a plurality of straight lines can be easily extracted by this Hough transform. An example of a method for obtaining the equation of a straight line passing through defects arranged in a dotted line is illustrated by a feature extraction method using the Hough transform. Further, the image analysis may include preprocessing such as contour extraction.

[0042] When the total number of pixels defined as black by binarization in an image of a composite semipermeable membrane stained with a dye is Y, the value of max(X) / Y can be used as a parameter indicating the ratio representing the influence of the flow path material on the defects of the composite semipermeable membrane.

[0043] In one embodiment of the present invention, before performing the above step A, a step of obtaining the removal rates for at least two types of solutes by a salt permeation performance measurement method may be performed on the composite semipermeable membrane or a fragment of the composite semipermeable membrane. By performing this step and using max(X) and Y obtained in step B after step A, the influence of physical defects other than the supply-side flow path material on the removal rates for the at least two types of solutes can be quantified.

[0044] The deterioration of the composite semipermeable membrane is roughly classified into chemical deterioration caused by contact with an oxidizing agent leaked from the pretreatment step or an acid or alkali used for washing, and physical deterioration. The above salt permeation performance measurement method is a method capable of quantitatively separating the contributions of chemical deterioration and physical deterioration in the performance degradation of the composite semipermeable membrane by obtaining the removal rates (permeation rates) for at least two types of solutes.

[0045] In the case of chemical degradation, when the removal performance of at least two types of solutes decreases, the decrease follows a certain relationship. For example, it is known that there is a nearly proportional relationship between the decrease in NaCl removal performance and boron removal performance. This relationship is similar for other solutes as well, and the changes in the removal performance of two types follow a certain relationship. In other words, the separation performance of test water containing at least two types of solutes in the state before use of the separation membrane can be measured or predicted in advance, and the degradation state of the separation membrane can be determined based on the deviation from that value. Alternatively, multiple test waters containing each solute individually may be used as the test water containing at least two types of solutes, and the removal rate of each solute may be measured individually.

[0046] For the at least two solutes used as indicators, it is preferable to use ionic substances with different valencies, and it is also preferable that they be substances with different molecular weights. Furthermore, when actually applying the method, it is good to target components that show a large difference in removal performance or that are easy to measure. Also, since the surface of the separation membrane is often charged, it is preferable that the at least two solutes are the same type of ionic substance with different valencies. That is, for example, a monovalent cation and a divalent cation, or a monovalent anion and a divalent anion. The monovalent ionic substance is not particularly limited, but it is preferable that it completely dissociates when dissolved in water such as pure water and is neutral. For example, Na and Mg, Cl and SO2 are particularly suitable because they are abundant in nature, easy to handle, and relatively inexpensive. ４ This is highly preferable. Furthermore, since natural water treatment separation membranes for processing natural water such as seawater and river water generally contain naturally occurring organic matter that is weakly anionic, the membrane surface is often given a negative charge, and in such cases, it is preferable to use the cations Na and Mg. In particular, sodium chloride is preferably used as a monovalent ionic substance. The divalent ionic substance used in this invention does not need to be particularly limited, as it just needs to completely dissociate when dissolved in water such as pure water and be neutral, but for similar reasons, magnesium sulfate is preferably used. Selecting these simultaneously is particularly preferable because the cations and anions have different properties.

[0047] When chemical degradation begins to occur in the reverse osmosis membrane, the permeation rate of monovalent ionic substances becomes larger than that of divalent ionic substances first. As the chemical degradation progresses further, the permeation rate of divalent ionic substances also increases, and a phenomenon is observed where the difference from the permeation rate of monovalent ionic substances becomes smaller. Therefore, it is also possible to diagnose the initial state (slight degradation) of chemical degradation due to chemical agent contact from the degree of deterioration of the separation performance between monovalent ionic substances and divalent ionic substances.

[0048] Simply put, for example, when diagnosing using monovalent ionic substances and divalent ionic substances with respect to a reverse osmosis membrane, when the concentration of the monovalent ionic substance is 0.9% by mass or more of the concentration of the monovalent ionic substance in the raw water, and the concentration of the divalent ionic substance in the permeated water in the collecting pipe is 0.2% by mass or less of the concentration of the divalent ionic substance in the raw water, it can be diagnosed that the main cause of the deterioration of the reverse osmosis membrane element is chemical degradation.

[0049] On the other hand, in the case of physical degradation, phenomena such as membrane damage and large holes occur, and the supply water (test water or water to be treated) leaks. That is, the permeated water concentration deteriorates depending on the composition of the supply water regardless of the membrane separation performance, so it can be roughly obtained by calculation.

[0050] After creating a relationship profile in advance between chemical degradation due to chemical agent contact and physical degradation due to damage, etc., as shown in FIG. 9, by comparing with the measured separation performance of the separation membrane and decomposing it into two arrows, the relationship between the change rates of monovalent ionic substances and divalent ionic substances in the case of chemical degradation and the relationship between the change rates of monovalent ionic substances and divalent ionic substances in physical degradation, it becomes possible to determine the contributions of chemical degradation and physical degradation.

[0051] Regarding the above salt permeation performance measurement method, by correcting the solute permeation rate using the values of max(X) and Y and calculating the ratio of physical degradation using that value, it is possible to estimate the ratio of physical degradation excluding the influence of the flow path material. That is, the amount of solute leaking from the membrane defect having the area of Y can be estimated to be less than the amount leaking from the defect corresponding to the area obtained by subtracting max(X) from Y.

[0052] In the above estimation, it is advisable to create a calibration curve in advance regarding the relationship between stained area and removal rate (solute permeability) using multiple composite semipermeable membranes that have been physically scratched beforehand as comparison membranes. Specifically, a calibration curve is created in advance regarding the relationship between the removal rate for at least two types of solutes of multiple comparison membranes, each having physical scratches on the separation functional layer side of an unused composite semipermeable membrane, and the stained area calculated from stained images obtained by staining these comparison membranes with a staining agent. By using this calibration curve and max(X) and Y obtained by analyzing the composite semipermeable membrane under analysis, the removal rate for at least two types of solutes in the composite semipermeable membrane or its fragments can be calculated, thereby quantifying the effect of physical scratches other than those on the supply side flow channel material. The comparison membranes used to create the above calibration curve are preferably composite semipermeable membranes manufactured using the same method as the composite semipermeable membrane used in the analysis, and preferably have not undergone chemical degradation. There are no particular restrictions on how the physical scratches are made, but scratches caused by rubbing or indentation are preferable as they are similar to the scratches on the target of analysis.

[0053] The above estimation can also be performed without using a calibration curve. Assuming that the rate at which the permeability (removal rate) of the two types of solutes increases during physical degradation is constant, the solute permeability coefficient originating from causes other than film defects can be calculated by multiplying the solute permeability due to physical degradation by (Y - max(X)) / Y.

[0054] The present invention will be described in more detail below with reference to examples, but the present invention is not limited in any way by these examples.

[0055] <Reference Example 1> In a water treatment plant using reverse osmosis membrane elements, a significant increase in pressure loss and a tendency toward deterioration of the water quality of the produced water were observed. Therefore, it was decided to remove the reverse osmosis membrane elements from the vessel and inspect them. The goal was to investigate the extent of physical and chemical deterioration of the reverse osmosis membrane itself, but this could not be accurately determined by performance measurements in the element state. Therefore, the reverse osmosis membrane elements were disassembled, and the reverse osmosis membrane and the supply-side flow channel material (net) were carefully removed to avoid causing further damage. The reverse osmosis membrane was cut into circular pieces with a diameter of 75 mm and set in the evaluation cell of the performance evaluation device. A 1500 mg / L sodium chloride aqueous solution was stored in the treated water tank of the performance evaluation device, and supplied for 15 minutes at a supply pressure of 1.5 MPa, a concentrated water flow rate of 3.5 L / min, a water temperature of 25°C, and a pH of 7, separating it into concentrated water and permeate. The permeate was collected, its electrical conductivity was measured, and the concentration was determined from the relationship between sodium chloride and electrical conductivity. After draining the treated water, the treated water was converted to pure water and supplied to the evaluation apparatus. After flushing out the sodium chloride, a 2000 mg / L magnesium sulfate aqueous solution was prepared in the treated water tank and supplied for 15 minutes at a supply pressure of 1.5 MPa, a concentrated water flow rate of 3.5 L / min, a water temperature of 25°C, and a pH of 7. The solution was then separated into concentrated water and permeate. The permeate was collected, its electrical conductivity was measured, and the concentration was determined from the relationship between magnesium sulfate and electrical conductivity.

[0056] As a result, the removal performance of sodium chloride was 99.50% (transmittance 0.50%), and the removal performance of magnesium sulfate was 99.76% (transmittance 0.24%). Prior to this, the removal performance of the reverse osmosis membrane during manufacturing was measured to be 99.86% (transmittance 0.14%), and the removal performance of magnesium sulfate was 99.95% (transmittance 0.05%).

[0057] After draining the treated water, a 500 mg / L methyl violet aqueous solution (staining agent) was stored in the treated water tank, and staining was performed by supplying it at a supply pressure of 0.40 MPa and a concentrated water flow rate of 3.5 L / min for 15 minutes, separating it into concentrated water and permeate. After draining the treated water, pure water was stored in the treated water tank and supplied at a supply pressure of 0.15 MPa, washing until the color of the concentrated water, which was colored by the residual dyeing agent, disappeared.

[0058] The extracted reverse osmosis membrane was dried, and an image was taken from the back side (support membrane side) of the separation functional layer using an image scanner at a resolution of 600 dpi vertically x 600 dpi horizontally, 24-bit full color, and then converted to 8-bit grayscale. At the same time as the reverse osmosis membrane, a grayscale color sample of the Kodak Color Separation Guide & Gray Scale (Q-13) was scanned. Using the image processing software GIMP, level correction was performed so that the parts of the color sample that are white were set to 255 and the parts that are black were set to 0, out of 256 gradations from 0 (black) to 255 (white). Furthermore, binarization was performed with a threshold of 247. The binarized stained image is shown in Figure 1.

[0059] <Example 1> A Hough transform was performed on the image from Reference Example 1 (the stained image of the reverse osmosis membrane shown in Figure 1), and line detection was performed within the image. As a result, line 1, shown in Figure 2, was detected. The entire image was rotated so that line 1 was in the vertical direction, and the image in Figure 3 was obtained. The time required for the preliminary calculations up to this point was less than 5 seconds using a personal computer equipped with an 11th generation Intel Core™ i5-1135G7 processor and 8GB of RAM. The same computer was used for the following steps.

[0060] Next, the dimensions of the supply-side channel material (net) were measured, and an image created at the same scale as the image in Figure 3 is shown on the left side of Figure 4. On the right side of Figure 4, four line segments are shown in the vertical, horizontal, and parallel to the strands that make up the net. Three images obtained by rotating the left side of Figure 4 clockwise at angles θ, 90°, and 180°-θ shown on the right side of Figure 4, along with the original left side of Figure 4, were used as comparison targets. Each of these images was superimposed on the image in Figure 3, and the total number of pixels that were black in both images, X, was calculated by arbitrarily shifting them in parallel. The overlap that maximized X is shown in Figure 5. Of the total 3,240,000 pixels, 37,628 pixels were black, i.e., defects in the composite semipermeable membrane. Of these, the number of pixels that overlapped with the supply-side channel material, Max(X), i.e., the influence of the supply-side channel material, was 21,950 pixels, and the number of pixels that did not overlap was 15,678 pixels. The time required for this series of calculations to find Max(X) was 67 minutes.

[0061] As a relationship between chemical degradation, a relationship between the permeability of sodium chloride and magnesium sulfate was created based on the separation performance (permeability) of a reverse osmosis membrane prepared using a membrane that had been forcibly chemically degraded by immersing it in hypochlorous acid. As a relationship between physical degradation, a relationship between the permeability of sodium chloride and magnesium sulfate was created from the results of measuring the permeability of sodium chloride and magnesium sulfate in reverse osmosis membranes with abrasion scratches and / or indentation scratches of various areas. Furthermore, calibration curves were created for the relationship between the stained area of ​​scratches and the permeability of sodium chloride, and for the relationship between the stained area of ​​scratches and the permeability of magnesium sulfate.

[0062] The transmittance of sodium chloride was 0.50%, and the transmittance of magnesium sulfate was 0.30%. The values ​​measured before degradation (at the time of manufacture) were 0.14% for sodium chloride and 0.05% for magnesium sulfate. When these values ​​and the relationship between physical and chemical degradation were plotted in Figure 6, the contribution ratio of chemical degradation to physical degradation was calculated to be 33:67, indicating that physical degradation contributed more to the decrease in separation performance.

[0063] Next, the transmittances of sodium chloride and magnesium sulfate were corrected using the image analysis results and the calibration curve mentioned above. As a result, the transmittance of sodium chloride was 0.36% and the transmittance of magnesium sulfate was 0.13%. When these were plotted in Figure 7, the contribution ratio of chemical degradation to physical degradation was calculated to be 67:33, and excluding the influence of the supply-side flow channel material, the result indicated that chemical degradation was the main cause.

[0064] <Example 2> In Example 1, the same analysis was performed as in Example 1, except that the entire image was rotated after visually detecting straight lines and measuring the inclination of the image from Reference Example 1 (Figure 1), and the same results were obtained. The time required for the pre-calculation was 2 minutes. The time required for the main calculation to find Max(X) was 67 minutes, the same as in Example 1.

[0065] <Example 3> In Example 1, the same analysis was performed as in Example 1, except that line detection and prior rotation processing were not performed on the image from Reference Example 1 (Figure 1), and the angle with the image from Figure 4 was arbitrarily changed while also performing translation, and the total number of pixels X that are black in both images was calculated by brute force. The same results were obtained. Since the prior calculation was omitted, the time required was 0, but the main calculation to find Max(X) afterwards took 6030 minutes.

[0066] <Example 4> In Example 1, instead of using a calibration curve to correct the transmittance of sodium chloride and magnesium sulfate using the image analysis results, it was assumed that the rate at which the transmittance (removal rate) of the two solutes increases when physical degradation occurs is constant, and the transmittance of the solute due to physical degradation was corrected by multiplying it by (Y - max(X)) / Y. Otherwise, the analysis was performed in the same manner as in Example 1. As a result, the transmittance of sodium chloride was 0.33% and the transmittance of magnesium sulfate was 0.13%. When plotted similarly in Figure 8, the contribution ratio of chemical degradation to physical degradation was calculated to be 59:41, and excluding the influence of the supply side flow channel material, the result showed that chemical degradation was the main cause.

[0067] <Comparative Example 1> The analysis was performed in the same manner as in Example 1, except that image comparison with the supply-side channel material was not performed. The influence of the supply-side channel material on membrane defects in the composite semipermeable membrane could not be determined. As shown in Figure 6, the contribution ratio of chemical degradation to physical degradation was calculated to be 33:67, and the calculation result showed that the contribution of physical degradation was greater in the decrease in separation performance. It was not possible to quantify the influence of physical defects other than those in the supply-side channel material.

[0068] Although various embodiments have been described above with reference to the drawings, it goes without saying that the present invention is not limited to these examples. It is clear to those skilled in the art that various modifications or alterations can be conceived within the scope of the claims, and these will naturally also fall within the technical scope of the present invention. Furthermore, the components of the above embodiments may be combined in any way without departing from the spirit of the invention.

[0069] This application is based on the Japanese Patent Application No. 2024-168172 filed on September 27, 2024, the contents of which are incorporated by reference within this application.

[0070] 1 straight line

Claims

1. A method for analyzing membrane defects in a composite semipermeable membrane, comprising a laminate comprising a composite semipermeable membrane having a separation functional layer on a support membrane, a supply-side channel material, and a permeable-side channel material, wherein the method analyzes the effect of membrane defects occurring in the separation functional layer, comprising: step A acquiring a stained image by staining the composite semipermeable membrane or a fragment of the composite semipermeable membrane with a dye; and step B comparing the stained image with a shape image of the supply-side channel material, and based on the comparison result, extracting from the stained image the location of the membrane defect that occurred in the separation functional layer of the composite semipermeable membrane due to the supply-side channel material provided on the separation functional layer side of the laminate, and calculating the effect of the supply-side channel material on the membrane defect in the separation functional layer.

2. The method for analyzing defects in a composite semipermeable membrane according to claim 1, wherein in step B, the stained image is converted into stained image data that is binarized into black and white, the shape image of the supply-side channel material is converted into shape image data of the supply-side channel material that is binarized into black and white, when comparing the binarized stained image data and the binarized shape image data of the supply-side channel material, the total number of pixels that are black in both image data at the position of comparison is defined as X, either the binarized shape image data of the supply-side channel material or the binarized stained image data is arbitrarily translated and / or rotated with respect to the other image data to be compared, and the maximum value of X, max(X), is calculated, and when the total number of pixels defined as black in the binarized stained image data is defined as Y, the influence of the supply-side channel material on defects in the composite semipermeable membrane is estimated using max(X) and Y.

3. The method for analyzing defects in a composite semipermeable membrane according to claim 2, comprising a step prior to step A, in which the removal rate for at least two types of solutes is determined for the composite semipermeable membrane or a fragment of the composite semipermeable membrane by a salt permeability performance measurement method, and using the max(X) and Y estimated in step B, the influence of physical defects other than the influence of the supply-side flow channel material on the removal rate for at least two types of solutes in the defects in the composite semipermeable membrane.

4. The method for analyzing membrane defects in a composite semipermeable membrane according to claim 3, wherein the at least two solutes are ionic substances with different valencies or substances with different molecular weights.

5. The method for analyzing membrane defects in a composite semipermeable membrane according to claim 4, wherein the ionic substances with different valencies are at least a substance composed of monovalent ions and a substance composed of divalent ions.

6. A method for analyzing membrane defects in a composite semipermeable membrane according to any one of claims 2 to 5, wherein a calibration curve is prepared in advance relating the removal rate for at least two types of solutes of a plurality of comparison membranes having physical scratches on the separation functional layer side of an unused composite semipermeable membrane, and the stained area calculated from stained images obtained by staining the plurality of comparison membranes with the dye, and the effect of physical scratches other than the supply-side flow channel material on the removal rate for at least two types of solutes in the composite semipermeable membrane or the fragment of the composite semipermeable membrane to be analyzed is quantified using the max(X) and Y of the composite semipermeable membrane or the fragment of the composite semipermeable membrane to be analyzed and the calibration curve.

7. A method for analyzing membrane defects of a composite semipermeable membrane according to any one of claims 2 to 6, wherein, before staining the composite semipermeable membrane or a fragment of the composite semipermeable membrane with the dye, any deposits adhering to the surface of the composite semipermeable membrane or a fragment of the composite semipermeable membrane are removed.

8. A method for analyzing membrane defects in a composite semipermeable membrane according to any one of claims 2 to 7, wherein when acquiring a stained image after staining the composite semipermeable membrane with a staining agent, the image is taken from the support membrane side of the stained composite semipermeable membrane.

9. A method for analyzing membrane defects in a composite semipermeable membrane according to any one of claims 2 to 8, wherein, as a preliminary image processing, either the binarized shape image data of the supply-side flow channel material or the binarized stained image data is rotated with respect to the other image data to be compared, so that the regular membrane defects in the stained image are aligned parallel to the horizontal or vertical axis of the image.

10. A method for analyzing membrane defects in a composite semipermeable membrane according to claim 9, comprising: calculating the orientation of membrane defects based on the result of Hough transform of the stained image data; and determining the rotation angle in the prior image processing in which the binarized stained image data is arbitrarily rotated with respect to the other image data to be compared.

Images