A defect detection method, a repair device, an FMM production method, and a production line

Abstract

The application discloses a kind of defect detection method, repair equipment, FMM production method and production line, it is related to the field of semiconductor, defect detection method includes: providing metal base material with strip, the main surface of metal base material has the mask pattern formed by photosensitive material;Using first light to irradiate the main surface of metal base material, the image of main surface and the first image of mask pattern are collected;Using second light to irradiate the main surface of metal base material, the second image of mask pattern is collected, and the second light is the light that can make mask pattern produce fluorescence;The image of main surface is identified and analyzed, whether the main surface exists defect is judged;First image and second image are identified and analyzed, and the identification analysis result of twice is compared and summarized, whether mask pattern exists defect is judged.The defect detection method of the application can improve the identification accuracy of defect, and improve the yield of FMM.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and specifically to a defect detection method, equipment, FMM production method, and production line in FMM production. Background Technology

[0002] Fine Metal Mask (FMM) is a core material in the display manufacturing process, primarily used to precisely transfer mask patterns during vapor deposition. FMMs are made from strip-shaped metal substrates through multiple processing steps.

[0003] The manufacturing process of a film-coated metal matrix (FMM) involves pretreatment, thinning, film lamination, exposure, etching, and other steps. FMM demands extremely high processing standards, resulting in a low yield rate during production. To improve the yield rate, defect detection is performed on the metal substrate after development, and defects are repaired. This improves the yield rate. If defects are deemed unrepairable during detection, the metal substrate is scrapped, preventing further processing and reducing production costs and increasing efficiency. However, during defect detection, the precise identification of defects and the mask pattern on the metal substrate can limit the potential improvement in yield rate. Summary of the Invention

[0004] This invention aims to address one of the technical problems in related technologies to a certain extent. To this end, this invention provides a defect detection method that improves the accuracy of defect identification and increases the yield of FMMs (Factory Models).

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A defect detection method, comprising: A strip-shaped metal substrate is provided, wherein a mask pattern formed of a photosensitive material is present on the main surface of the metal substrate; The main surface of the metal substrate is illuminated with a first light, and an image of the main surface and a first image of the mask pattern are acquired. The main surface of the metal substrate is illuminated with a second light to capture a second image of the mask pattern. The second light is a light that can cause the mask pattern to fluoresce. The image of the main surface is identified and analyzed to determine whether there are defects on the main surface. The first and second images are identified and analyzed. The results of the two identification and analysis are compared and summarized to determine whether there are defects in the mask pattern.

[0006] This invention discloses a defect detection method applicable to the FMM (Film-Made Metal Components) production process for detecting defects in metal substrates, thereby improving the yield of FMM finished products. During FMM production, after exposure and development, the photosensitive material on the main surface of the metal substrate forms a mask pattern on the substrate's surface. The portion of the metal substrate not covered by the mask pattern is directly exposed, allowing it to be etched by the etching liquid in subsequent FMM production processes. Defect detection involves identifying the mask pattern to determine its integrity or the presence of excess photosensitive material. Furthermore, the portion of the main surface not covered by the mask pattern is also identified as a defect.

[0007] During defect detection, a first light beam and a second light beam are shone onto the main surface. Under the first light beam, the main surface and the mask pattern produce reflected light. The main surface is made of metal, while the mask pattern is made of a resin-based photosensitive material. The two materials have different reflectivities. Since the surface condition of the defective area differs from that of the normal area, it also exhibits a different reflectivity. By identifying the images of the main surface and the first image, the intensity of the reflected light can be converted into a digital signal, allowing for defect identification. Under the second light beam, the mask pattern fluoresces. The fluorescence intensity at the defective area differs from that at the normal area. By analyzing the second image and converting the fluorescence intensity into a digital signal, defect identification can also be performed.

[0008] The first light source can simultaneously detect defects on both the main surface and the mask pattern, while the second light source detects defects only on the mask pattern. Comparing and summarizing the results of the two defect detections provides a more accurate reflection of the actual condition on the metal substrate. By utilizing the different optical characteristics of the two light sources, complementary verification is achieved, compensating for the shortcomings of a single light source. This improves the defect detection rate and significantly enhances the yield of the final product.

[0009] Optionally, the step of identifying and analyzing the image of the main surface to determine whether there are defects on the main surface includes: Identify the intensity of the reflected light from the main surface to the first ray and generate the corresponding first grayscale value; The first grayscale value is compared with the grayscale value of the standard main surface to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect.

[0010] Optionally, the first image and the second image are subjected to recognition analysis, and the results of the two recognition analyses are compared and summarized to determine whether there are defects in the mask pattern, including: Analyze the first image, identify the intensity of the reflected light from the mask pattern to the first light ray, and generate the corresponding second grayscale value based on the intensity of the reflected light; Analyze the second image, identify the fluorescence intensity generated by the mask pattern, and generate the corresponding third grayscale value based on the fluorescence intensity; The second and third grayscale values ​​are compared and summarized to form the fourth grayscale value; The fourth grayscale value is compared with the grayscale value of the standard mask pattern to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect.

[0011] Optionally, the brightness of the first light is greater than the brightness of the second light, and the brightness / darkness threshold of the first light is lower than the brightness / darkness threshold of the second light.

[0012] The first light beam has a lower brightness / darkness threshold. Because the metal substrate and the mask pattern have different reflectivities for the first light beam, and the metal substrate has a larger surface roughness after thinning, resulting in an overall darker appearance, the difference in reflected light between normal and defective areas is not significant. If the threshold is set too high, slight defective dark areas may be mistaken for normal mask pattern dark areas. Setting a lower brightness / darkness threshold aims to reduce the judgment gap, allowing even subtle changes in grayscale values ​​to be detected and avoiding missed detections. The second light beam has a higher brightness / darkness threshold. This setting filters out weak interference signals, preventing the weakly fluorescent edges of the mask pattern from being identified as defects, ensuring the accuracy of mask pattern defect identification. A higher brightness setting for the first light beam increases the difference in reflected light intensity between normal and defective areas, making defects more noticeable.

[0013] Optionally, the first light source is white light, and the second light source is ultraviolet light. The white light has an intensity of 80 cd to 220 cd, a color temperature of 6000 K to 7000 K, and a brightness / darkness threshold of 15 / 15. The white light source is either coaxial or side-lit. The ultraviolet light has an intensity of 20 cd to 120 cd and a brightness / darkness threshold of 20 / 20. The white light is full-spectrum visible light. By increasing the intensity of the white light, its brightness is made higher than that of the ultraviolet light, enhancing the difference in reflected light between the main surface and the mask pattern. The photosensitive material exhibits fluorescence under ultraviolet light. The mask pattern fluoresces and is thus identified under ultraviolet light irradiation. The main surface does not undergo any special changes under ultraviolet light irradiation, thereby allowing defects in the photosensitive material to be identified through ultraviolet light irradiation.

[0014] Optionally, the defect detection method further includes: Determine whether the defect needs to be repaired.

[0015] Optionally, based on the defect identification results of the main surface and the mask pattern, the defect type, size, and location can be determined; Count the number of different types of defects and compare them with preset values; If the quantity exceeds a preset value, it is determined that it cannot be reworked; if the quantity does not exceed the preset value, it is determined that it needs to be reworked. Repairing defects also incurs costs. After defect detection, a determination is made as to whether repair is necessary. Each defect type is categorized, and the severity of the defect is considered to determine whether rework is required. This process eliminates parts that do not require repair, preventing them from participating in subsequent processing steps, reducing production costs, and improving production efficiency. Simultaneously, repair improves the quality of semi-finished products before etching, significantly increasing the final yield.

[0016] The present invention also provides an FMM production method, comprising: A strip-shaped metal substrate is provided, the metal substrate having a first surface and a second surface disposed opposite to each other; Pretreatment of the metal substrate; The first and second surfaces are thinned. Photosensitive material is attached to the first and second surfaces; The photosensitive materials on the first and second surfaces are exposed to light. The photosensitive materials on the first and second surfaces are subjected to development treatment; The first and second surfaces are etched to form through-holes in the vapor deposition process. The FMM production method employs the aforementioned defect detection method to perform defect detection on the metal substrate after development treatment. The FMM production method also includes a defect repair method for repairing defects, the defect repair method comprising: Based on the defect type of the first surface and its mask pattern, select the corresponding repair method to repair the defect; Based on the defect type of the second surface and its mask pattern, the corresponding repair method is selected to repair the defect.

[0017] The FMM production method provided by this invention adopts the aforementioned defect detection method and has the same beneficial effects as the aforementioned defect detection method. Since the reasoning process is similar, it will not be repeated here.

[0018] Optionally, the defect types include defects such as missing surface material on the metal substrate, missing mask pattern, and redundant mask pattern. The defects such as missing surface material on the metal substrate and missing mask pattern are filled and repaired by injecting photosensitive material, and the redundant mask pattern is removed and repaired by laser vaporization.

[0019] Optionally, the removal and repair include: Based on the specifications of the defect, determine the size of the excess photosensitive material; align the laser head with the defect location and generate laser to vaporize the excess photosensitive material; The filling and repair includes: Calculate the amount of photosensitive material to be injected based on the specifications of the defect; inject the photosensitive material into the center of the defect; if the defect type is a photosensitive material deficiency defect, remove the excess photosensitive material by laser vaporization; if the defect type is a metal substrate surface material deficiency defect, retain the excess photosensitive material; and cure the injected photosensitive material.

[0020] During the photosensitive material removal and repair process, the laser head position is adjusted according to the location of the defect and the size of the photosensitive material. Combined with precise graphic parameters, excess photosensitive material is removed, ensuring the repaired mask pattern matches the normal mask pattern and eliminating the adverse effects of defects on subsequent etching. In photosensitive material filling and repair, photosensitive material is injected to complete the mask pattern, preventing excessive etching of the metal substrate in subsequent processes. Excess photosensitive material is injected during the repair process to ensure complete repair of the defect. The excess photosensitive material is removed by laser vaporization, ensuring the quality of subsequent etching. If the defect is in the metal substrate itself, injection repair prevents over-etching during subsequent etching. The photosensitive material used for repairing the metal substrate does not need to be removed to avoid affecting the repair effect. Furthermore, because the recessed areas of the metal substrate are inherently lacking material, the presence of excess photosensitive material in these areas will not affect the etching result.

[0021] Optionally, when the first surface and its mask pattern are repaired, the first surface is arranged facing upwards, and when the second surface and its mask pattern are repaired, the second surface is arranged facing upwards.

[0022] Optionally, after the first surface and its mask pattern are repaired, an anti-etching film is attached to the second surface, and the first surface is etched to form a first recessed area. An adhesive film is completely attached to the first surface, and the adhesive film fills the first recessed area to adhere to impurities formed on the first surface due to etching. Remove the adhesive film on the first surface and the anti-etching film on the second surface; A photosensitive material is attached to the etched first surface and cured, and the photosensitive material fills the first recessed area. After the second surface and its mask pattern are repaired, the second surface is etched to form a second recessed area. The first recessed area and the second recessed area correspond to each other and are interconnected to form the vapor deposition hole.

[0023] During the etching process of the first surface, the anti-etching film can protect the second surface, preventing the etching liquid from contacting the second surface and etching the unrepaired second surface when it flows. During the etching process of the second surface, the photosensitive material that has been cured on the first surface can protect the first surface. Moreover, the photosensitive material is fluid before curing and can completely adhere to the first surface. Thus, during the etching process of the second recessed area, when the second recessed area is connected to the first recessed area, the cured photosensitive material can block the etching liquid entering the first recessed area, preventing it from flowing to other parts of the first surface on the surface of the photosensitive material.

[0024] After etching, the first surface will have residual etched impurities. By attaching an adhesive film and filling the first recessed area with the adhesive film, the adhesive film can completely adhere to the first surface and stick to the etched impurities. After removing the film from the first surface, the impurities on the first surface will be removed along with the adhesive film. Then, photosensitive material is attached to and cured on the first surface to maintain the cleanliness of the first surface.

[0025] Optionally, the defect repair method is performed before etching the metal substrate; and / or, the first surface and the second surface are etched respectively, with the first surface facing downwards during etching and the second surface facing downwards during etching. Setting the defect repair method to be performed before etching the metal substrate allows defects on the surface of the metal substrate and the mask pattern to be repaired before etching. Thus, during etching, the etching liquid only etches the areas not protected by the mask pattern, preventing over-etching. If the surface to be etched is positioned downwards during etching, the etching liquid is less likely to come into contact with the other surface, thereby avoiding over-etching.

[0026] This invention also provides a defect detection and repair device, including a defect detection apparatus and a defect repair apparatus located downstream of the defect detection apparatus. The defect detection and repair apparatus uses the aforementioned defect detection method to detect defects in the metal substrate used in the production of FMMs, and uses the defect repair method disclosed in one of the aforementioned FMM production methods to repair defects in the metal substrate used in the production of FMMs. The defect detection and repair device provided by this invention employs the aforementioned defect detection and defect repair methods and has the same beneficial effects as the aforementioned defect detection and defect repair methods. Because the reasoning process is similar, it will not be repeated here.

[0027] Optionally, the defect repair device includes a removal module and a filling module. The defect types include defects such as missing surface material on the metal substrate, missing mask pattern, and redundant mask pattern. The defects such as missing surface material on the metal substrate and missing mask pattern are filled and repaired by the filling module, and the redundant mask pattern is removed and repaired by the removal module.

[0028] Optionally, the filling and repair includes a cavity and a nozzle communicating with the cavity. The cavity contains a photosensitive material and a first piezoelectric ceramic on the photosensitive material. The nozzle includes two relatively movable spray seats, forming a photosensitive material channel between the two spray seats. The photosensitive material injection mechanism also includes a second piezoelectric ceramic connected to the spray seats.

[0029] Piezoelectric ceramics deform after becoming conductive. The first piezoelectric ceramic, after deformation, compresses the photosensitive material within its cavity, forcing the material out of the cavity and injecting it into the defect area through a nozzle. The direction and magnitude of the piezoelectric ceramic deformation are related to the direction and magnitude of the voltage. Therefore, after determining the defect size, the required amount of photosensitive material can be calculated, and the voltage input to the first piezoelectric ceramic can be adjusted to control its deformation and precisely control the amount of injected photosensitive material. Furthermore, by adjusting the voltage input to the second piezoelectric ceramic in conjunction with the defect size, the size of the photosensitive material channel can be adjusted, preventing the photosensitive material from being injected outside the defect before it is fully filled, thus affecting the repair of the defect.

[0030] This invention also provides an FMM production line for processing strip-shaped metal substrates into FMMs. The FMM production line includes, from upstream to downstream, a pretreatment device, a thinning device, an exposure device, a developing device, and an etching device. The FMM production line also includes the aforementioned defect detection and repair equipment, which is used to detect and repair defects in the metal substrate during the FMM production process. The FMM production line uses the aforementioned FMM production method to produce FMMs. The FMM production line provided by this invention has the same beneficial effects as the aforementioned defect detection and repair equipment and FMM production method; due to the similar reasoning process, it will not be repeated here.

[0031] These features and advantages of the present invention will be disclosed in detail in the following specific embodiments and accompanying drawings. The preferred embodiments or means of the present invention will be shown in detail in conjunction with the accompanying drawings, but are not intended to limit the technical solutions of the present invention. In addition, each of these features, elements and components appearing in the following text and drawings is a plurality of, and different symbols or numbers are used for convenience of representation, but all represent parts with the same or similar construction or function. Attached Figure Description

[0032] The present invention will be further described below with reference to the accompanying drawings: Figure 1 This is a flowchart of the FMM production method in this invention; Figure 2 This is a flowchart of the defect detection method in this invention; Figure 3 This is a schematic diagram of the defect detection device in this invention; Figure 4 This is a detailed flowchart of step S40 in the present invention; Figure 5 This is a detailed flowchart of the repair of the first surface and the mask pattern on its surface in this invention; Figure 6 This is a detailed flowchart of the second surface and the mask pattern repair on its surface in this invention; Figure 7 This is a schematic diagram of the defect detection device in this invention; Figure 8 This is a schematic diagram of the photosensitive material injection mechanism in this invention; Figure 9 This is a schematic diagram of the photosensitive material injection mechanism in this invention during the injection of photosensitive material; Figure 10 This is a flowchart of steps S50 and S06 in this invention; Figure 11 These are comparison images before and after defect repair; Figure 12 This invention exposes the alignment point coordinate system; Figure 13 This is a flowchart illustrating the repair process in this invention; Figure 14 This invention relates to surface protrusion defects on metal substrates and their impact. Figure 15 This invention relates to surface depressions in metal substrates and their impact.

[0033] Figure label: Metal substrate 100, first surface 110, second surface 120; Feeding device 200, correction sensor 210; Defect detection device 300, first illumination unit 310, second illumination unit 320, identification probe 330; The defect repair device 400, the removal module 410, the photosensitive material injection mechanism 420, the cavity 421, the nozzle 422, the channel 423, the first piezoelectric ceramic 424, the vibrating plate 4241, the spray base 425, the second piezoelectric ceramic 426, the photosensitive material replenishment channel 427, and the curing mechanism 430. Detailed Implementation

[0034] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described are intended to explain the present invention and should not be construed as limiting the invention.

[0035] The terms "an embodiment," "example," or "trademark" used in this specification refer to a particular feature, structure, or characteristic described in connection with the embodiment itself that may be included in at least one embodiment disclosed in this invention. The phrase "in an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0036] Reference Figures 1 to 15 This invention discloses a defect detection and repair method, defect detection and repair equipment, FMM production method, and FMM production line in FMM production. The FMM production line employs an FMM production method to prepare a strip-shaped metal substrate 100 into an FMM finished product. The metal substrate 100 requires multiple processing steps to be prepared into FMM, including pretreatment, thinning, lamination, exposure, development, etching, film removal, cleaning, and cutting. These steps utilize processing equipment such as pretreatment devices, thinning devices, exposure devices, development devices, and etching devices. The defect detection and repair equipment is used during the FMM production process to detect and repair defects in the metal substrate 100 after development treatment, thereby improving the yield rate of FMM. The defect detection and repair equipment uses a defect detection method to detect defects in the metal substrate 100 and a defect repair method to repair defects in the metal substrate 100.

[0037] In the above-mentioned process, the pretreatment stage of the metal substrate 100 involves surface cleaning, leveling, rust removal, and protrusion removal on the two opposing main surfaces of the metal substrate 100—the first surface 110 and the second surface 120—as preparation for production. During the pretreatment stage, the metal substrate 100 is processed by a pretreatment device.

[0038] After pretreatment, the metal substrate 100 undergoes a thinning process to ensure its thickness meets the requirements for FMM fabrication. During this thinning process, material is simultaneously removed from both the first surface 110 and the second surface 120, resulting in double-sided thinning. This not only improves thinning efficiency but also releases internal stress on the two main surfaces, reducing deformation of the metal substrate 100. In the thinning stage, the metal substrate 100 is processed using a thinning device.

[0039] After the thinning is completed, photosensitive material will be attached to the first surface 110 and the second surface 120.

[0040] After the film is applied, the metal substrate 100 enters the exposure stage. The photosensitive materials on the first surface 110 and the second surface 120 are exposed using an exposure device on the FMM production line. Simultaneously, the photomask assembly of the exposure device exposes the photosensitive materials on both surfaces, transferring the mask pattern from the photomask assembly onto the first surface 110 and the second surface 120. After exposure, a portion of the photosensitive material on the metal substrate 100 is cured, and the pattern formed by the cured photosensitive material becomes the mask pattern transferred from the photomask assembly onto the metal substrate 100.

[0041] The developing process of the metal substrate 100 involves cleaning up the uncured photosensitive material, leaving the cured photosensitive material on the metal substrate 100. This forms a mask pattern of photosensitive material on the first surface 110 and the second surface 120, directly exposing any parts of the metal substrate 100 not covered by the mask pattern. During the developing stage, the metal substrate 100 is processed by a developing apparatus.

[0042] In the etching stage of the metal substrate 100, the metal substrate 100 is etched using an etching liquid. The etched areas are the parts of the metal substrate 100 not covered by the mask pattern. After etching, multiple through-holes are formed on the metal substrate 100. The mask pattern protects the covered metal substrate from further etching. During the etching stage, the metal substrate 100 is processed using an etching apparatus. On some FMM production lines, laser drilling equipment is also used to create openings in the metal substrate during the etching stage.

[0043] During the preparation process, the metal substrate 100 is conveyed by the feeding device 200 in a roll-to-roll manner to shorten the production line length and the space occupied. The metal substrate 100 is wound into a roll, and then the feeding device 200 performs unwinding-conveying-winding. The wound metal substrate 100 is transferred to the next processing equipment, and the metal substrate 100 is processed during the conveying process.

[0044] The defect detection and repair equipment includes a defect detection device 300 and a defect repair device 400 located downstream of the defect detection device 300. The defect detection device 300 includes an illumination module, a vision module, and an analysis module. The defect detection device 300 is used to identify the mask pattern on the metal substrate 100 and the surface of the metal substrate 100 to detect defects, which are generally mask pattern excess defects, mask pattern missing defects, and metal substrate surface missing defects. The defect repair device 400 includes a removal module 410 and a filling module. The removal module 410 is used to remove excess photosensitive material, and the filling module is used to fill in the surface pits or missing parts of the mask pattern on the metal substrate 100.

[0045] like Figure 1As shown, the FMM production method includes the following steps: S01, providing a metal substrate 100 and pre-treating the metal substrate 100; S02, the first surface 110 and the second surface 120 are thinned; S03, photosensitive material is attached to the first surface 110 and the second surface 120; S04, the photosensitive materials on the first surface 110 and the second surface 120 are exposed; S05, develop the photosensitive material on the first surface 110 and the second surface 120; S06, the first surface 110 and the second surface 120 are etched to form a through-hole.

[0046] Defect detection and repair are performed after step S05.

[0047] like Figure 2 As shown, the defect detection method and defect repair method include the following steps: S10, a metal substrate 100 is provided, and a mask pattern formed of a photosensitive material is provided on the first surface 110 and the second surface 120 of the metal substrate 100. S20, using the first light to illuminate the first surface 110 and the second surface 120, and acquiring an image of the main surface and a first image of the mask pattern; S30, the first surface 110 and the second surface 120 are illuminated with a second light to acquire a second image of the mask pattern, and the mask pattern fluoresces under the illumination of the second light; S40, Analyze the images acquired in steps S20 and S30 to determine whether there are defects in the main surface and mask pattern; S50. If a defect exists, the corresponding repair method shall be selected according to the defect type to repair the defect.

[0048] The metal substrate in this application is double-sided exposed, and the defect detection method of this application is also applicable to single-sided exposed metal substrates.

[0049] Step S50 includes: Based on the defect type of the first surface 110 and its mask pattern, the corresponding repair method is selected to repair the defect; Based on the defect type of the second surface 120 and its mask pattern, the corresponding repair method is selected to repair the defect.

[0050] During defect repair, the first surface 110 is repaired first, followed by the second surface 120. If the metal substrate 100 does not have defects during the inspection in step S40, etching is performed directly. After the first surface 110 is repaired, a protective film is attached to the second surface 120 before etching the first surface 110. The protective film protects the second surface 120, preventing the etching liquid from contacting and etching the unrepaired second surface 120. After the second surface 120 is repaired, a protective film is also attached to the first surface 110 before etching. The protective film prevents the etching liquid from contacting the already etched first surface 110, thus preventing over-etching. To improve the protection effect, the protective film can be tightly adhered to the metal substrate and protrude from the mask pattern on the metal substrate through methods such as extrusion.

[0051] During the etching process, the metal substrate 100 is transported laterally—when etching the first surface 110, the first surface 110 is set facing upwards, and when etching the second surface 120, the second surface 120 is set facing upwards. This can prevent the etching liquid from flowing randomly on the surface of the metal substrate 100 during the etching process.

[0052] In the above steps, S20 to S50 are performed by the defect detection and repair equipment and the etching device.

[0053] In steps S20 to S40, by identifying the mask patterns on the first surface 110 and the second surface 120 and the mask patterns on the two surfaces, it is determined whether the mask patterns are complete or contain excess photosensitive material. In addition, after development, the metal substrates on the first surface 110 and the second surface 120 that are not covered by the mask patterns are also identified as defects.

[0054] Step S40 includes: S41, perform image recognition and analysis on the main surface to determine whether there are defects on the main surface; S42, perform recognition analysis on the first image and the second image, compare and summarize the results of the two recognition analyses, and determine whether there are defects in the mask pattern.

[0055] In steps S20 and S30, a first light beam and a second light beam are irradiated onto the first surface 110 and the second surface 120. Under the illumination of the first light beam, the main surface and the mask pattern will generate reflected light. The main surface is made of metal, and the mask pattern is made of a photosensitive resin material. The two have different reflectivities. Since the surface condition of the defective area is different from that of the normal area, it will also have a different reflectivity. By identifying the image of the main surface and the first image, the intensity of the reflected light can be converted into a digital signal for defect identification. Under the illumination of the second light beam, the mask pattern can generate fluorescence. The fluorescence intensity generated by the defective area is different from that of the normal area. By analyzing the second image and converting the fluorescence intensity into a digital signal for defect identification, the defect can be identified.

[0056] The first light source can simultaneously detect defects on both the main surface and the mask pattern, while the second light source detects defects only on the mask pattern. Comparing and summarizing the results of the two defect detections provides a more accurate reflection of the actual condition on the metal substrate. By utilizing the different optical characteristics of the two light sources, complementary verification is achieved, compensating for the shortcomings of a single light source. This improves the defect detection rate and significantly enhances the yield of the final product.

[0057] Because the metal substrate 100 has a large surface roughness after thinning, its surface is relatively dark, and the intensity difference of reflected light in different areas is not large enough, resulting in inaccurate defect identification. During defect detection, increasing the brightness of the first light beam expands the contrast between defects and normal areas, making defects more obvious and their boundaries clearer. This allows for better defect repair during the repair process, significantly improving the yield of the final product.

[0058] Repairing defects also incurs costs. In step S40, after detecting defects, the defects are also judged. Based on the detection results and the severity of the defects, a decision is made on whether to rework them. This eliminates parts that do not need to be repaired, preventing them from participating in subsequent processing steps, reducing production costs and improving production efficiency. At the same time, the quality of semi-finished products before etching is improved through repair, resulting in a significant increase in the yield of the final product.

[0059] Reference Figure 2 and Figure 3Based on the above embodiments, in one embodiment of the present invention, the illumination module includes a first illumination unit 310 and a second illumination unit 320 arranged in pairs relative to the metal substrate 100. The two first illumination units 310 respectively illuminate the first surface 110 and the second surface 120 with first light, and the two second illumination units 320 respectively illuminate the first surface 110 and the second surface 120 with second light. The vision module includes a pair of recognition probes 330, which respectively acquire images of the first surface 110 and the second surface 120 and the mask patterns on them. The analysis module includes a chip for analyzing and processing images, data, etc.

[0060] In this system, the first light source is white light, and the second light source is ultraviolet light. White light is full-spectrum visible light; by increasing its intensity, its brightness is made higher than that of ultraviolet light, enhancing the difference in reflected light between the main surface and the mask pattern. The photosensitive material exhibits fluorescence under ultraviolet light; the mask pattern fluoresces and is thus identified when irradiated by ultraviolet light. The main surface does not show any special changes under ultraviolet light irradiation, allowing defects in the photosensitive material to be identified through ultraviolet light illumination. By utilizing the different optical characteristics of the two light sources, complementary verification is achieved, compensating for the limitations of a single light source.

[0061] The luminous intensity of white light is 80cd to 220cd, and the color temperature is 6000K to 7000K. The light source type for white light is a coaxial light source or a side light source.

[0062] In addition, other colored light sources can be added to the first lighting unit 310 according to actual needs.

[0063] The table below shows a comparison of the parameters of white light and ultraviolet light in this application.

[0064]

[0065] A lower brightness / darkness threshold is set for white light. Because the metal substrate and the mask pattern have different reflectivities for white light, and the metal substrate has a larger surface roughness after thinning, resulting in an overall darker appearance, the difference in reflected light between normal and defective areas is not significant. If the threshold is set too high, slight defective dark areas may be mistaken for normal mask pattern dark areas. Setting a lower brightness / darkness threshold aims to reduce the judgment gap, ensuring that even subtle changes in grayscale values ​​are detected and preventing missed detections.

[0066] A high brightness / darkness threshold is set for the ultraviolet light. This setting filters out weak interference signals, preventing the weak fluorescence edges of the mask pattern from being identified as defects, thus ensuring the accuracy of mask pattern defect identification.

[0067] Both white light and ultraviolet light can be used for defect detection independently. White light can perform macroscopic inspection of the surface of the metal substrate 100 and the mask pattern, but its effect on detecting details is poor. Since the metal substrate 100 does not fluoresce when irradiated with ultraviolet light, ultraviolet light irradiation can only be used to detect the mask pattern. In this application, the results of the two detection methods are combined and verified.

[0068] The identification probe 300 includes a first probe and a second probe, which can detect two different types of light, respectively. The first probe is used to acquire images of the main surface and the mask pattern under the first light illumination, while the second probe is used to acquire images of the mask pattern under the second light illumination.

[0069] Step S20 is actually: The first surface 110 and the second surface 120 are illuminated with white light, and images of the first surface 110 and its mask pattern (first image A) and the second surface 120 and its mask pattern (first image B) are acquired respectively.

[0070] Step S30 is actually: When the first surface 110 and the second surface 120 are irradiated with ultraviolet light, the mask patterns on the first surface 110 and the second surface 120 fluoresce under the second light. The second image A of the mask pattern on the first surface 110 and the second image B of the mask pattern on the second surface 120 are acquired respectively.

[0071] Reference Figure 4 Based on the above embodiments, in one embodiment of the present invention, step S40 further includes determining whether there are defects in the main surface and the mask pattern.

[0072] Step S41 includes: S411, Identify the intensity of the reflected light from the main surface to the first light ray, and generate the corresponding first grayscale value; S412, compare the first grayscale value with the grayscale value of the standard main surface to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect.

[0073] Step S42 includes: S421, Analyze the first image, identify the intensity of the reflected light from the mask pattern to the first light ray, and generate the corresponding second grayscale value based on the intensity of the reflected light; S422, Analyze the second image, identify the fluorescence intensity generated by the mask pattern, and generate the corresponding third grayscale value based on the fluorescence intensity; S423 compares and summarizes the second and third grayscale values ​​to form the fourth grayscale value; S424 compares the fourth grayscale value with the grayscale value of the standard mask pattern to determine if they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect.

[0074] Step S40 also includes: S43, summarize the results of steps S412 and S424 to form the fifth grayscale value. The fifth grayscale value reflects the actual situation of the first surface and its mask pattern, and the second surface and its mask pattern on the metal substrate.

[0075] Among them, the first gray level value, the second gray level value, the third gray level value, the fourth gray level value, and the fifth gray level value all include group A and group B, which correspond to the two sides of the metal substrate 100 respectively.

[0076] In the above, steps S41 and S42 are the process of converting the grayscale value into a digital signal.

[0077] Reference Figure 4 Based on the above embodiments, in another embodiment of the present invention, step S40 further includes: determining whether to repair the defect.

[0078] Includes the following steps: S44. Based on the fifth grayscale value obtained in step S43, determine the defect type, size, and location; S45, count the number of different types of defects and compare them with preset values; S46. If the quantity exceeds the preset value, it is determined to be non-repairable and marked as scrap. If the quantity does not exceed the preset value, it is determined to be repairable and marked as requiring repair.

[0079] The cost of repairing defects includes material costs, time costs, and labor costs. The more defects there are, the higher the repair cost. Before repair, the metal substrate 100 that requires high repair costs is scrapped in advance to reduce costs and improve the production efficiency of FMM finished products.

[0080] The preset values ​​are not fixed and need to be determined based on the type of defect. For example, if the defect is a missing mask pattern or a pit on the surface of the metal substrate (100), the preset values ​​may be incorrect. Figure 15 The repair costs for these three types of defects—(1), (2), (3), and (4), (5), and (6), are similar, and the defects will not affect subsequent processing after repair. The total number of each type of defect can be counted to determine whether it is repairable. After production, an FMM (Film Mesh) is cut into multiple pieces, with the preset value generally being half the number of pieces.

[0081] If the defect is a protrusion on the surface of the metal substrate 100, such as Figure 14As shown, the protrusions push up the protective film, preventing it from adhering tightly to the metal substrate 100 and the mask pattern. This causes the etching liquid to flow along the protective film to other areas, resulting in over-etching. Such defects are irreparable and will inevitably have adverse effects during the etching process; therefore, their quantity is disregarded, and the product is directly deemed scrap.

[0082] Since a single FMM (Film Mask) is cut into multiple pieces after production, scrapping an entire FMM due to surface protrusions of the metal substrate 100 would result in significant losses. Therefore, the entire mask pattern is divided into multiple regions (corresponding to the number of pieces cut) based on the subsequent cutting method. Parts with such defects are marked as scrap and can be skipped during repair. For example, if an FMM is subsequently cut into two pieces (the actual number is much larger than two), the mask pattern is divided into two parts during inspection. If one part has surface protrusions of the metal substrate 100, regardless of the number of defects, it is marked as scrap, and only the other part is repaired.

[0083] Protruding defects can be compared Figure 11 , Figure 14 and Figure 15 , Figure 14 As can be seen, the shadow of the raised defect is lighter in the center and gradually darkens around the edges, which is consistent with... Figure 15 The pitting defects on the middle surface create a significant contrast.

[0084] The specific steps of step S40 are as follows: S411, Identify the intensity of the reflected light from the main surface to the first light ray, and generate the corresponding first grayscale value; S412, compare the first grayscale value with the grayscale value of the standard main surface to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect. S421, Analyze the first image, identify the intensity of the reflected light from the mask pattern to the first light ray, and generate the corresponding second grayscale value based on the intensity of the reflected light; S422, Analyze the second image, identify the fluorescence intensity generated by the mask pattern, and generate the corresponding third grayscale value based on the fluorescence intensity; S423 compares and summarizes the second and third grayscale values ​​to form the fourth grayscale value; S424: Compare the fourth grayscale value with the grayscale value of the standard mask pattern to determine if they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect. S43, summarize the results of steps S412 and S424 to form the fifth grayscale value. The fifth grayscale value reflects the actual situation of the first surface and its mask pattern, and the second surface and its mask pattern on the metal substrate.

[0085] S44. Based on the fifth grayscale value obtained in step S43, determine the defect type, size, and location; S45, count the number of different types of defects and compare them with preset values; S46. If the quantity exceeds the preset value, it is determined to be non-repairable and marked as scrap. If the quantity does not exceed the preset value, it is determined to be repairable and marked as requiring repair.

[0086] Reference Figures 5 to 7 Based on the above embodiments, in one embodiment of the present invention, in step S50, defect repair is divided into removal repair and filling repair, which are performed by the removal module 410 and the filling module, respectively.

[0087] The removal module 410 includes a laser head and a position adjustment mechanism for adjusting the position and angle of the laser head. The laser head generates a high-energy laser to vaporize excess photosensitive material on the mask pattern. The position adjustment mechanism adjusts the position of the laser head, regulating the laser movement path and the location of the defect corresponding to the laser head. The laser generates extremely high heat to vaporize and remove the solid photosensitive material. Because defects can be accurately identified, excess photosensitive material can be precisely removed by combining accurate graphic parameters, ensuring that the repaired mask pattern matches the normal mask pattern and eliminating the adverse effects of defects on subsequent etching processes.

[0088] The filling module includes a photosensitive material injection mechanism 420 and a curing mechanism 430. After the photosensitive material is injected by the photosensitive material injection mechanism 420 to repair the defect, the curing mechanism 430 cures the photosensitive material. After the photosensitive material is injected to repair the defect, the filling module uses the curing mechanism 430 to cure the injected photosensitive material, thereby fixing the shape of the photosensitive material and preventing the photosensitive material from flowing and causing the defect to recur.

[0089] Photosensitive material removal and repair includes the following steps: SA11, Based on the analysis results of the defect specifications in step S44, determine the defect location and the size of excess photosensitive material; SA12: The laser head is aligned with the defective area, and the laser vaporizes the excess photosensitive material.

[0090] Photosensitive material filling and repair includes the following steps: SB11, Based on the analysis results of the defect specifications in step S44, determine the location and size of the defect; SB12, calculate the amount of photosensitive material injected, and inject the photosensitive material into the center of the defect area; SB13: If the defect type is a lack of photosensitive material, the excess photosensitive material is removed by laser vaporization; if the defect type is a depression on the surface of the metal substrate 100, the excess photosensitive material is retained. SB14 is used to cure the injected photosensitive material.

[0091] The dimensions of the defect include length, width, area, and depth. In step SB12, the amount of photosensitive material injected is calculated based on the defect type and size, ensuring that the injected amount exceeds the volume. During the repair process, an excess of photosensitive material is injected to ensure complete repair of the defect. In SB14, the photosensitive material is cured using UV lamp irradiation.

[0092] During the repair process, defects such as missing or excess photosensitive material are repaired simultaneously. The removal module 410 and the filling module are located in the same position. Thus, after the photosensitive material is injected, the removal module 410 can be used to remove the excess photosensitive material, followed by curing. Although the injected photosensitive material is liquid, it is relatively viscous and will not deform in a short time, so the repair sequence of injection, removal, and curing can be performed.

[0093] During the repair process, the metal substrate 100 is transported laterally. When repairing the first surface 110 and its mask pattern, the first surface 110 is set upwards. When repairing the second surface 120 and its mask pattern, the second surface 120 is set upwards. This can prevent the liquid photosensitive material from flowing randomly on the surface of the metal substrate 100 during the repair process and causing deformation, which would affect the repair effect.

[0094] If the defect is due to a lack of photosensitive material on the mask pattern, the mask pattern can be made complete by injecting photosensitive material, preventing excessive etching of the metal substrate 100 during subsequent etching processes. Excess photosensitive material is removed by laser vaporization, ensuring the quality of subsequent etching. If the defect is in the metal substrate 100 itself, it can be repaired by injection, preventing over-etching during subsequent etching. The photosensitive material used for repairing the metal substrate 100 does not need to be removed to avoid affecting the repair effect. Furthermore, since the recessed area of ​​the metal substrate 100 itself lacks material, even if there is excess photosensitive material there, it will not affect the etching effect.

[0095] There are two methods for calculating the amount of photosensitive material injected: 1. Photosensitive material deficiency: Photosensitive material injection amount = defect area × photosensitive material thickness × 1.2; 2. If the metal substrate 100 has pits, the amount of photosensitive material injected = defect area × photosensitive material thickness × 2.

[0096] Reference Figure 8 and Figure 9Based on the above embodiments, in one embodiment of the present invention, the photosensitive material injection mechanism 420 includes a cavity 421 and a nozzle 422. The cavity 421 is provided with liquid photosensitive material, and the nozzle 422 includes a channel 423 communicating with the cavity 421. The filling module also includes a first driving member and a second driving member. The first driving member is used to provide a quantitative amount of photosensitive material from the cavity 421 to the nozzle 422, and the second driving member is used to adjust the cross-sectional size of the channel 423.

[0097] Specifically: The filling module also includes a circuit module. The first driving element includes a first piezoelectric ceramic 424 electrically connected to the circuit module. The first piezoelectric ceramic 424 is located inside the cavity 421. The photosensitive material is located in the space enclosed by the first piezoelectric ceramic 424 and the inner wall of the cavity 421. The filling module adjusts the voltage applied to the first piezoelectric ceramic 424 by the circuit module according to the size of the defect.

[0098] The nozzle 422 includes two relatively movable spray seats 425, with a channel 423 formed between the two spray seats 425. A second drive member is connected to the spray seats 425 to drive at least one of the spray seats 425 to move in order to adjust the cross-sectional size of the channel 423.

[0099] The filling module also includes a circuit module. The second driving element includes a second piezoelectric ceramic 426 electrically connected to the circuit module. There are two second piezoelectric ceramics 426, each connected to a spray base 425. The filling module adjusts the magnitude and direction of the voltage applied to the second piezoelectric ceramic 426 by the circuit module according to the size of the defect, so as to drive the two spray bases 425 to move relative to each other.

[0100] Piezoelectric ceramics can deform after becoming conductive, and the direction and magnitude of the deformation are related to the direction and magnitude of the voltage. After deformation, the first piezoelectric ceramic 424 can compress the photosensitive material in the cavity 421, causing the photosensitive material to be squeezed out of the cavity 421 and injected into the defect area through the nozzle 422. A vibrating plate 4241 attached to the first piezoelectric ceramic 424 is also provided in the cavity 421. The vibrating plate 4241 can deform synchronously with the deformation of the first piezoelectric ceramic 424, and the vibrating plate 4241 maintains contact with the photosensitive material.

[0101] After determining the size of the defect, the required amount of photosensitive material can be calculated. The voltage input to the first piezoelectric ceramic 424 can then be adjusted to control its deformation, thus precisely controlling the amount of photosensitive material injected. After deformation, the second piezoelectric ceramic 426 can compress the spray holder 425, causing it to move and adjust the distance between the two spray holders 425. This adjusts the cross-sectional size of the channel 423, ensuring it matches the size of the defect. This prevents photosensitive material from being injected outside the defect before it is fully filled, thus affecting the repair process. The voltage input to the second piezoelectric ceramic 426 is adjusted according to the defect size to regulate the deformation direction and amount of the piezoelectric ceramic, thereby enlarging or shrinking the channel 423.

[0102] The photosensitive material injection mechanism 420 also includes a photosensitive material replenishment channel 427 that connects to the cavity 421. The position of the first piezoelectric ceramic 424 in the cavity 421 remains unchanged. The photosensitive material replenishment channel 427 can replenish the photosensitive material in the cavity 421 in a timely manner after the photosensitive material in the cavity 421 is squeezed out by the first piezoelectric ceramic 424, so that the photosensitive material is kept at a fixed amount and the deformation of the first piezoelectric ceramic 424 matches the amount of photosensitive material injected.

[0103] Step SB12 includes: SB121 adjusts the voltage input to the first piezoelectric ceramic 424 according to the amount of photosensitive material injected, so that the deformation of the first piezoelectric ceramic 424 matches the amount of photosensitive material injected; SB122, according to the defect size, adjusts the voltage input to the second piezoelectric ceramic 426, so that the deformation of the second piezoelectric ceramic 426 adjusts the distance between the two spray seats 425, so that the size of the photosensitive material channel 423 matches the defect size.

[0104] In addition, the number of second piezoelectric ceramics 426 can be set to only one and connected to one of the sprayers 425 to drive it to move closer to or away from the other sprayer 425, which can also adjust the size of the channel 423.

[0105] Reference Figure 10 Based on the above embodiments, in one embodiment of the present invention, step S50 includes the following steps: S51, Based on the analysis results of the defect specifications in step S44, the first surface 110 and the mask pattern on its surface are repaired. S52, based on the analysis results of the defect specifications in step S44, the second surface 120 and its mask pattern are repaired.

[0106] After step S51 is executed and before step S52 is executed, step S06 includes the following steps: S061, an anti-etching film is attached to the second surface 120; S062, the first surface 110 is etched to form a first recessed area; S063, an adhesive film is completely attached to the first surface 110, and the adhesive film fills the first recessed area to adhere to the impurities formed by etching on the first surface 110. S064, Remove the adhesive film on the first surface 110 and the anti-etching film on the second surface 120; S065, a photosensitive material is attached to the first surface 110, the photosensitive material fills the first recessed area, and the photosensitive material is cured.

[0107] The anti-etching film is the protective film mentioned above, which can separate the etching liquid from the second surface 120.

[0108] After performing step S52, step S06 further includes the following steps: S066, the second surface 120 is etched to form a second recessed area, which is connected to the first recessed area.

[0109] After etching, the first surface 110 will retain etching impurities. By attaching an adhesive film and filling the first recessed area, the adhesive film can completely adhere to the first surface 110 and adhere to the etching impurities. After removing the film from the first surface 110, the impurities on the first surface 110 will be removed along with the adhesive film. Then, photosensitive material is attached to and cured on the first surface 110 to maintain its cleanliness. The removal of the film from the first surface 110 and the attachment of the photosensitive material are both performed in a vacuum environment, without the introduction of external impurities. Thus, after the film removal, the first surface 110 is in a dust-free state, and the attachment of the photosensitive material can maintain the dust-free state of the first surface 110 for a long time.

[0110] During the etching process of the second surface 120, the photosensitive material that has been cured on the first surface 110 can protect the first surface 110. Before curing, the photosensitive material is fluid and can completely adhere to the first surface 110. Thus, during the etching process of the second recessed area, when the second recessed area is connected to the first recessed area, the cured photosensitive material can block the etching liquid entering the first recessed area, preventing it from flowing to other parts of the first surface 110 on the surface of the photosensitive material.

[0111] Unlike the above embodiments, in another embodiment of the present invention, it is not necessary to attach a protective film to the first and second surfaces.

[0112] This embodiment discloses two implementation methods: Option 1 involves performing the defect repair method before etching the metal substrate.

[0113] By setting the defect repair method to be performed before the metal substrate etching process, defects on the surface of the metal substrate and the mask pattern can be repaired before etching. In this way, during the etching process, the etching liquid will only etch the areas not protected by the mask pattern, and even if the etching liquid accidentally touches another surface during the etching process, over-etching will not occur.

[0114] Option 2 involves etching the first and second surfaces separately, with the first surface facing down during etching and the second surface facing down during etching. Option 2 streamlines the etching process, preventing the etching liquid from easily contacting the other surface and thus avoiding over-etching.

[0115] Option 1 and Option 2 can be implemented separately or simultaneously.

[0116] Reference Figure 3 , Figure 7 and Figure 12 Based on the above embodiments, in one embodiment of the present invention, during the FMM production process, the metal substrate 100 is held in a stretched state by the tension of the feeding device 200 during the conveying process. Under the action of the tension, the surface flatness of the metal substrate 100 can be maintained, which can better detect defects on its surface, and also facilitate etching and defect repair of the metal substrate 100.

[0117] The feeding device 200 includes a tension sensor and a correction sensor 210 disposed on the conveying path of the metal substrate 100. The tension sensor is used to detect the tension on the metal substrate 100 during the conveying process, so as to adjust the tension applied by the feeding device 200 according to the detection signal of the tension sensor. The correction sensor 210 is used to detect the offset of the metal substrate 100 during the conveying process, so as to adjust the orientation of the metal substrate 100 according to the detection signal of the correction sensor 210.

[0118] Although the metal substrate 100 can maintain a relatively flat state under tensile force, it will also produce a certain amount of deformation, which will cause the mask pattern and defect size identified by the defect detection device 300 to be larger than the actual size.

[0119] Step S40 also includes the following steps: SC11: Obtain the actual coordinates of the photomask alignment during the exposure process of the metal substrate 100 in the production FMM, and compare the obtained actual coordinates with the theoretical alignment coordinates to obtain the expansion and contraction coefficient. SC12 performs expansion / contraction processing on grayscale values ​​based on the expansion / contraction coefficient.

[0120] Step SC11 is executed before step S411; step SC12 is executed after steps S411, S421, and S422 respectively, and is used to perform expansion and contraction processing on the first grayscale value, the second grayscale value, and the third grayscale value.

[0121] By performing expansion and contraction processing on the obtained digital signal after obtaining the expansion and contraction coefficient through steps SC11 and SC12, errors caused by the deformation of the metal substrate 100 can be eliminated.

[0122] like Figure 12 As shown, this is a coordinate system formed by the alignment points of the photomask during the exposure process, where X represents the length direction of the metal substrate and Y represents the width direction of the metal substrate. X-direction expansion / contraction coefficient = (X6 - X1) / preset value; The expansion / contraction coefficient in the Y direction = (Y1-Y6) / preset value.

[0123] The preset value is a fixed value selected based on factors such as the metal substrate and the transmission tension.

[0124] In summary, the defect detection and repair method disclosed in this invention includes the following steps: S10, a metal substrate 100 is provided, and a mask pattern formed of a photosensitive material is provided on the first surface 110 and the second surface 120 of the metal substrate 100. S20, use white light to illuminate the first surface 110 and the second surface 120, and acquire images of the two main surfaces and the first image of the mask pattern of the surface respectively; S30, use ultraviolet light to irradiate the first surface 110 and the second surface 120, the mask patterns on the first surface 110 and the second surface 120 generate fluorescence when irradiated by the second light, and acquire the second images of the mask patterns on the two main surfaces respectively. SC11: Obtain the actual coordinates of the photomask alignment during the exposure process of the metal substrate 100 in the production FMM, and compare the obtained actual coordinates with the theoretical alignment coordinates to obtain the expansion and contraction coefficient. S411, Identify the intensity of the reflected light from the main surface to the first light ray, and generate the corresponding first grayscale value; SC12 performs expansion / contraction processing on the first grayscale value based on the expansion / contraction coefficient; S412, compare the first grayscale value with the grayscale value of the standard main surface to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect. S421, Analyze the first image, identify the intensity of the reflected light from the mask pattern to the first light ray, and generate the corresponding second grayscale value based on the intensity of the reflected light; SC12 performs expansion / contraction processing on the second grayscale value based on the expansion / contraction coefficient; S422, Analyze the second image, identify the fluorescence intensity generated by the mask pattern, and generate the corresponding third grayscale value based on the fluorescence intensity; SC12 performs expansion / contraction processing on the third grayscale value based on the expansion / contraction coefficient; S423 compares and summarizes the second and third grayscale values ​​to form the fourth grayscale value; S424: Compare the fourth grayscale value with the grayscale value of the standard mask pattern to determine if they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect. S43, summarize the results of steps S412 and S424 to form the fifth grayscale value. The fifth grayscale value reflects the actual situation of the first surface and its mask pattern, and the second surface and its mask pattern on the metal substrate.

[0125] S44. Based on the fifth grayscale value obtained in step S43, determine the defect type, size, and location; S45, count the number of different types of defects and compare them with preset values; S46. If the quantity exceeds the preset value, it is determined to be non-repairable and marked as scrap. If the quantity does not exceed the preset value, it is determined to be repairable and marked as requiring repair. S51, Based on the analysis results of the defect specifications in step S44, the first surface 110 and the mask pattern on its surface are repaired. S52, based on the analysis results of the defect specifications in step S44, the second surface 120 and its mask pattern are repaired.

[0126] The FMM production method disclosed in this invention includes the following steps: S01, providing a metal substrate 100 and pre-treating the metal substrate 100; S02, the first surface 110 and the second surface 120 are thinned; S03, photosensitive material is attached to the first surface 110 and the second surface 120; S04, the photosensitive materials on the first surface 110 and the second surface 120 are exposed; S05, develop the photosensitive material on the first surface 110 and the second surface 120; S20, use white light to illuminate the first surface 110 and the second surface 120, and acquire images of the two main surfaces and the first image of the mask pattern of the surface respectively; S30, use ultraviolet light to irradiate the first surface 110 and the second surface 120, the mask patterns on the first surface 110 and the second surface 120 generate fluorescence when irradiated by the second light, and acquire the second images of the mask patterns on the two main surfaces respectively. SC11: Obtain the actual coordinates of the photomask alignment during the exposure process of the metal substrate 100 in the production FMM, and compare the obtained actual coordinates with the theoretical alignment coordinates to obtain the expansion and contraction coefficient. S411, Identify the intensity of the reflected light from the main surface to the first light ray, and generate the corresponding first grayscale value; SC12 performs expansion / contraction processing on the first grayscale value based on the expansion / contraction coefficient; S412, compare the first grayscale value with the grayscale value of the standard main surface to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect. S421, Analyze the first image, identify the intensity of the reflected light from the mask pattern to the first light ray, and generate the corresponding second grayscale value based on the intensity of the reflected light; SC12 performs expansion / contraction processing on the second grayscale value based on the expansion / contraction coefficient; S422, Analyze the second image, identify the fluorescence intensity generated by the mask pattern, and generate the corresponding third grayscale value based on the fluorescence intensity; SC12 performs expansion / contraction processing on the third grayscale value based on the expansion / contraction coefficient; S423 compares and summarizes the second and third grayscale values ​​to form the fourth grayscale value; S424: Compare the fourth grayscale value with the grayscale value of the standard mask pattern to determine if they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect. S43, summarize the results of steps S412 and S424 to form the fifth grayscale value. The fifth grayscale value reflects the actual situation of the first surface and its mask pattern, and the second surface and its mask pattern on the metal substrate.

[0127] S44. Based on the fifth grayscale value obtained in step S43, determine the defect type, size, and location; S45, count the number of different types of defects and compare them with preset values; S46. If the quantity exceeds the preset value, it is determined to be non-repairable and marked as scrap. If the quantity does not exceed the preset value, it is determined to be repairable and marked as requiring repair. S51, Based on the analysis results of the defect specifications in step S44, the first surface 110 and the mask pattern on its surface are repaired. S061, an anti-etching film is attached to the second surface 120; S062, the first surface 110 is etched to form a first recessed area; S063, an adhesive film is completely attached to the first surface 110, and the adhesive film fills the first recessed area to adhere to the impurities formed by etching on the first surface 110. S064, Remove the adhesive film on the first surface 110 and the anti-etching film on the second surface 120; S065, a photosensitive material is attached to the first surface 110, the photosensitive material fills the first recessed area, and the photosensitive material is cured. S52, Based on the analysis results of the defect specifications in step S44, the second surface 120 and the mask pattern on its surface are repaired. S066, the second surface 120 is etched to form a second recessed area, which is connected to the first recessed area.

[0128] After the above steps, the yield rate of FMM increased significantly (defect rate ≤15%), greatly reducing batch scrap. Ten batches of products were produced using the improved equipment and process. The defect statistics for each batch can be found in the chart.

[0129] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes, but is not limited to, the contents described in the accompanying drawings and the specific embodiments above. Any modifications that do not depart from the functional and structural principles of the present invention will be included within the scope of the claims.

Claims

1. A defect detection method, characterized in that, include: A strip-shaped metal substrate (100) is provided, the main surface of which has a mask pattern formed of a photosensitive material; The main surface of the metal substrate (100) is illuminated with a first light, and an image of the main surface and a first image of the mask pattern are acquired. The main surface of the metal substrate (100) is illuminated with a second light to capture a second image of the mask pattern. The second light is a light that can cause the mask pattern to fluoresce. The image of the main surface is identified and analyzed to determine whether there are defects on the main surface. The first and second images are identified and analyzed. The results of the two identification and analysis are compared and summarized to determine whether there are defects in the mask pattern. Wherein, the brightness of the first light is greater than that of the second light, the brightness / darkness threshold of the first light is lower than that of the second light, the first light is white light, the second light is ultraviolet light, the light intensity of the white light is 80cd to 220cd, the color temperature is 6000K to 7000K, and the brightness / darkness threshold is 15 / 15, the light source type of the white light is a coaxial light source or a side light, and the light intensity of the ultraviolet light is 20cd to 120cd, and the brightness / darkness threshold is 20 / 20.

2. The defect detection method according to claim 1, characterized in that, The step of identifying and analyzing the image of the main surface to determine whether there are defects on the main surface includes: Identify the intensity of the reflected light from the main surface to the first ray and generate the corresponding first grayscale value; The first grayscale value is compared with the grayscale value of the standard main surface to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect.

3. The defect detection method according to claim 1, characterized in that, The first and second images are analyzed for recognition. The results of the two analyses are compared and summarized to determine whether there are defects in the mask pattern, including: Analyze the first image, identify the intensity of the reflected light from the mask pattern to the first light ray, and generate the corresponding second grayscale value based on the intensity of the reflected light; Analyze the second image, identify the fluorescence intensity generated by the mask pattern, and generate the corresponding third grayscale value based on the fluorescence intensity; The second and third grayscale values ​​are compared and summarized to form the fourth grayscale value; The fourth grayscale value is compared with the grayscale value of the standard mask pattern to determine whether they are consistent. If they are inconsistent, there is a defect; if they are consistent, there is no defect.

4. The defect detection method according to any one of claims 1 to 3, characterized in that, The defect detection method further includes: Determine whether the defect needs to be repaired.

5. The defect detection method according to claim 4, characterized in that, Based on the defect identification results of the main surface and mask pattern, determine the defect type, size, and location; Count the number of different types of defects and compare them with preset values; If the quantity exceeds the preset value, it is determined that it cannot be repaired; if the quantity does not exceed the preset value, it is determined that it needs to be repaired.

6. An FMM production method, characterized in that, include: A strip-shaped metal substrate (100) is provided, the metal substrate (100) having a first surface (110) and a second surface (120) disposed opposite to each other. The metal substrate (100) is pretreated; The first surface (110) and the second surface (120) are thinned. Photosensitive material is attached to the first surface (110) and the second surface (120); The photosensitive materials on the first surface (110) and the second surface (120) are exposed. The photosensitive materials on the first surface (110) and the second surface (120) are developed; The first surface (110) and the second surface (120) are etched to form a through-hole; The FMM production method employs the defect detection method according to any one of claims 1 to 5 to perform defect detection on the metal substrate (100) after development treatment. The FMM production method further includes a defect repair method for repairing defects, the defect repair method comprising: Based on the defect type of the first surface (110) and its mask pattern, the corresponding repair method is selected to repair the defect; Based on the defect type of the second surface (120) and its mask pattern, the corresponding repair method is selected to repair the defect.

7. The FMM production method according to claim 6, characterized in that, The defect types include surface material defects of the metal substrate (100), mask pattern defects, and mask pattern excess defects. The surface material defects of the metal substrate (100) and the mask pattern defects are filled and repaired by injecting photosensitive material, and the mask pattern excess defects are removed and repaired by laser vaporization.

8. The FMM production method according to claim 7, characterized in that, The removal and repair include: Determine the size of excess photosensitive material based on the specifications of the defect; The laser head is aligned with the defective area, and the laser vaporizes the excess photosensitive material. The filling and repair includes: Calculate the amount of photosensitive material to be injected based on the specifications of the defect; Inject photosensitive material into the center of the defective area; If the defect type is a photosensitive material missing defect, the excess photosensitive material is removed by laser vaporization; If the defect type is a missing surface material defect of the metal substrate (100), then the excess photosensitive material is retained; The injected photosensitive material is then cured.

9. The FMM production method according to any one of claims 6 to 8, characterized in that, When the first surface (110) and its mask pattern are repaired, the first surface (110) is arranged facing upwards, and when the second surface (120) and its mask pattern are repaired, the second surface (120) is arranged facing upwards.

10. The FMM production method according to claim 9, characterized in that, The FMM production method further includes: After the first surface (110) and its mask pattern are repaired, an anti-etching film is attached to the second surface (120), and the first surface (110) is etched to form a first recessed area. An adhesive film is completely attached to the first surface (110), and the adhesive film fills the first recessed area to adhere to the impurities formed by etching on the first surface (110). Remove the adhesive film on the first surface (110) and the anti-etching film on the second surface (120); Photosensitive material is attached to the first surface (110) after etching and the photosensitive material is cured, and the photosensitive material fills the first recessed area; After the second surface (120) and its mask pattern are repaired, the second surface (120) is etched to form a second recessed area. The first recessed area and the second recessed area correspond to each other and are interconnected to form the vapor deposition hole.

11. The FMM production method according to any one of claims 6 to 8, characterized in that, The defect repair method is performed before etching the metal substrate (100); And / or, the first surface (110) and the second surface (120) are respectively etched, the first surface (110) is arranged facing down when the first surface (110) is etched, and the second surface (120) is arranged facing down when the second surface (120) is etched.

12. A defect detection and repair device, characterized in that, The device includes a defect detection device (300) and a defect repair device (400) located downstream of the defect detection device (300). The defect detection and repair device performs defect detection on the metal substrate (100) in the production of FMM using the defect detection method according to any one of claims 1 to 5. The defect detection and repair device performs defect repair on the metal substrate (100) in the production of FMM using the defect repair method disclosed in an FMM production method according to any one of claims 6 to 11.

13. The defect detection and repair equipment according to claim 12, characterized in that, The defect repair device (400) includes a removal module (410) and a filling module. The defect types include surface material loss defects of the metal substrate (100), mask pattern loss defects, and mask pattern excess defects. The surface material loss defects of the metal substrate (100) and the mask pattern loss defects are filled and repaired by the filling module, and the mask pattern excess defects are removed and repaired by the removal module (410).

14. The defect detection and repair equipment according to claim 13, characterized in that, The filling module includes a cavity (421) and a nozzle (422) communicating with the cavity (421). The cavity (421) contains a photosensitive material and a first piezoelectric ceramic (424) on the photosensitive material. The nozzle (422) includes two relatively movable spray seats (425), and a photosensitive material channel (423) is formed between the two spray seats (425). The filling module also includes a second piezoelectric ceramic (426) connected to the spray seats (425).

15. An FMM production line for processing strip-shaped metal substrates (100) into FMMs, characterized in that, The FMM production line includes a pretreatment device, a thinning device, an exposure device, a developing device, and an etching device from upstream to downstream. The FMM production line also includes the defect detection and repair equipment as described in claim 12. The defect detection and repair equipment is used to perform defect detection and repair on the metal substrate (100) during the FMM production process. The FMM production line uses the FMM production method described in claim 6 to produce FMM.

Images