Residual glue detection device and method

By using feature spectrum interval comparison technology in semiconductor wafer surface detection, the problem of low efficiency and low accuracy of residual glue detection in translucent structures is solved, and fast and accurate judgment of residual glue position is achieved, and detection efficiency and accuracy are improved. , avoid sample damage.

CN118937356APending Publication Date: 2024-11-12ZHUHAI GREE ELECTRONIC COMPONENTS CO LTD +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410870859.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-01
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

When detecting residual glue on the surface of semiconductor wafers with translucent structures such as silicon carbide, the prior art has problems such as low detection efficiency, low accuracy and sample damage, especially real-time detection is difficult to achieve.

Method used

By pre-acquisition of the characteristic spectral intervals of each model of residual glue, and during the detection process, the spectral data of the surface of the workpiece to be detected is obtained through the spectral camera, the second characteristic spectral interval corresponding to the characteristic spectral interval is extracted for comparison, and the residual glue position is judged by , improve detection accuracy and efficiency.

Benefits of technology

It realizes rapid and accurate detection of residual glue positions, reduces detection time and comparison workload, avoids sample damage, and improves detection efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118937356A_ABST
    Figure CN118937356A_ABST
Patent Text Reader

Abstract

The invention provides a residual glue detection device and method. The method comprises the following steps: determining a first characteristic spectrum interval of residual glue through a test workpiece; the characteristic spectrum interval comprises a characteristic wave band and corresponding spectrum data; acquiring spectral data of the surface of the to-be-detected workpiece; extracting a second characteristic spectrum interval corresponding to the first characteristic spectrum interval in the spectrum data, and comparing the first characteristic spectrum interval with the second characteristic spectrum interval; and if the similarity between the two is greater than a first threshold value, judging that the shooting position is a residual glue area, and if the similarity between the two is less than or equal to the first threshold value, judging that the shooting position is a non-residual glue area. According to the method, the first characteristic spectrum interval obtained in advance is compared with the second characteristic spectrum interval in the spectrum data shot in real time, so that the adhesive residue position on the surface of the to-be-detected workpiece is rapidly obtained, and meanwhile, the detection precision is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of residual glue detection, and in particular to a residual glue detection device and method. Background Art

[0002] Semiconductor power devices have the characteristics of high voltage resistance, high current resistance and low on-resistance. In recent years, the third-generation compound semiconductor materials represented by silicon carbide have rapidly emerged and become an important branch of power semiconductor preparation materials. Unlike silicon wafers, silicon carbide wafer substrates are usually transparent / translucent, which makes it difficult to measure defects on the wafer surface. Photolithography (PHOTO) is the core process in semiconductor manufacturing. The transfer of graphics from the mask (MASK) to the semiconductor surface is completed through steps such as glue coating, exposure, and development on the wafer surface. However, in actual production, there are often problems such as uneven photoresist coating, insufficient exposure or focal plane error in the photolithography process, resulting in local residues of the glue layer on the wafer surface, thus affecting subsequent processes.

[0003] At present, the detection methods of residual adhesive on the surface mainly include: manual visual inspection under an optical microscope with a high-magnification microscope lens, or using AOI detection equipment to take high-definition images of the sample surface through the microscopic light path, and compare them with the standard image to obtain differential data, and then obtain the detection results. This method is widely used and mature and reliable. However, since the new generation of compound semiconductors (such as silicon carbide) are mostly translucent structures, it is difficult to focus on the surface; the refined scanning brought by high-precision detection usually results in low detection efficiency; at the same time, the wafer is easily contaminated during the sample delivery process, and the fixed position of the machine also makes real-time detection difficult to achieve.

[0004] Traditional wafer surface optical inspection often requires the use of automatic optical inspection equipment (AOI), scanning electron microscope (SEM), atomic force microscope (AFM) and other equipment. These devices can scan finely but cannot obtain scanning results in a short time, resulting in the problem of measuring sites taking too long; at the same time, surface morphology measurement by SEM, AFM and other inspection methods will damage the sample surface, thus causing irreversible damage to the product. Summary of the invention

[0005] In order to overcome the problems existing in the related art, the present invention provides a residual glue detection method, which quickly obtains the residual glue position on the surface of the workpiece to be inspected by comparing a first characteristic spectral interval acquired in advance with a second characteristic spectral interval in the spectral data captured in real time, while improving the detection accuracy.

[0006] A method for detecting residual adhesive, comprising:

[0007] Determine the first characteristic spectrum interval of residual adhesive by testing the workpiece;

[0008] Acquire spectral data of the surface of the workpiece to be detected; the surface of the workpiece to be detected includes a residual glue area and a non-residual glue area;

[0009] A second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data is extracted, and the first characteristic spectral interval and the second characteristic spectral interval are compared; if the similarity between the two is greater than a first threshold, the shooting position is judged to be a residual glue area; if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

[0010] Furthermore, the determining the first characteristic spectrum interval of the residual adhesive by testing the workpiece specifically includes: acquiring characteristic spectrum intervals of the residual adhesive of each type to form a residual adhesive data set;

[0011] When inspecting the surface of a workpiece to be inspected, the type of residual adhesive on the surface of the workpiece to be inspected is determined, and a characteristic spectrum interval corresponding to the residual adhesive is retrieved from the residual adhesive data set.

[0012] Furthermore, the forming of the residual adhesive data set specifically includes:

[0013] A hyperspectral camera is used to photograph the surface of the test workpiece to obtain a hyperspectral image; and the photographed area includes both the residual adhesive area and the non-residual adhesive area;

[0014] Acquire the spectral segment data and corresponding spectral data of the hyperspectral image by reading a three-dimensional vector;

[0015] Determine a pixel point in the residual glue area or the non-residual glue area as a reference point, and confirm that each pixel point is a residual glue pixel point or a non-residual glue pixel point by comparing the spectral data of the reference point with other pixel points; all the residual glue pixels constitute the residual glue area;

[0016] According to the spectrum data of the residual glue area and the corresponding spectrum data, a first characteristic spectrum interval corresponding to the residual glue is obtained;

[0017] The first characteristic spectral intervals of different residual adhesive models are obtained in turn, and the characteristic spectral intervals of different residual adhesive models are obtained to form a residual adhesive data set.

[0018] Further, determining a pixel point in the residual glue area as a reference point;

[0019] The spectral data of the reference point and other pixel points are compared by the spectral matching method to confirm whether each pixel point is a residual glue pixel point or a non-residual glue pixel point.

[0020] Furthermore, the obtaining of a first characteristic spectrum interval corresponding to the residual glue according to the spectrum segment data of the residual glue area and the corresponding spectrum data includes:

[0021] The average brightness value of the residual glue area in each band is calculated, and the M peak intervals with the largest average brightness values ​​are selected as the characteristic bands of the residual glue. The characteristic bands and their corresponding spectral data constitute the characteristic spectral intervals of the residual glue.

[0022] Furthermore, spectral data is acquired through a spectral camera, and a filter is arranged between the spectral camera and the workpiece to be inspected, the filter corresponds one-to-one to the residual glue model on the surface of the workpiece to be inspected, and the filter can prevent bands other than the characteristic band of the residual glue from being imaged; the spectral data formed after passing through the filter includes a second characteristic spectral interval corresponding to the residual glue model.

[0023] Furthermore, the characteristic spectral interval includes a characteristic band and corresponding spectral data.

[0024] The present invention also provides a residual adhesive detection device, comprising a spectral camera, a storage component and an analysis component, wherein the storage component stores characteristic spectral intervals of various types of residual adhesive, and the characteristic spectral intervals include characteristic wavebands and corresponding spectral data;

[0025] The spectral camera is used to photograph the surface of the workpiece to be inspected, form spectral data, and transmit the spectral data to the analysis component; the surface of the workpiece to be inspected includes a residual glue area and a non-residual glue area;

[0026] The analysis component retrieves the first characteristic spectral interval corresponding to the residual glue from the storage component, and simultaneously extracts the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data, and compares the first characteristic spectral interval with the second characteristic spectral interval; if the similarity between the two is greater than a first threshold, the shooting position is judged to be a residual glue area, and if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

[0027] Furthermore, it also includes a filter assembly, in which a filter sheet corresponding to the type of residual glue is provided, and the filter sheet can prevent the band other than the characteristic band of the residual glue from being imaged;

[0028] The filter assembly also includes a turntable, in which a plurality of filters are arranged. The turntable is connected to a first driving member, and the first driving member is used to drive the turntable to move so that the selected filter is aligned with the field of view of the spectral camera.

[0029] Furthermore, it also includes a support platform and a second driving member, wherein the support platform is located on a side of the filter assembly away from the spectral camera, and the support platform is connected to the second driving member, and the second driving member is used to drive the position of the workpiece to be detected on the support platform to change.

[0030] The beneficial effects of the present invention are:

[0031] The present invention provides a method for detecting residual glue. Before detecting a workpiece to be detected, it is necessary to first confirm the type of residual glue on the surface of the workpiece to be detected, and determine the first characteristic spectral interval of the residual glue by testing the workpiece, wherein the characteristic spectral interval includes a characteristic band and corresponding spectral data; the characteristic band refers to the band area in which the corresponding spectral data peak is relatively obvious; the present application can pre-store characteristic spectra of various residual glues, and during the detection process, it is only necessary to confirm the type of residual glue and call up the corresponding characteristic spectrum; after entering the detection stage, the spectral data of the surface of the workpiece to be detected is obtained; the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data is extracted, and the first characteristic spectral interval and the second characteristic spectral interval are compared; since the surface of the workpiece to be detected includes a residual glue area and a non-residual glue area; if the similarity after the comparison is greater than the first threshold, the position is judged to be a residual glue area, and if the similarity after the comparison is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area. This application advances the process of obtaining the characteristic spectral range of residual glue to ensure that during the detection stage, only spectral data needs to be obtained and automatically compared to obtain the position of the residual glue; this application concentrates the objects of analysis and comparison in the characteristic spectral range, which can not only improve the detection accuracy, but also reduce the comparison workload and improve the detection efficiency.

[0032] The present application also provides a residual glue detection device, including a spectral camera, a storage component and an analysis component, wherein the storage component stores characteristic spectral intervals of various types of residual glue; the spectral camera is used to photograph the surface of the workpiece to be detected, form spectral data, and transmit the spectral data to the analysis component; the analysis component retrieves the first characteristic spectral interval corresponding to the residual glue from the storage component, and simultaneously extracts the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data, and compares the first characteristic spectral interval and the second characteristic spectral interval; the residual glue position is quickly marked according to the comparison result. Compared with the existing detection device, the present application only needs to use a single spectral camera for a single data acquisition to realize the residual photoresist detection on the surface of the workpiece to be detected, and has the characteristics of simple structure and fast detection speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Schematic diagram of two types of film layers to be tested;

[0034] Figure 2 This is a schematic diagram of the residual adhesive detection principle for this application;

[0035] Figure 3 This is a schematic diagram of the structure of the detection device of this application;

[0036] Figure 4 Schematic diagram of the detection method of this application.

[0037] Reference numerals:

[0038] 11. Spectral camera; 12. Filter; 13. Turntable; 14. Workpiece to be inspected; 15. Support table. DETAILED DESCRIPTION

[0039] The preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0040] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0041] It should be understood that although the terms "first", "second", "third", etc. may be used to describe various information in the present invention, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0042] Example 1

[0043] The present application provides a method for detecting residual adhesive, comprising:

[0044] S1: determining a first characteristic spectrum interval of residual adhesive by testing a workpiece; the characteristic spectrum interval includes a characteristic waveband and corresponding spectrum data;

[0045] S2: Acquiring spectral data of the surface of the workpiece to be inspected; the surface of the workpiece to be inspected includes a residual adhesive area and a non-residual adhesive area;

[0046] S3: Extracting the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data, and comparing the first characteristic spectral interval and the second characteristic spectral interval; if the similarity between the two is greater than a first threshold, the shooting position is judged to be a residual glue area; if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

[0047] Before testing the workpiece to be tested, it is necessary to first confirm the type of residual glue on the surface of the workpiece to be tested, and determine the first characteristic spectral interval of the residual glue by testing the workpiece, wherein the characteristic spectral interval includes a characteristic band and corresponding spectral data; a characteristic band refers to a band region in which the corresponding spectral data peak is relatively obvious; the present application can pre-store the characteristic spectral intervals of various residual glues, and during the testing process, it is only necessary to confirm the type of residual glue and retrieve the corresponding characteristic spectral interval. In other words, the above step S1 can be a step completed in advance, or it can be a step completed synchronously with the surface testing of the workpiece to be tested. In order to improve the detection speed of the surface of the workpiece to be tested, the present application can pre-determine the characteristic spectral intervals of each type of residual glue and store them. In the subsequent testing process, it is only necessary to retrieve the characteristic spectral interval of the corresponding model from the stored database.

[0048] The characteristic spectrum interval includes characteristic bands and corresponding spectral data; that is, under the characteristic bands, the peak of the spectral data of the residual glue is more obvious. This application uses these characteristic bands with obvious peaks and their corresponding spectral data as characteristic signals of residual glue. By comparing these characteristic signals, it can be determined whether there is residual glue in the shooting area. If it is a band outside the characteristic spectrum interval, the peak of the spectral data is not obvious and it is not easy to distinguish it from the surface position of the workpiece to be detected.

[0049] After entering the detection stage, the spectral data of the surface of the workpiece to be detected is obtained; the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data is extracted, and the first characteristic spectral interval and the second characteristic spectral interval are compared; since the surface of the workpiece to be detected includes a residual glue area and a non-residual glue area; if the similarity after the comparison is greater than the first threshold, the position is judged to be a residual glue area, and if the similarity after the comparison is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

[0050] The spectral data obtained by shooting in this application is based on the field of view of the spectral camera. Since the field of view of the spectral camera is relatively small relative to the surface of the workpiece to be inspected, the workpiece to be inspected or the spectral camera can be moved during the specific shooting process to ensure that the spectral data at various positions on the surface of the workpiece to be inspected can be obtained.

[0051] Before photographing the surface of the workpiece to be detected, the type of residual glue on the surface of the workpiece to be detected can be known, so the characteristic spectrum interval used for comparison is also determined and can be directly retrieved from the database. There are two ways to extract the second characteristic spectrum interval of the residual glue on the surface of the workpiece to be detected: the first is to use the same method as that of obtaining the characteristic spectrum interval in step S1. The second is to design a corresponding filter when the characteristic spectrum interval is known in advance, and obtain spectral data through the filter. In this way, the obtained spectral data only contains spectral data of the characteristic band, which can be directly compared with the retrieved first characteristic spectrum interval.

[0052] It should be noted that the position of the residual glue on the surface of the workpiece to be detected is uncertain. Each pixel point can be compared with the characteristic spectral interval as an object. If the similarity between the two is greater than the first threshold, it means that the pixel point matches the characteristic spectral interval of the residual glue, and the probability of it being a residual glue pixel point is high, and it is regarded as a residual glue pixel point. Similarly, if the similarity between the two is less than the first threshold, it means that the pixel point does not match the characteristic spectral interval of the residual glue, and the probability of it being a non-residual glue pixel point is high, and it is regarded as a non-residual glue pixel point. All the residual glue pixels are connected to form a residual glue area.

[0053] In this application, the first threshold value can be set according to the residual glue model and the accuracy requirements of the detection method. When the residual glue and the workpiece surface material are determined, if the accuracy requirements of the detection method are high, the first threshold value can be set to be larger, and if the accuracy requirements of the detection method are lower, the first threshold value can be set to be smaller. At the same time, the size of the first threshold value is also related to the residual glue model or the workpiece surface material. In actual operation, the specific value of the first threshold value can be determined based on empirical values ​​or test data.

[0054] In the present application, the comparison process can be completed by setting an analysis component, which automatically compares the corresponding second characteristic spectral interval in the captured spectral data with the first characteristic spectral interval of the residual glue model acquired in advance, and outputs the comparison result. At the same time, the residual glue position is marked on the surface of the workpiece to be inspected according to the comparison result.

[0055] This application advances the process of obtaining the characteristic spectral range of residual glue to ensure that during the detection stage, only spectral data needs to be obtained and automatically compared to obtain the position of the residual glue; this application concentrates the objects of analysis and comparison in the characteristic spectral range, which can not only improve the detection accuracy, but also reduce the comparison workload and improve the detection efficiency.

[0056] Example 2

[0057] This application is described by taking the workpiece to be inspected as a wafer as an example and the residual glue as a photoresist as an example.

[0058] The spectrum (visible light reflection spectrum) is an inherent property of the object itself, and changes due to the composition, material and external energy of the object. In the semiconductor manufacturing process, the same film layer or photoresist layer on the surface of the wafer is usually a fixed component. After being irradiated by a fixed light source, the film layer and the photoresist / different types of photoresists will also show different spectral characteristics. Taking the common Iline photoresist AZ-MiR900 as an example, its main components are resin, photosensitizer, solvent, organic dye, etc. After being attached to different substrates, the photoresist film layer will also show different colors. In contrast, the components of the substrate are mostly oxide films or semiconductor interfaces, and their colors are mostly light gray (Si) or light green (SiC), etc., which are different from the surface photoresist layer. The differences in color, composition and other characteristics between different film layers are manifested in the spectrum as different peaks and troughs of the curve, or the local intensity differences of the image between different spectral bands, which can be obtained by comparison after instrument acquisition and analysis.

[0059] Please see attached Figure 1 The main factors that cause residual adhesive on the wafer surface during the photolithography process can be divided into the following two types: one is the residual adhesive caused by insufficient development in the photolithography process, and the other is insufficient dry de-stripping in the etching process. In the former, the residual adhesive is usually located on the surface of the film layer, from top to bottom, the surface photoresist and the dielectric layer under the adhesive, while in the latter, the residual adhesive presents a double-layer or multi-layer structure depending on the etching depth.

[0060] This embodiment provides a method for obtaining characteristic spectral intervals of various types of residual glue in advance to form a residual glue data set, so that when the surface of the workpiece to be inspected is inspected, the type of residual glue on the surface of the workpiece to be inspected is determined, and the corresponding characteristic spectral interval of the residual glue is retrieved from the residual glue data set; the detection efficiency can be significantly improved.

[0061] Specifically, the residual adhesive data set is formed including:

[0062] S11: For one specific type of residual adhesive, a hyperspectral camera is used to photograph the surface of the test workpiece to obtain a hyperspectral image; and the photographed area includes both the residual adhesive area and the non-residual adhesive area.

[0063] The test workpiece refers to a carrier used to obtain a specific type of residual adhesive. In order to distinguish it from the workpiece to be tested in the above and below texts, it is named as the test workpiece. It is not used for testing, but only for obtaining the characteristic spectrum interval of a specific type of residual adhesive. The characteristic spectrum interval obtained by the test workpiece is defined as the first characteristic spectrum interval, and the characteristic spectrum interval obtained by the surface of the workpiece to be tested is defined as the second characteristic spectrum interval in the following text. The two are essentially the same, but the sources are different.

[0064] Hyperspectral images are a kind of data set that adds different spectra to two-dimensional image data. They can be regarded as a set of three-dimensional data cubes, as shown in the attached figure. Figure 2As shown. The acquisition of the hyperspectral image is completed by a microscopic hyperspectral camera. The obtained hyperspectral image should include the three-dimensional spectral image data and characteristic point spectrum curve data within the effective area of ​​the test workpiece. Taking the wafer with adhesive after development as an example, a high spatial resolution spectral camera is used to collect the surface hyperspectral image data of the adhesive area and the non-adhesive area of ​​the wafer in a vertical shooting manner. Please refer to the attached Figure 1 . Taking the GaiaMicro series of microscopic hyperspectral systems as an example, the instrument uses slit spectrometry inside, and the slit size is 30μm*14.2mm, which can achieve spectral data acquisition with a magnification of 1X to 8X and a maximum of 1040 channels. During the data acquisition process, the wafer is placed on a support table, and the microscopic hyperspectral camera shoots from top to bottom; when shooting, first adjust the instrument objective lens to the appropriate magnification, and ensure that the spectral resolution is not less than 10nm; adjust the camera position to ensure that the field of view contains both the residual glue to be tested and the film layer to complete a hyperspectral data pre-acquisition.

[0065] S12: Acquire the spectral segment data and corresponding spectral data of the hyperspectral image by reading a three-dimensional vector;

[0066] S113: Determine a pixel point in the residual glue area or the non-residual glue area as a reference point, and confirm that each pixel point is a residual glue pixel point or a non-residual glue pixel point by comparing the spectral data of the reference point with other pixels; all residual glue pixels constitute the residual glue area. In other words, when extracting data, it is necessary to first select a reference point, that is, any spatial point in the residual glue area and the non-residual glue area, and obtain the spectral segment data and corresponding spectral data of the hyperspectral image by reading a three-dimensional vector, so as to obtain a reference spectral vector of the residual glue area and a reference spectral vector of the non-residual glue area.

[0067] Specifically, the reference spectral vectors of the photoresist and the film layer under the glue have been provided above. It is only necessary to use a suitable algorithm to compare the reference spectral vector with the spectral vectors of other pixels in the field of view to determine the location of the residual glue. The commonly used method is the spectral matching method, that is, the classification of objects is achieved by comparing the similarity of two spectral curves. The present application provides a method for determining the matching degree of two vectors by using the cosine value of the angle between the vectors, and the calculation formula is:

[0068]

[0069] Among them, X represents the reference spectral vector, Y represents the spectral vectors of other pixels, and n represents the number of bands in the spectral vector, which is determined by the hyperspectral camera parameters.

[0070] As a specific embodiment, a pixel point in the residual glue area is determined as a reference point; the spectral data of the reference point and other pixel points are compared by the spectral matching method to confirm that each pixel point is a residual glue pixel point or a non-residual glue pixel point. Specifically, assuming that the resolution of the hyperspectral camera is a×b, the reference spectral vector of the residual glue area is taken as vector X, and the spectral curve data corresponding to the positions of other pixel points is vector Y. During the calculation process, each pixel point needs to be scanned and compared, so there are a total of a×b groups of data that need to be matched with vector X. A set of matching results is obtained by calculating the cosine value of the angle. The smaller the calculated result, the closer the trends of the two groups of vectors are, that is, the greater the probability of consistent components. Normally, if there are differences in the composition of the object, there will be a mutation in the spectral matching result. At this time, the pixel point corresponding to the mutation point is the residual glue position.

[0071] The cosine value of the angle calculated in this application can be compared with the first threshold. The first threshold can be set according to the residual glue model and the accuracy requirements of the detection method. When the residual glue and the surface material of the workpiece are determined, if the accuracy requirements of the detection method are high, the first threshold can be set larger. If the accuracy requirements of the detection method are low, the first threshold can be set smaller. At the same time, the size of the first threshold is also related to the model of the residual glue or the surface material of the workpiece. In actual operation, the value of the specific first threshold can be determined based on empirical values ​​or test data. Specifically, for example, the first threshold is set to any value between 0.1-0.3, such as 0.2, and a pixel point in the residual glue area is taken as the reference point. When the cosine value of the angle between the remaining pixels and the reference point is less than or equal to 0.2, it indicates that the pixel point is a residual glue pixel point; when the cosine value of the angle between the remaining pixels and the reference point is greater than 0.2, it indicates that the pixel point is a non-residual glue pixel point.

[0072] As another specific embodiment, a pixel point in the non-residue glue area is determined as a reference point; the spectrum data of the reference point and other pixels are compared by spectral matching method to confirm that each pixel point is a residual glue pixel point or a non-residue glue pixel point. The method is as described above and will not be described in detail here.

[0073] S14: According to the spectrum data of the residual glue area and the corresponding spectrum data, a first characteristic spectrum interval corresponding to the residual glue is obtained. Specifically, the average brightness value of the residual glue area in each band is calculated, and the M peak intervals with the largest average brightness values ​​are selected as the characteristic bands of the residual glue. The characteristic bands and their corresponding spectrum data constitute the characteristic spectrum interval of the residual glue.

[0074] Specifically, in this application, the strength of the spectral feature of an object is measured by the local average brightness of the image. The calculation method is:

[0075] The average brightness of the residual glue area = the sum of the intensity values ​​of all pixels in the area / the number of pixels.

[0076] The average brightness value of the residual glue area in each band is calculated cyclically, and the calculation results between each band are compared. The 1 to 5 significant peak intervals with the largest average brightness calculation results are taken as the spectral characteristic interval of the residual glue.

[0077] S15: Repeat the above steps to obtain characteristic spectrum intervals of different types of residual adhesive, and form a residual adhesive data set.

[0078] Example 3

[0079] like Figure 4 As shown, the present application provides a method for detecting residual adhesive, comprising:

[0080] S1: Determine a first characteristic spectrum interval of residual adhesive by testing a workpiece; the characteristic spectrum interval includes a characteristic waveband and corresponding spectrum data. The method for obtaining the characteristic spectrum interval is as described in Example 2, and will not be described in detail here.

[0081] S2: Acquire spectral data of the surface of the workpiece to be detected; the surface of the workpiece to be detected includes a residual glue area and a non-residual glue area. Specifically, a filter is arranged between the spectral camera and the workpiece to be detected, and the filter corresponds to the residual glue type on the surface of the workpiece to be detected one by one, and the filter can prevent the bands other than the residual glue characteristic band from being imaged. Specifically, the following steps are included:

[0082] S21: Loading the wafer to be tested on the support table. Specifically, a wafer clamp may be used to place the wafer to be tested in a groove inside the support table, with the flat side facing downward.

[0083] S22: The spectral camera is installed above the wafer. After the spectral camera is adjusted to a suitable magnification, the photolithography alignment mark on the wafer surface is searched with the left side of the flat edge as a reference.

[0084] S23: The spectral camera samples the surface in a vertical shooting mode. The field of view is the size of a single shot of lithography. The appropriate magnification is selected to ensure that sufficient image details can be obtained.

[0085] S3: Extract the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data, and compare the first characteristic spectral interval with the second characteristic spectral interval. Among them, the spectral data formed after passing through the filter includes the second characteristic spectral interval corresponding to the residual glue model, that is, the spectral data obtained in this embodiment only includes the spectral data corresponding to the characteristic band, and the characteristic band here refers to the specific band corresponding to the photoresist model on the wafer surface. Before detection, the model of the photoresist can be known according to the wafer processing technology, and the characteristic spectral interval corresponding to the photoresist can be known from Example 2.

[0086] The present application can use an analysis component to automatically compare the captured spectral data with the first characteristic spectral interval and the second characteristic spectral interval; if the similarity between the two is greater than the first threshold, the shooting position is judged to be a residual glue area, and if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area. The spectral camera captures the spectral data of the wafer surface at a fixed frame rate, and the spectral data is captured after passing through a filter, and only contains the band corresponding to the characteristic band of the photoresist. After algorithm analysis, the residual glue position is marked, and the result is output to the computer, which can realize the visualization operation of the analysis result.

[0087] This application realizes the detection of residual adhesive on the surface of the wafer based on the spectral characteristics of the object itself. Compared with conventional detection devices on the market, this method only needs to use a single detector for single data collection to realize the detection of residual photoresist on the surface of the wafer, and has the characteristics of simple structure and fast detection speed; at the same time, the spectrum is the physical property of the object itself, and the reflection spectrum presented by different materials under light must be different. For example, the spectral curves and multi-spectral images of two objects, such as wafers and photoresists, will be different in a certain spectral range. It is only necessary to build a suitable imaging system and cooperate with the data analysis algorithm to dig out the difference interval and complete the detection. This method can avoid the imaging interference caused by traditional cameras due to problems such as optical path and mechanical jitter, and can ensure high-precision detection.

[0088] Example 4

[0089] like Figure 3 As shown, the present application also provides a residual glue detection device, including a spectral camera 11, a storage component and an analysis component, wherein the storage component stores characteristic spectral intervals of various types of residual glue, and the characteristic spectral intervals include characteristic bands and corresponding spectral data;

[0090] The spectral camera 11 is used to photograph the surface of the workpiece to be inspected, form spectral data, and transmit the spectral data to the analysis component; the surface of the workpiece 14 to be inspected includes a residual glue area and a non-residual glue area. In this application, the imaging system consists of a spectral camera 11, a front objective lens, and a locator. The spectral camera has 5 million pixels and a resolution of 2560×1920. The spectral camera 11 collects data in a vertical shooting manner, and can achieve fixed track movement in conjunction with a locator. In this application, the spectral camera can be an industrial camera, and the industrial camera can achieve spectral imaging in combination with a filter chip. The spectral data obtained by the spectral camera is different from the hyperspectral image obtained by the hyperspectral camera, which usually has hundreds of channels. However, the number of data channels obtained after the spectral camera is imaged is relatively small, usually less than 10, so this application defines the data obtained by the spectral camera as spectral data. .

[0091] The analysis component retrieves the first characteristic spectral interval corresponding to the residual glue from the storage component, and simultaneously extracts the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data, and compares the first characteristic spectral interval with the second characteristic spectral interval; if the similarity between the two is greater than a first threshold, the shooting position is judged to be a residual glue area, and if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

[0092] The detection device of the present application also includes a filter component, in which a filter 12 corresponding to the residual glue model is provided, and the filter 12 can prevent the band other than the characteristic band of the residual glue from being imaged; wherein the filter 12 is the core component of the system, which is customized according to the results of the spectral data analysis of the residual glue, and its different specifications represent different characteristic ranges of the photoresist.

[0093] The filter assembly further includes a turntable 13, in which a plurality of filters 12 are arranged, and the turntable 13 is connected to a first driving member, and the first driving member is used to drive the turntable 13 to move so that the selected filter is aligned with the field of view of the spectral camera. The turntable 13 is a carrier of the filter 12, and the channel switching is realized under the drive of the built-in motor of the first driving member, wherein the driving signal is generated by the central processing unit in the analysis assembly.

[0094] The detection device of the present application also includes a support table 15 and a second driving member. The support table 15 is located on the side of the filter component away from the spectral camera, and the support table 15 is connected to the second driving member. The second driving member is used to drive the position of the workpiece 14 to be detected on the support table 15 to change. The support table 15 is a cylindrical structure, and a wafer-specific vacuum suction cup is installed on the surface, which can be compatible with 6-inch, 8-inch, and 12-inch wafers at the same time; a limit device is set at the edge position and a gap is left, which can support manual placement of the wafer and prevent the wafer from slipping due to external interference. The second driving member and the imaging system cooperate with each other to achieve shooting at various positions on the wafer surface.

[0095] The detection device of the present application also includes a light source, which uses an integrating sphere light source with adjustable color temperature and brightness to ensure high uniformity and wide dynamic range of imaging, and is combined with a collimating lens to form a front optical path system.

[0096] When the workpiece to be inspected is a workpiece other than a wafer, the fixing and inspection of the workpiece to be inspected are similar to those of the wafer, and will not be described in detail here.

[0097] Compared with the existing detection devices, the present application only needs to use a single spectral camera to perform a single data acquisition to realize the residual photoresist detection on the surface of the workpiece to be inspected, and has the characteristics of simple structure and fast detection speed.

[0098] Unless otherwise specifically stated, the relative arrangement, numerical expressions and numerical values ​​of the parts and steps set forth in these embodiments do not limit the scope of the present application. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to those of ordinary skill in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be considered as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so that once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings. In the description of the present application, it should be understood that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, perpendicular, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present application and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the device or element referred to must have a specific direction or be constructed and operated in a specific direction, and therefore cannot be understood as limiting the scope of protection of the present application; the directional words "inside and outside" refer to the inside and outside relative to the contours of each component itself.

[0099] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0100] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only for the convenience of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of this application.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting residual glue, characterized in that: include: Determine the first characteristic spectrum interval of residual adhesive by testing the workpiece; Acquire spectral data of the surface of the workpiece to be inspected; A second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data is extracted, and the first characteristic spectral interval and the second characteristic spectral interval are compared; if the similarity between the two is greater than a first threshold, the shooting position is judged to be a residual glue area; if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

2. A residual adhesive detection method according to claim 1, characterized in that: The determining of the first characteristic spectrum interval of the residual adhesive by testing the workpiece specifically includes: obtaining characteristic spectrum intervals of the residual adhesive of each type to form a residual adhesive data set; When inspecting the surface of a workpiece to be inspected, the type of residual adhesive on the surface of the workpiece to be inspected is determined, and a characteristic spectrum interval corresponding to the residual adhesive is retrieved from the residual adhesive data set.

3. A residual adhesive detection method according to claim 2, characterized in that: The forming of the residual adhesive data set specifically includes: A hyperspectral camera is used to photograph the surface of the test workpiece to obtain a hyperspectral image; and the photographed area includes both the residual adhesive area and the non-residual adhesive area; Acquire the spectral segment data and corresponding spectral data of the hyperspectral image by reading a three-dimensional vector; Determine a pixel point in the residual glue area or the non-residual glue area as a reference point, and confirm that each pixel point is a residual glue pixel point or a non-residual glue pixel point by comparing the spectral data of the reference point with other pixel points; all the residual glue pixels constitute the residual glue area; According to the spectrum data of the residual glue area and the corresponding spectrum data, a first characteristic spectrum interval corresponding to the residual glue is obtained; The first characteristic spectral intervals of different residual adhesive models are obtained in turn to form a residual adhesive data set.

4. A residual adhesive detection method according to claim 3, characterized in that: Determine a pixel point in the residual glue area as a reference point; The spectral data of the reference point and other pixel points are compared by the spectral matching method to confirm whether each pixel point is a residual glue pixel point or a non-residual glue pixel point.

5. A method for detecting residual adhesive according to claim 3, characterized in that: The step of obtaining the first characteristic spectrum interval corresponding to the residual glue according to the spectrum segment data of the residual glue area and the corresponding spectrum data includes: The average brightness value of the residual glue area in each band is calculated, and the M peak intervals with the largest average brightness values ​​are selected as the characteristic bands of the residual glue. The characteristic bands and their corresponding spectral data constitute the characteristic spectral intervals of the residual glue.

6. A method for detecting residual adhesive according to claim 1, characterized in that: Spectral data is acquired through a spectral camera, and a filter is arranged between the spectral camera and the workpiece to be inspected, the filter corresponds one-to-one to the residual glue model on the surface of the workpiece to be inspected, and the filter can prevent bands other than the characteristic band of the residual glue from being imaged; the spectral data formed after passing through the filter includes a second characteristic spectral interval corresponding to the residual glue model.

7. A method for detecting residual adhesive according to claim 1, characterized in that: The characteristic spectrum interval includes a characteristic waveband and corresponding spectrum data.

8. A residual glue detection device, characterized in that: It includes a spectral camera, a storage component and an analysis component, wherein the storage component stores characteristic spectral intervals of various types of residual adhesive, and the characteristic spectral intervals include characteristic wavebands and corresponding spectral data; The spectral camera is used to photograph the surface of the workpiece to be inspected, form spectral data, and transmit the spectral data to the analysis component; The analysis component retrieves the first characteristic spectral interval corresponding to the residual glue from the storage component, and simultaneously extracts the second characteristic spectral interval corresponding to the first characteristic spectral interval in the spectral data, and compares the first characteristic spectral interval with the second characteristic spectral interval; if the similarity between the two is greater than a first threshold, the shooting position is judged to be a residual glue area, and if the similarity between the two is less than or equal to the first threshold, the shooting position is judged to be a non-residual glue area.

9. The residual adhesive detection device according to claim 8, characterized in that: It also includes a filter assembly, in which a filter sheet corresponding to the type of residual glue is arranged, and the filter sheet can prevent the band other than the characteristic band of the residual glue from being imaged; The filter assembly also includes a turntable, in which a plurality of filters are arranged. The turntable is connected to a first driving member, and the first driving member is used to drive the turntable to move so that the selected filter is aligned with the field of view of the spectral camera.

10. The residual adhesive detection device according to claim 9, characterized in that: It also includes a support platform and a second driving member, wherein the support platform is located on a side of the filter assembly away from the spectral camera, and the support platform is connected to the second driving member, and the second driving member is used to drive the position of the workpiece to be detected on the support platform to change.

Citation Information

Cited By

  • Back adhesive quality detection method, device and system

    CN120451094A