Wafer processing detection method and system
By detecting the three-dimensional edge overlap of the wafer and generating bond offset reminder information, the problem of inaccurate detection of three-dimensional bonding and analyzing bond offset in the prior art is solved, and the accurate evaluation of wafer processing quality is achieved and improve.
Patent Information
- Application Number
- CN202410955191.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-17
AI Technical Summary
The prior art cannot accurately detect three-dimensional bonding and analyze bonding offsets during multi-layer wafer connections, resulting in the inability to accurately evaluate bonding quality during wafer processing.
By obtaining the three-dimensional integrated design information of the first wafer and the second wafer, detecting its edge coincidence, and using the three-dimensional simulation data set to generate a three-dimensional detection image, identifying the vertical coincidence of the edge after bonding, and generating a bond offset reminder information.
Accurate evaluation and timely correction of bonding processing quality are achieved, and the reliability and efficiency of wafer processing are improved.
Smart Images

Figure CN118943036A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of processing detection technology, and in particular to a wafer processing detection method and system. Background Art
[0002] Wafer processing inspection is an important link aimed at ensuring the quality and reliability of the wafer processing process. When connecting multiple layers of wafers in traditional wafer processing, the evaluation of edge overlap usually relies on manual visual inspection or simple measurement methods. However, this method is subjective and inaccurate, and it is difficult to meet the needs of efficient production. In addition, when there is a position deviation between the upper and lower integrated wafers, it may cause unstable or incomplete connections between adjacent wafers, thereby affecting the overall wafer processing quality.
[0003] Therefore, it is necessary to accurately evaluate edge overlap and provide reminder information of bonding offset during wafer processing, so as to achieve automation and improve accuracy in the wafer processing process, thereby improving production efficiency and quality reliability. Summary of the invention
[0004] The present application provides a wafer processing detection method and system, aiming to solve the technical problems that the prior art cannot accurately detect three-dimensional bonded wafers when connecting multi-layer wafers, and lacks offset analysis, resulting in the inability to accurately evaluate the bonding quality during wafer processing.
[0005] In view of the above problems, the present application provides a wafer processing detection method and system.
[0006] The first aspect disclosed in the present application provides a wafer processing detection method, the method comprising: obtaining three-dimensional integrated design information of a first wafer and a second wafer; performing overlap detection based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer, respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; if the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than a preset edge overlap, obtaining a second detection instruction; using the three-dimensional integrated design information to generate a three-dimensional simulation data set, and controlling a wafer detection device to detect a three-dimensional wafer body with the second detection instruction, and outputting a three-dimensional detection image set, wherein the three-dimensional wafer body is a wafer after three-dimensional through-hole bonding processing of the first wafer and the second wafer; performing edge recognition according to the three-dimensional detection image set, and obtaining bonding processing overlap with the edge recognition data, wherein the bonding processing overlap is to identify the vertical overlap of the edges of the first wafer and the second wafer after bonding; generating bonding offset reminder information according to the bonding processing overlap.
[0007] Another aspect disclosed in the present application provides a wafer processing detection system, which is used for the above method, and the system includes: a design information acquisition unit, which is used to acquire three-dimensional integrated design information of a first wafer and a second wafer; an overlap detection unit, which is used to perform overlap detection based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer, respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; a detection instruction generation unit, which is used to obtain a second detection instruction if the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than a preset edge overlap; a detection image acquisition unit, which is used to detect the edge overlap of the first wafer and the second wafer based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer, respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; a detection instruction generation unit, which is used to detect the edge overlap of the first wafer and the second wafer, respectively, to obtain the edge overlap of the first wafer and the second wafer, respectively. The detection image acquisition unit is used to generate a three-dimensional simulation data set using the three-dimensional integrated design information, and control the wafer detection equipment to detect the three-dimensional wafer body with the second detection instruction, and output a three-dimensional detection image set, wherein the three-dimensional wafer body is a wafer after the first wafer and the second wafer are processed by three-dimensional through-hole bonding; the coincidence acquisition unit, the coincidence acquisition unit is used to perform edge recognition according to the three-dimensional detection image set, and obtain the bonding processing coincidence with the edge recognition data, wherein the bonding processing coincidence is the vertical coincidence of the edges after the first wafer and the second wafer are bonded; the reminder information generation unit, the reminder information generation unit is used to generate bonding offset reminder information according to the bonding processing coincidence.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: By performing overlap detection based on the thickness edges of the first wafer and the second wafer, the edge overlap can be accurately calculated, which can help judge the quality of the bonding process and provide quantitative metrics; using three-dimensional integrated design information, a three-dimensional simulation data set is generated, which contains the geometric information of the wafer after bonding and can be used for subsequent wafer detection; by performing edge recognition on the three-dimensional detection image set, the overlap of the bonding process can be calculated, which provides a reliable assessment of the bonding quality and can detect the bonding offset in time; according to the bonding process overlap, a bonding offset reminder information is generated, which helps the operator to take timely measures to correct the offset problem and avoid further damage to the quality and reliability of the wafer processing. In summary, the wafer processing detection method solves the edge projection overlap detection and bonding offset reminder problems in the prior art, realizes accurate assessment and timely correction of the bonding process quality, and thus improves the reliability and efficiency of wafer processing.
[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 A schematic diagram of a wafer processing detection method flow is provided for an embodiment of the present application; Figure 2 A schematic diagram of the structure of a wafer processing detection system is provided for an embodiment of the present application.
[0011] Explanation of the accompanying drawings: design information acquisition unit 10, coincidence detection unit 20, detection instruction generation unit 30, detection image acquisition unit 40, coincidence acquisition unit 50, reminder information generation unit 60. DETAILED DESCRIPTION
[0012] The embodiment of the present application provides a wafer processing detection method to solve the technical problems that the prior art cannot accurately detect three-dimensional bonded wafers when connecting multi-layer wafers, and lacks offset analysis, resulting in the inability to accurately evaluate the bonding quality during wafer processing.
[0013] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.
[0014] Embodiment 1 like Figure 1 As shown, an embodiment of the present application provides a wafer processing detection method, the method comprising: Acquiring three-dimensional integrated design information of the first wafer and the second wafer; The design files of the first wafer and the second wafer are read by computer-aided design software, such as CAD software, to obtain three-dimensional integrated design information. The three-dimensional integrated design information refers to design data on the geometric shape, size, structure, bonding, etc. of the wafer, including the geometric model, position coordinates, size parameters, etc. of the wafer.
[0015] Performing overlap detection based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; Through optical detection equipment, such as a CMOS collector in an image acquisition module, images of the first wafer and the second wafer are acquired, including the upper and lower edges of the two wafers. Using an edge detection algorithm, such as Canny edge detection, edge detection operations are performed on the acquired images to identify the edge contours of the wafers. The edge contours of the first wafer and the second wafer are matched, for example, using a shape matching algorithm, the overlap between the edges is calculated, and a metric value representing the edge overlap is obtained. According to the results of the edge matching, the edge overlap of the first wafer and the edge overlap of the second wafer are calculated by comparing the matched edge lengths, overlapping areas, etc. The acquisition of this edge overlap provides a reference for subsequent wafer inspection and processing.
[0016] If the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than a preset edge overlap, obtaining a second detection instruction; According to the edge overlap calculation result, the edge overlap of the first wafer and the edge overlap of the second wafer are compared with the preset edge overlap. If the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than the preset edge overlap, the condition is met, and according to the result of meeting the condition, a second detection instruction is obtained, which is used to control the relevant parameters of the wafer detection device for detecting the three-dimensional wafer body.
[0017] Generate a three-dimensional simulation data set using the three-dimensional integrated design information, and control a wafer detection device to detect a three-dimensional wafer body using the second detection instruction, and output a three-dimensional detection image set, wherein the three-dimensional wafer body is a wafer after the first wafer and the second wafer are processed by three-dimensional through-hole bonding; Based on the acquired design information, a three-dimensional simulation data set is generated by converting the design information into geometric elements such as points, lines, and surfaces in a three-dimensional coordinate system, and then three-dimensional modeling is performed to generate a three-dimensional model representing the first wafer and the second wafer.
[0018] Holes for through-hole bonding are prepared in advance on the joining surfaces of the first wafer and the second wafer. The first wafer and the second wafer are aligned and positioned by optical alignment equipment and mechanical clamping devices. The through holes of the two wafers are aligned and bonded under appropriate pressure and temperature to combine them into a whole, thereby generating the three-dimensional wafer body.
[0019] Using the wafer inspection equipment, according to the second inspection instruction, the three-dimensional wafer body is inspected, including using the image acquisition module to collect the top view image and the side view image of the wafer, and using the vertical inspection module to detect the verticality of the wafer. By integrating the data obtained from the inspection of the three-dimensional wafer body, a three-dimensional inspection image set is generated as an output result to display information such as the processing quality of the wafer and the effect of through-hole bonding.
[0020] Furthermore, the wafer inspection equipment includes: An image acquisition module, wherein the image acquisition module comprises a CMOS collector, and the CMOS collector is used to acquire an image of the bonded three-dimensional wafer body; A vertical detection module is connected to the image acquisition module and is used to detect the verticality of the CMOS collector and the measured plane of the three-dimensional wafer. When a preset vertical angle is met, the CMOS collector is triggered to collect top view and side view images and output the three-dimensional detection image set.
[0021] The image acquisition module is part of the wafer inspection equipment, which includes a CMOS collector. The main function of this module is to collect images of the bonded three-dimensional wafer body for subsequent analysis and processing. Among them, the CMOS (complementary metal oxide semiconductor) collector is an image sensor composed of many photodiodes that can convert optical signals into electrical signals. The bonded three-dimensional wafer body is placed in the image acquisition module and illuminated by an appropriate light source. The CMOS collector scans the three-dimensional wafer body and converts the optical signal on it into an electrical signal. The collected image data is converted from analog to digital and transmitted to the subsequent processing unit for further analysis and processing.
[0022] The vertical detection module is connected to the image acquisition module through communication. The vertical detection module detects the vertical angle of the CMOS collector relative to the measured plane by using a sensor, and sets a preset vertical angle according to specific application requirements and process requirements. When it is detected that the verticality of the CMOS collector and the measured plane meets the preset conditions, that is, when the detected vertical angle is less than or equal to the preset vertical angle, the CMOS collector is triggered to collect overhead and side images, and the overhead and side images corresponding to the vertical angle are obtained. After processing and integration, these images form a three-dimensional detection image set as the output result.
[0023] Furthermore, the wafer inspection device further includes a light compensation control module, the light compensation control module includes an LED light group, and the method includes: Collecting the real-time light environment of the image acquisition module to obtain ambient light information; Performing illumination uniformity identification according to the ambient light information to obtain illumination uniformity; A light compensation control parameter is generated according to the light uniformity, and the light compensation control parameter is input into the light compensation control module, so as to uniformly adjust the ambient light by adjusting the LED light group.
[0024] The light compensation control module is another component of the wafer inspection equipment, including the LED light group, which is used to control the brightness and color of the ambient light to achieve uniform adjustment of the lighting. Among them, the LED light group is a group of light-emitting diodes. By controlling the brightness and color of each LED, the characteristics of the light it emits can be changed.
[0025] Start the image acquisition module to capture the image of the wafer surface, which contains reflected and scattered light from the ambient light. Preprocess the acquired image, including removing noise, adjusting the brightness and contrast of the image, etc. Perform light analysis on the preprocessed image, such as calculating the average brightness value, color distribution, light gradient, etc. of the image to obtain ambient light information.
[0026] Based on the ambient light information, the illumination uniformity is identified. For example, the image is divided into different areas, and the brightness difference of the pixels in each area is calculated. A smaller difference indicates a higher illumination uniformity. The distribution of different color channels in the image is analyzed. If the distribution of each channel is relatively uniform, it indicates that the illumination uniformity is good. According to the calculation result of the brightness difference and the distribution of the color channel, the illumination uniformity is obtained to indicate the uniformity of the ambient light.
[0027] According to the illumination uniformity, light compensation control parameters are generated. For example, according to the illumination uniformity parameters, the LED light group is adjusted in different regions to make the light in each region more uniform; the color of the LED light group is adjusted, such as changing the brightness ratio of the red, green, and blue channels to achieve a more uniform illumination effect; according to the illumination uniformity parameters, the overall brightness of the LED light group is adjusted to illuminate the wafer surface more evenly. The generated light compensation control parameters are input into the light compensation control module, and the LED light group is adjusted through the light compensation control module. According to the generated light compensation control parameters, the brightness, color, or illumination intensity of the local area of the LED light group is adjusted to achieve a more uniform ambient lighting effect. In this way, more consistent illumination conditions can be obtained in each region, thereby improving the accuracy and reliability of wafer detection.
[0028] Perform edge recognition according to the three-dimensional detection image set, and obtain bonding process overlap using edge recognition data, wherein the bonding process overlap is a vertical overlap of edges of the first wafer and the second wafer after bonding; Furthermore, edge recognition is performed according to the three-dimensional detection image set, and bonding processing overlap is obtained using edge recognition data, including: Acquire edge sample data of a top view image and edge sample data of a side view image of the first wafer; Acquire the three-dimensional detection image set, wherein the three-dimensional detection image set includes a top view image set and a side view image set; Acquire a first processing overlap by extracting edge data of the overhead image and comparing it with edge sample data of the overhead image; Acquire a second processing overlap by extracting edge data of the side view image and comparing it with edge sample data of the side view image; The bonding processing overlap is obtained according to the first processing overlap and the second processing overlap.
[0029] Use a CMOS collector to collect top view and side view images of the first wafer, output the top view and side view images of the first wafer, apply edge detection algorithms, such as Canny edge detection, to extract edges from the top view image, obtain edge information of the first wafer, select representative edge samples from the edge information, such as key feature points, edge segments, etc., the edge samples represent the edge shape and position of the first wafer, and form an edge sample data set of the top view image. In the same way, obtain an edge sample data set of the side view image. The acquired edge sample data is used as a reference standard when evaluating the degree of edge matching in the actual detection image.
[0030] In the aforementioned steps, a CMOS collector in a wafer inspection device is used to collect top view images and side view images of a three-dimensional wafer body, and the three-dimensional inspection image set is output, wherein the three-dimensional inspection image set includes a top view image set and a side view image.
[0031] The edge detection algorithm is also used to extract the edge of the top view image to obtain edge data representing the edge position of the three-dimensional wafer body. The edge sample data of the top view image of the first wafer is used as a reference sample data set, the extracted edge data is compared with the edge sample data, and a matching algorithm, such as a correlation matching method, is used to calculate the difference between the edge data and the sample data. Based on the obtained difference, the first processing overlap is obtained, which reflects the degree of edge matching.
[0032] The edge data extracted from the side view image is compared with the edge sample data, and the relative spacing between every two adjacent edge lines of the two wafer edge lines is identified to obtain the relative spacing uniformity, thereby determining the spacing offset. The spacing offset is used as the second processing overlap, which reflects the vertical overlap of the edges.
[0033] Different weights are assigned to the first processing overlap and the second processing overlap, and then a weighted comprehensive calculation is performed. The choice of weights can be determined based on specific circumstances and needs. For example, weights are allocated based on edge importance, accuracy requirements, etc. Based on the result of the weighted calculation, a numerical representation of the bonding processing overlap is generated. This value can represent the degree of vertical overlap of the edges of the first wafer and the second wafer after bonding, reflecting the accuracy and quality of the bonding process.
[0034] Furthermore, extracting edge data of the overhead image and comparing it with edge sample data of the overhead image to obtain a first processing overlap includes: By performing grayscale conversion on the top view image, an edge grayscale pixel coordinate set in the top view image is obtained based on wafer edge detection; Obtaining a sample edge grayscale pixel coordinate set according to the edge sample data of the overhead image; The edge grayscale pixel coordinate set and the sample edge grayscale pixel coordinate set are used to compare the grayscale values of the coordinate points to obtain a coordinate coincidence rate, and the coordinate coincidence rate is output as the first processing coincidence rate.
[0035] The top view image is converted into a grayscale image by converting the RGB channels of the color image into grayscale values. The grayscale image is processed using edge detection algorithms, such as Canny edge detection, Sobel operator, etc. These algorithms can extract areas in the image where the grayscale changes dramatically and identify them as edges.
[0036] After edge detection is completed, a binary image representing the edge is obtained, in which the edge part is marked as white and the non-edge part is black. By traversing the pixels of the image, the coordinates of the white pixels (edge pixels) are found and saved in the edge grayscale pixel coordinate set, which represents the position of the wafer edge detected in the top view image.
[0037] The same method is used to perform grayscale conversion on the edge sample data of the overhead image to obtain a sample edge grayscale pixel coordinate set.
[0038] For each coordinate point in the edge grayscale pixel coordinate set, obtain its corresponding grayscale value, find the coordinate point at the same position in the sample edge grayscale pixel coordinate set, and obtain its corresponding grayscale value, perform a coordinate point grayscale value comparison, such as performing a correlation calculation, and obtain the similarity between the two grayscale values. According to the results of the coordinate point grayscale value comparison, count the number of coordinate points with a similarity higher than a preset threshold, compare the number of these well-matched coordinate points with the total number of coordinate points, and obtain the coordinate coincidence rate, which is expressed as a percentage. The calculated coordinate coincidence rate is used as a metric for the first processing coincidence degree, which can represent the matching degree and overlap between the edge in the overhead image and the sample edge, reflecting the accuracy and quality of the first processing.
[0039] Furthermore, extracting edge data of the side view image and comparing it with edge sample data of the side view image to obtain a second processing coincidence degree includes: By performing grayscale conversion on the side view image, two wafer edge lines of the side view image are obtained based on wafer edge detection; Identifying the relative spacing between each two adjacent edge lines of the two wafer edge lines to obtain relative spacing uniformity; A pitch deviation is determined according to the relative pitch uniformity, and the pitch deviation is output as the second machining overlap.
[0040] The side view image is converted into grayscale using the same method as above, and the grayscale image is processed using an edge detection algorithm to obtain an edge image, and a curve fitting method is used to extract the edge line to obtain the respective edge lines of the first wafer and the second wafer after bonding.
[0041] For every two adjacent wafer edge lines, the relative spacing between them is calculated by the Euclidean distance between the straight lines. Based on the calculated relative spacing, the spacing between all adjacent wafer edge lines is calculated and summed up. The total spacing is divided by the number of wafer edge lines to obtain the average spacing. For the spacing between each adjacent wafer edge line, the difference between each spacing and the average spacing is calculated by subtracting the average spacing from each spacing, that is, the uniformity of the relative spacing. A smaller difference indicates a more uniform spacing. This uniformity information is used to determine whether the spacing distribution between wafer edge lines is uniform, thereby determining the shape and processing quality of the wafer.
[0042] The relative spacing uniformity is calculated in absolute value to obtain their absolute values, which are used to represent the spacing deviation. A larger deviation value indicates uneven spacing, while a smaller value indicates more uniform spacing. The calculated spacing deviation is output as a metric of the second processing coincidence. The second processing coincidence reflects the spacing uniformity and processing quality of the wafer edge in the side view image. The spacing deviation can be used to quantitatively evaluate this uniformity difference.
[0043] Bonding offset reminder information is generated according to the bonding processing overlap.
[0044] Based on the numerical value of the bonding processing overlap, a threshold is set according to actual needs. When the bonding processing overlap is greater than or equal to the threshold, it indicates that the bonding meets the preset requirements; when the bonding processing overlap is less than the threshold, it indicates that there is an offset problem in the bonding. In this case, a bonding offset reminder message is generated, such as a warning message, an error prompt or a guidance suggestion, to notify the operator of the existence of a bonding offset, thereby prompting the operator to take corresponding corrective measures.
[0045] Furthermore, it also includes: If the bonding process overlap is greater than or equal to a preset bonding process overlap, determining whether to integrate a third wafer, and if the third wafer is integrated, performing bonding offset identification based on the first wafer, the second wafer, and the third wafer; If the bonding process overlap is less than the preset bonding process overlap, a bonding offset reminder message is generated.
[0046] The calculated bonding process overlap is compared with the preset bonding process overlap. If it is greater than or equal to the preset bonding process overlap, it means that the bonding process overlap meets the conditions, and further judgment is made as to whether the third wafer is to be integrated. If the third wafer needs to be integrated, the next step of bonding offset identification is entered; otherwise, the processing flow is terminated.
[0047] When it is decided to integrate the third wafer, the edge lines or feature information of the first wafer, the second wafer and the third wafer are used to identify the bonding offset, and the aforementioned image acquisition, image processing and edge detection operations are repeated. By comparing the positional relationship and alignment degree between the edge lines, the offset in the bonding process is determined and quantitatively evaluated.
[0048] When the bonding processing overlap is less than the preset bonding processing overlap, it indicates that there is a bonding offset problem. In this case, a bonding offset reminder message is generated, such as a warning message, an error prompt, or a guidance suggestion, to inform the operator of the bonding offset, thereby prompting the operator to take corresponding corrective measures.
[0049] In summary, the wafer processing detection method and system provided by the embodiments of the present application have the following technical effects: 1. By performing overlap detection based on the thickness edges of the first wafer and the second wafer, the edge overlap can be accurately calculated, which can help determine the quality of the bonding process and provide quantitative metrics; 2. Using the 3D integrated design information, generate 3D simulation data sets, which contain the geometric information of the bonded wafers and can be used for subsequent wafer inspection; 3. By performing edge recognition on the 3D inspection image set, the overlap of the bonding process can be calculated, which provides a reliable assessment of the bonding quality and can detect bonding deviation in a timely manner; 4. Generate bonding offset reminder information based on bonding process overlap, which helps operators take timely measures to correct the offset problem and avoid further damage to the quality and reliability of wafer processing.
[0050] In summary, the wafer processing detection method solves the edge projection coincidence detection and bonding offset reminder problems in the prior art, realizes accurate evaluation and timely correction of bonding processing quality, thereby improving the reliability and efficiency of wafer processing.
[0051] Embodiment 2 Based on the same inventive concept as the wafer processing detection method in the above-mentioned embodiment, Figure 2 As shown, the present application provides a wafer processing detection system, the system comprising: A design information acquisition unit 10, wherein the design information acquisition unit 10 is used to acquire three-dimensional integrated design information of the first wafer and the second wafer; An overlap detection unit 20, the overlap detection unit 20 is used to perform overlap detection based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer, respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; A detection instruction generating unit 30, wherein the detection instruction generating unit 30 is used to obtain a second detection instruction if the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than a preset edge overlap; A detection image acquisition unit 40, the detection image acquisition unit 40 is used to generate a three-dimensional simulation data set using the three-dimensional integrated design information, and control the wafer detection equipment to detect the three-dimensional wafer body with the second detection instruction, and output a three-dimensional detection image set, wherein the three-dimensional wafer body is a wafer after the first wafer and the second wafer are subjected to three-dimensional through-hole bonding processing; An overlap acquisition unit 50, the overlap acquisition unit 50 is used to perform edge recognition according to the three-dimensional detection image set, and obtain bonding process overlap with edge recognition data, wherein the bonding process overlap is a vertical overlap of the edges of the first wafer and the second wafer after bonding; The reminder information generating unit 60 is used to generate bonding offset reminder information according to the bonding processing overlap.
[0052] Furthermore, the wafer inspection equipment includes: An image acquisition module, wherein the image acquisition module comprises a CMOS collector, and the CMOS collector is used to acquire an image of the bonded three-dimensional wafer body; A vertical detection module is connected to the image acquisition module and is used to detect the verticality of the CMOS collector and the measured plane of the three-dimensional wafer. When a preset vertical angle is met, the CMOS collector is triggered to collect top view and side view images and output the three-dimensional detection image set.
[0053] Furthermore, the wafer inspection device further includes a light compensation control module, the light compensation control module includes an LED light group, and the inspection image acquisition unit 40 includes the following operation steps: Collecting the real-time light environment of the image acquisition module to obtain ambient light information; Performing illumination uniformity identification according to the ambient light information to obtain illumination uniformity; A light compensation control parameter is generated according to the light uniformity, and the light compensation control parameter is input into the light compensation control module, so as to uniformly adjust the ambient light by adjusting the LED light group.
[0054] Furthermore, the system further includes a processing overlap acquisition module to perform the following operation steps: Acquire edge sample data of a top view image and edge sample data of a side view image of the first wafer; Acquire the three-dimensional detection image set, wherein the three-dimensional detection image set includes a top view image set and a side view image set; Acquire a first processing overlap by extracting edge data of the overhead image and comparing it with edge sample data of the overhead image; Acquire a second processing overlap by extracting edge data of the side view image and comparing it with edge sample data of the side view image; The bonding processing overlap is obtained according to the first processing overlap and the second processing overlap.
[0055] Furthermore, the system further comprises a first processing coincidence output module to perform the following operation steps: By performing grayscale conversion on the top view image, an edge grayscale pixel coordinate set in the top view image is obtained based on wafer edge detection; Obtaining a sample edge grayscale pixel coordinate set according to the edge sample data of the overhead image; The edge grayscale pixel coordinate set and the sample edge grayscale pixel coordinate set are used to compare the grayscale values of the coordinate points to obtain a coordinate coincidence rate, and the coordinate coincidence rate is output as the first processing coincidence rate.
[0056] Furthermore, the system further includes a second processing coincidence output module to perform the following operation steps: By performing grayscale conversion on the side view image, two wafer edge lines of the side view image are obtained based on wafer edge detection; Identifying the relative spacing between each two adjacent edge lines of the two wafer edge lines to obtain relative spacing uniformity; A pitch deviation is determined according to the relative pitch uniformity, and the pitch deviation is output as the second machining overlap.
[0057] Furthermore, the system further includes an offset reminder information generation module to perform the following operation steps: If the bonding process overlap is greater than or equal to a preset bonding process overlap, determining whether to integrate a third wafer, and if the third wafer is integrated, performing bonding offset identification based on the first wafer, the second wafer, and the third wafer; If the bonding process overlap is less than the preset bonding process overlap, a bonding offset reminder message is generated.
[0058] Through the above-mentioned detailed description of a wafer processing detection method in this specification, those skilled in the art can clearly understand a wafer processing detection method and system in this embodiment. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.
[0059] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A wafer processing detection method, It is characterized in that The method comprises: Acquiring three-dimensional integrated design information of the first wafer and the second wafer; Performing overlap detection based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; If the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than a preset edge overlap, obtaining a second detection instruction; Generate a three-dimensional simulation data set using the three-dimensional integrated design information, and control a wafer detection device to detect a three-dimensional wafer body using the second detection instruction, and output a three-dimensional detection image set, wherein the three-dimensional wafer body is a wafer after the first wafer and the second wafer are processed by three-dimensional through-hole bonding; Perform edge recognition according to the three-dimensional detection image set, and obtain bonding process overlap using edge recognition data, wherein the bonding process overlap is a vertical overlap of edges of the first wafer and the second wafer after bonding; Bonding offset reminder information is generated according to the bonding processing overlap.
2. The method according to claim 1, It is characterized in that The wafer inspection equipment comprises: An image acquisition module, wherein the image acquisition module comprises a CMOS collector, and the CMOS collector is used to acquire an image of the bonded three-dimensional wafer body; A vertical detection module is connected to the image acquisition module and is used to detect the verticality of the CMOS collector and the measured plane of the three-dimensional wafer. When a preset vertical angle is met, the CMOS collector is triggered to collect top view and side view images and output the three-dimensional detection image set.
3. The method according to claim 2, It is characterized in that The wafer inspection device further includes a light compensation control module, the light compensation control module includes an LED light group, and the method includes: Collecting the real-time light environment of the image acquisition module to obtain ambient light information; Performing illumination uniformity identification according to the ambient light information to obtain illumination uniformity; A light compensation control parameter is generated according to the light uniformity, and the light compensation control parameter is input into the light compensation control module, so as to uniformly adjust the ambient light by adjusting the LED light group.
4. The method according to claim 1, It is characterized in that Performing edge recognition according to the three-dimensional detection image set and obtaining bonding processing overlap using edge recognition data, the method further includes: Acquire edge sample data of the top view image and edge sample data of the side view image of the first wafer; Acquire the three-dimensional detection image set, wherein the three-dimensional detection image set includes a top view image set and a side view image set; Acquire a first processing overlap by extracting edge data of the overhead image and comparing it with edge sample data of the overhead image; Acquire a second processing overlap by extracting edge data of the side view image and comparing it with edge sample data of the side view image; The bonding processing overlap is obtained according to the first processing overlap and the second processing overlap.
5. The method according to claim 4, It is characterized in that By extracting edge data of the overhead image and comparing it with edge sample data of the overhead image, a first processing overlap is obtained, the method comprising: By performing grayscale conversion on the top view image, an edge grayscale pixel coordinate set in the top view image is obtained based on wafer edge detection; Obtaining a sample edge grayscale pixel coordinate set according to the edge sample data of the overhead image; The edge grayscale pixel coordinate set and the sample edge grayscale pixel coordinate set are used to compare the grayscale values of the coordinate points to obtain a coordinate coincidence rate, and the coordinate coincidence rate is output as the first processing coincidence rate.
6. The method according to claim 4, It is characterized in that By extracting edge data from the side view image and comparing it with edge sample data of the side view image, a second processing coincidence degree is obtained, and the method further includes: By performing grayscale conversion on the side view image, two wafer edge lines of the side view image are obtained based on wafer edge detection; Identifying the relative spacing between each two adjacent edge lines of the two wafer edge lines to obtain relative spacing uniformity; A pitch deviation is determined according to the relative pitch uniformity, and the pitch deviation is output as the second machining overlap.
7. The method according to claim 1, It is characterized in that The method further comprises: If the bonding process overlap is greater than or equal to a preset bonding process overlap, determining whether to integrate a third wafer, and if the third wafer is integrated, performing bonding offset identification based on the first wafer, the second wafer, and the third wafer; If the bonding process overlap is less than the preset bonding process overlap, a bonding offset reminder message is generated.
8. A wafer processing detection system, It is characterized in that A wafer processing detection method for implementing any one of claims 1 to 7, comprising: A design information acquisition unit, the design information acquisition unit is used to acquire three-dimensional integrated design information of the first wafer and the second wafer; An overlap detection unit, the overlap detection unit is used to perform overlap detection based on the upper edge and the lower edge of the thickness of the first wafer and the second wafer, respectively, to obtain the edge overlap of the first wafer and the edge overlap of the second wafer; A detection instruction generating unit, wherein the detection instruction generating unit is used to obtain a second detection instruction if the edge overlap of the first wafer and the edge overlap of the second wafer are both greater than a preset edge overlap; A detection image acquisition unit, the detection image acquisition unit is used to generate a three-dimensional simulation data set using the three-dimensional integrated design information, and control the wafer detection equipment to detect the three-dimensional wafer body with the second detection instruction, and output a three-dimensional detection image set, wherein the three-dimensional wafer body is a wafer after the first wafer and the second wafer are subjected to three-dimensional through-hole bonding processing; An overlap acquisition unit, the overlap acquisition unit is used to perform edge recognition according to the three-dimensional detection image set, and obtain bonding process overlap with edge recognition data, wherein the bonding process overlap is a vertical overlap of the edges of the first wafer and the second wafer after bonding; A reminder information generating unit, wherein the reminder information generating unit is used to generate bonding offset reminder information according to the bonding processing overlap.
Citation Information
Patent Citations
Wafer offset detection method and system, storage medium and electronic equipment
CN115861183A
Method for wafer bonding alignment through edge detection
CN116313971A
Wafer surface defect detection method and system
CN116773548A
Wafer alignment method and device, computer equipment and storage medium
CN117438360A
Alignment method and device for multi-layer lamination of wafer
CN117672940A
