Wafer defect inspection equipment
The wafer defect detection is achieved through the image sensor array on the entire wafer, which solves the problems of low detection efficiency and large land use of existing equipment, improves detection efficiency and cost-effectiveness, and is suitable for the rapid detection of highly integrated semiconductor devices.
Patent Information
- Application Number
- CN202011073358.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-10-09
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2040-10-09
AI Technical Summary
The existing wafer defect detection equipment has low detection efficiency, large area, high cost, and requires complex calibration structures and optical magnifier scanning methods, resulting in increased equipment volume and energy consumption.
The image acquisition module is adopted, including an image sensor array on the entire wafer, and the entire surface image of the wafer to be detected is obtained through one shot, simplifying the detection process, reducing the device size and energy consumption, and determining whether the wafer surface is determined by the defect judgment module. There are defects.
It improves the efficiency of defect detection, reduces cost, simplifies the equipment structure, reduces the occupied volume and energy consumption, realizes fast focus and fast shooting, and is suitable for high-integration detection of semiconductor devices.
Smart Images

Figure CN112201596B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wafer defect detection, and in particular to a high-speed wafer detection device. Background Art
[0002] With the rapid development of semiconductor manufacturing technology, semiconductor devices are developing towards higher integration in order to achieve faster computing speeds, larger storage capacities and more functions. The higher the integration of semiconductor chips, the smaller the critical dimension (CD) of the semiconductor device.
[0003] Photolithography is the most important process in semiconductor integrated manufacturing. When manufacturing integrated circuits, multi-layer processes are required, and each layer must be accurately aligned. Therefore, in order to ensure that the patterns between the layers of photolithography are accurately stacked, the chip must be inspected after the development process is completed. After-development inspection can detect errors in the photolithography process and correct them in time. This is one of the few steps in the chip manufacturing process that can be corrected. Once a wafer with defective photoresist is sent to the next step of pattern formation (etching), it is very easy to cause the wafer to be scrapped. Specifically, the after-development inspection generally includes checking the coverage, alignment, exposure, and development of the photoresist one by one, and judging whether the photoresist performance meets the process requirements.
[0004] The existing post-development inspection is carried out through detection equipment, specifically through optical magnifying glass scanning on the detection equipment. The existing detection and scanning methods have low detection efficiency, low output per unit time, and relatively lagging data processing. In addition, the optical magnifying glass scanning method requires the equipment to be equipped with a larger magnifying glass and a larger stage and stage drive device, which makes the detection equipment occupy a large area and is expensive. Summary of the invention
[0005] The technical problem to be solved by the present invention is how to improve the efficiency of defect detection and reduce the area occupied by the detection equipment.
[0006] The present invention provides a wafer defect detection device, comprising:
[0007] Wafer stage, used to fix the wafer to be inspected;
[0008] An image acquisition module, wherein the image acquisition module comprises a first wafer, wherein the first wafer comprises a first surface and a second surface opposite to each other, wherein an image sensor array is formed on the first surface, and wherein the image acquisition module obtains a detection image corresponding to the entire surface of the wafer to be detected by taking a single shot through the image sensor array;
[0009] A defect judgment module is used to judge whether there are defects on the surface of the wafer to be inspected according to the inspection image obtained by the image acquisition module.
[0010] Optionally, the image sensor array is formed in the first wafer through a semiconductor integrated manufacturing process, and the image sensor array includes a plurality of image sensors, and the plurality of image sensors are arranged in rows and columns and distributed in the first wafer.
[0011] Optionally, the image acquisition module also includes a second wafer, which is bonded to the second surface of the first wafer, and a plurality of signal processing circuits are formed in the second wafer. The plurality of signal processing circuits are used to perform image processing on a plurality of images corresponding to the same shooting of a plurality of image sensors to obtain a detection image corresponding to the entire surface of the wafer to be inspected.
[0012] Optionally, the image processing includes image stitching, and the image stitching includes an image preprocessing step, an image registration step, a change model establishment step, a same coordinate change step, and a fusion reconstruction step.
[0013] Optionally, the image processing also includes one or more of image grayscale processing, image clarity processing, and synthesis processing of multiple defocused images.
[0014] Optionally, the image sensor further includes a filter layer and a plurality of microlenses located on the filter.
[0015] Optionally, the image sensor array further includes a lens assembly array corresponding to a plurality of image sensors and arranged on the first surface of the first wafer.
[0016] Optionally, the size of the image sensor array corresponds to the size of the wafer to be inspected.
[0017] Optionally, the defect judgment module includes a standard unit and a comparison unit, the standard unit stores a standard wafer image or a defect-free wafer image, and the comparison unit is used to compare or match the detection image obtained by the image acquisition module with the standard wafer image or the defect-free wafer image, and then determine whether there are defects on the detection image and the location of the defects; the defect judgment module also learns and obtains the type of defects during the process of defect judgment.
[0018] Optionally, the wafer stage can rotate, move vertically, move horizontally, and deflect.
[0019] Compared with the prior art, the technical solution of the present invention has the following advantages:
[0020] The wafer defect detection device of the present invention comprises: a wafer stage for fixing a wafer to be detected; an image acquisition module, the image acquisition module comprises a first wafer, the first wafer comprises a first surface and a second surface relative to each other, the first surface is formed with an image sensor array, the image acquisition module obtains a detection image corresponding to the entire surface of the wafer to be detected by taking a single shot through the image sensor array; a defect judgment module, the defect judgment module judges whether there is a defect on the surface of the wafer to be detected according to the detection image obtained by the image acquisition module. Since the image acquisition module comprises a first wafer, and an image sensor array is formed in the first wafer, the positions between adjacent image sensors in the image sensor array are fixed, so that when the image acquisition module obtains the detection image of the entire surface of the wafer to be detected through the image sensor array, the front, back, left, right and height positions between adjacent image sensors do not need to be calibrated, which simplifies the operation process of the wafer defect detection device, and does not need to set up an additional auxiliary calibration structure, thereby simplifying the structure of the image acquisition module and reducing the volume of the image acquisition module. In addition, since the entire first wafer is used to form the image sensor array, there is no need to set cutting paths like the existing integrated circuit manufacturing wafer, and then divide the wafer into individual chips. Therefore, there is more space on the first wafer for forming image sensors and functional circuits connected to the image sensors.
[0021] Furthermore, the image acquisition module obtains the detection image corresponding to the entire surface of the wafer to be detected by shooting once through the image sensor array, so the wafer stage does not need to perform scanning movement (movement in the horizontal direction) when acquiring the detection image, and the wafer stage described in this application only needs to move in a small range or distance in the vertical direction and / or in the horizontal direction when aligning with the image sensor array, so the wafer stage and the corresponding drive unit or drive device can be small in size, and the volume occupied by the image sensor array is also small, thereby reducing the volume occupied by the entire wafer defect detection equipment and reducing energy consumption (driving small volume reduces energy consumption). Furthermore, in this application, the detection image corresponding to the entire surface of the wafer to be detected is obtained by shooting once through the image acquisition module through the image sensor array, and the defect judgment module judges whether there is a defect on the surface of the wafer to be detected based on the detection image obtained by the image acquisition module. When performing defect detection, the time for acquiring the detection image is greatly reduced (one-time transient imaging), which improves the efficiency of defect detection and reduces costs (the cost of the image sensor array is greatly reduced compared to the optical scanning magnifier).
[0022] Furthermore, each image sensor in the image sensor array has the same size, so that when performing defect detection, the image sensor array in the image acquisition module can quickly focus and shoot the wafer to be inspected, and it is convenient to stitch images obtained by several image sensors.
[0023] Furthermore, a filter layer and a plurality of microlenses located on the filter are formed on the surface of one side of the first wafer on which the image sensor array is formed. The filter can filter out unnecessary light. The microlenses are used to converge the light reflected from the surface of the wafer to be detected to the corresponding image sensor during measurement.
[0024] Furthermore, a lens assembly array corresponding to the image sensor array is also provided on a surface of one side of the first wafer on which the image sensor array is formed, so that the focal length and magnification of the corresponding image sensor in the image sensor array can be adjusted. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 A schematic diagram of the structure of a wafer defect detection device according to an embodiment of the present invention;
[0026] Figure 2 Schematic diagram of the top view of the image acquisition module in one embodiment of the present invention;
[0027] Figure 3 It is a schematic structural diagram of a wafer defect detection device in another embodiment of the present invention. DETAILED DESCRIPTION
[0028] The present invention provides a wafer defect detection device, comprising: the wafer defect detection device of the present invention, comprising: a wafer stage for fixing a wafer to be detected; an image acquisition module, the image acquisition module comprising a first wafer, the first wafer comprising a first surface and a second surface relative to each other, the first surface being formed with an image sensor array, the image acquisition module taking a single shot through the image sensor array to obtain a detection image corresponding to the entire surface of the wafer to be detected; a defect judgment module, the defect judgment module judging whether there is a defect on the surface of the wafer to be detected according to the detection image obtained by the image acquisition module. Since the image acquisition module comprises a first wafer, and the image sensor array is formed in the first wafer, the positions between adjacent image sensors in the image sensor array are fixed, so that when the image acquisition module obtains the detection image of the entire surface of the wafer to be detected through the image sensor array, the front, back, left, right and height positions between adjacent image sensors do not need to be calibrated, thereby simplifying the operation process of the wafer defect detection device, and there is no need to set up an additional auxiliary calibration structure, thereby simplifying the structure of the image acquisition module and reducing the volume of the image acquisition module. In addition, since the entire first wafer is used to form the image sensor array, there is no need to set cutting paths like the existing integrated circuit manufacturing wafer, and then separate the wafer into individual chips, so there is more space on the first wafer for forming image sensors and functional circuits connected to the image sensors. In addition, the volume occupied by the entire wafer defect detection equipment can be reduced, and energy consumption can be reduced (small driving volume reduces energy consumption), and the efficiency of defect detection is improved, and the cost is reduced (the cost of the image sensor array is greatly reduced compared to the optical scanning magnifier).
[0029] In order to make the above-mentioned objects, features and advantages of the present invention more clearly understood, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. When describing the embodiments of the present invention in detail, for the sake of convenience, the schematic diagrams will not be partially enlarged according to the general proportion, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.
[0030] An embodiment of the present invention provides a wafer defect detection device, please refer to Figure 1 and Figure 2 , Figure 2 The top view of the image acquisition module is a schematic diagram, including:
[0031] Wafer stage 101, used for fixing wafer 102 to be inspected;
[0032] An image acquisition module 103, wherein the image acquisition module 103 includes a first wafer 105, wherein the first wafer 105 includes a first surface and a second surface opposite to each other, wherein an image sensor array 104 is formed on the first surface, and the image acquisition module 103 obtains a detection image corresponding to the entire surface of the wafer 102 to be detected by taking a single shot through the image sensor array 104;
[0033] The defect judgment module 108 judges whether there are defects on the surface of the wafer 102 to be inspected according to the inspection image obtained by the image acquisition module 103 .
[0034] Specifically, the wafer stage 101 is used to fix the wafer to be detected 102, and the wafer stage 101 can fix the wafer to be detected 102 by vacuum adsorption, electrostatic adsorption or mechanical support. Specifically, the wafer defect detection device also includes a transmission unit, which is used to transfer the wafer to be detected 102 to the wafer stage 101, and then perform defect detection.
[0035] In one embodiment, the wafer stage 101 can be moved. Specifically, before the detection image is taken, the wafer stage 101 can be rotated, and can move in the vertical direction and / or move in a small range or a small distance in the horizontal direction (the specific small range or small distance movement range is plus or minus 10-25 mm), so that the wafer 102 to be detected on the wafer stage 101 is aligned with the image sensor array on the image acquisition module 103, and the wafer stage 101 can also be deflected when the detection image is taken so that the wafer to be detected is in a tilted state to obtain a dark field image. In a specific embodiment, the wafer stage 101 can be connected to a corresponding drive unit or drive device, and the drive unit or drive device can drive the wafer stage to perform corresponding actions (including rotation, small range movement in the vertical direction, small range movement in the horizontal direction, and deflection), and the specific drive unit or drive device can be a combination of a micromotor and a micromachine.
[0036] The wafer 102 to be inspected is a wafer that needs to be inspected for defects after undergoing a certain semiconductor process. Specifically, the wafer 102 to be inspected can be a wafer after undergoing a photolithography coating process, a wafer after undergoing a photolithography development process, a wafer after undergoing an etching process (including wet etching or dry etching), a wafer after undergoing a chemical mechanical polishing process, a wafer after undergoing a chemical vapor deposition process, or a wafer after undergoing a physical vapor deposition process. The material of the wafer 102 to be inspected can be silicon (Si), germanium (Ge), or silicon germanium (GeSi), silicon carbide (SiC); it can also be silicon on insulator (SOI), germanium on insulator (GOI); or it can also be other materials, such as III-V compounds such as gallium arsenide. The size (diameter) of the wafer 102 to be inspected can be 6 inches, 8 inches, 12 inches, or 18 inches.
[0037] In the present application, the subsequent defect detection on the wafer 102 to be inspected includes: detecting whether the pattern formed on the surface of the wafer 102 to be inspected is good (such as whether the pattern is complete, whether the pattern has defects), detecting whether the film layer formed on the surface of the wafer 102 to be inspected is good (such as whether the surface morphology of the film layer is good), and detecting whether there are particle defects, scratch defects or other types of defects on the surface of the wafer 102 to be inspected. Specifically, for example, after the photolithography and development process, the post-development defect detection (ADI, After Development Inspection) of the wafer 102 to be inspected specifically includes: detecting whether the photoresist pattern after development is good, the photoresist pattern includes a photoresist pattern used as a subsequent etching mask, a photoresist pattern used to measure a feature size (CD), and a photoresist pattern used to measure an overlay error (overlay), and checking whether there are particle defects, scratch defects or other types of photoresist defects. For another example, the post-etch defect inspection (AEI, After Etch Inspection) performed on the wafer to be inspected after the etching process specifically includes: inspecting whether the etching pattern is good, inspecting whether there are particle defects, scratch defects or other types of etching defects.
[0038] The image acquisition module 103 includes a first wafer 105, in which an image sensor array 104 is formed. The image sensor array 104 includes a plurality of image sensors 107, which are arranged in rows and columns and distributed in the first wafer 105 (see Figure 2 ).
[0039] The material of the first wafer 105 can be silicon (Si), germanium (Ge), or silicon germanium (GeSi), silicon carbide (SiC); it can also be silicon on insulator (SOI), germanium on insulator (GOI); or it can also be other materials, such as III-V group compounds such as gallium arsenide. In this embodiment, the first wafer 105 is a silicon wafer.
[0040] The image sensor array 104 (or a plurality of image sensors 107) is formed in the first wafer 105 through a semiconductor integrated manufacturing process, so the positions between adjacent image sensors 107 in the image sensor array 104 are fixed, so when the image acquisition module 103 acquires the detection image of the entire surface of the wafer 102 to be detected through the image sensor array 104, the front and back left and right positions and the height positions between adjacent image sensors 107 do not need to be calibrated, which simplifies the operation process of the wafer defect detection equipment, and does not require the additional auxiliary calibration structure, thereby simplifying the structure of the image acquisition module 103 and reducing the volume of the image acquisition module 103. In addition, since the entire wafer of the first wafer 105 is used to form the image sensor array 104, there is no need to set a cutting path like the existing integrated circuit manufacturing wafer, and the wafer needs to be divided into individual chips later, so there can be more space on the first wafer 105 for forming the image sensor 107 and the functional circuit connected to the image sensor 107.
[0041] In one embodiment, each of the image sensors 107 includes a plurality of photosensitive areas arranged in rows and columns in the first wafer, and the photosensitive areas are used to sense external incident light to generate electrical signals.
[0042] In one embodiment, the plurality of image sensors 107 in the first wafer 105 are CMOS image sensors.
[0043] In this embodiment, the size of the image sensor array 104 corresponds to the size of the wafer 102 to be inspected, that is, the size of the image sensor array 104 can be equal to, approximately equal to, slightly larger than, or slightly smaller than the size of the wafer 102 to be inspected. It is only necessary that the combined viewing angle of the image sensors 107 in the image sensor array 104 can cover the entire surface of the wafer 102 to be inspected.
[0044] In one embodiment, the number of image sensors 107 in the image sensor array 104 is greater than or equal to 5, and can be specifically 5 (the specific arrangement method can be to set 1 image sensor in the middle of the first wafer 105, and set one image sensor on each of the four edges), 9 (the specific arrangement method can be to arrange 9 image sensors in a 3x3 array on the first wafer 105), 13 (the specific arrangement method can be to arrange 9 image sensors in a 3x3 array in the middle of the first wafer 105, and arrange 1 image sensor on each of the four edges), 24 (the specific arrangement method can be to arrange 16 image sensors in a 4x4 array in the middle of the first wafer 105, and arrange 2 image sensors on each of the four edges), 37 (the specific arrangement method can be to arrange 25 image sensors in a 5x5 array in the middle of the first wafer 105, and arrange 3 image sensors on each of the four edges), 69 (the specific arrangement method can be to arrange 7x4 array in the middle of the first wafer 105, and ... 7 arrays of 49 image sensors, with 5 image sensors arranged on each of the four edges). It should be noted that the number of image sensors 107 in the image sensor array 104 can also be other numbers, and the specific number can be set according to the size of the wafer 102 to be inspected, the clarity of the image, the accuracy of the inspection, etc., as long as the combined viewing angle of the image sensors in the image sensor array 104 can cover the entire surface of the wafer 102 to be inspected.
[0045] In one embodiment, continue to refer to Figure 1 and Figure 2 Each image sensor 107 in the image sensor array 104 has the same size, so that when performing defect detection, the image sensor array 104 in the image acquisition module 103 can quickly focus and shoot the wafer 102 to be inspected, and it is convenient to stitch the images obtained by several image sensors 107.
[0046] In this embodiment, the image acquisition module 103 further includes a second wafer 106, the second wafer 106 is bonded to the second surface of the first wafer 105, and a plurality of signal processing circuits 111 are formed in the second wafer 106, and the plurality of signal processing circuits 111 are used to perform image processing on a plurality of images corresponding to the same shooting of the plurality of image sensors 107 to obtain a detection image corresponding to the entire surface of the wafer 102 to be detected. The image acquisition module 103 further includes a plastic sealing layer 110 for plastic sealing the first wafer 105 and the second wafer 106. In other embodiments, the plurality of signal processing circuits may also be formed on the first wafer 105.
[0047] The image processing includes image stitching, and the image stitching is used to stitch a number of images corresponding to the same shooting by the image sensor 107 to obtain a detection image corresponding to the entire surface of the wafer 102 to be inspected. In one embodiment, the image stitching includes an image preprocessing step, an image registration step, a change model establishment step, a same coordinate change step, and a fusion reconstruction step.
[0048] Specifically, the image preprocessing step includes basic operations of digital image processing (such as denoising, edge extraction, histogram processing, etc.), establishing a matching template for the image, and performing some transformation on the image (such as Fourier transform, wavelet transform, etc.).
[0049] The image registration step is to use a certain matching strategy or algorithm to find the corresponding position of the template or feature point in the image to be spliced in the reference image, and then determine the transformation relationship between the two images.
[0050] In one embodiment, the algorithm used for the image registration can adopt a frequency domain-based method (phase correlation method) and a time domain-based method. Among them, the time domain-based method can be specifically divided into a feature-based method and a region-based method. The feature-based method first finds the feature points (such as boundary points, inflection points, corner points) in the two images, and determines the correspondence between the feature points between the images, and then uses this correspondence to find the transformation relationship between the two images. This type of method does not directly use the grayscale information of the image, so it is not sensitive to light changes, but it relies heavily on the accuracy of the correspondence between the feature points. The region-based method uses a block in the overlapping area of one image as a template, and searches for the matching block that is most similar to this template in the other image. This algorithm has high accuracy.
[0051] The step of establishing the transformation model calculates the parameter values in the mathematical model according to the corresponding relationship between the templates or image features, thereby establishing a mathematical transformation model of the two images.
[0052] The unified coordinate transformation step transforms the image to be spliced into the coordinate system of the reference image according to the established mathematical transformation model to complete the unified coordinate transformation.
[0053] The fusion and reconstruction step fuses the overlapping areas of the images to be stitched to obtain a smooth and seamless panoramic image that is stitched and reconstructed.
[0054] In other embodiments, the graphic processing may also include one or more of image grayscale processing, image clarity processing, and synthesis processing of multiple defocused images.
[0055] In one embodiment, the image sensor 107 further includes a filter layer 112 and a plurality of micro lenses 113 located on the filter 112. The filter 112 may be a color filter to filter out unnecessary light. The micro lenses are used to converge the light reflected from the surface of the wafer 102 to be inspected into the corresponding image sensor 107 during measurement.
[0056] In one embodiment, the image sensor array 104 further includes an array of lens assemblies (not shown in the figure) corresponding to the plurality of image sensors 107 and disposed on the first surface of the first wafer 105. Specifically, a lens assembly may be disposed on each image sensor 107, or a lens assembly may be disposed on a plurality of adjacent (≥2) image sensors 107 (for example, a lens assembly may be disposed on two adjacent image sensors 107, or a lens assembly may be disposed on three adjacent image sensors 107, or a lens assembly may be disposed on four adjacent image sensors 107), so that the focal length and magnification of the image sensors 107 corresponding to the image sensor array 104 may be adjusted.
[0057] In one embodiment, when the aforementioned image sensor array 104 is used for shooting, all image sensors 107 in the image sensor array 104 are used for shooting at the same magnification, and when the signal processing circuit performs image processing, several images obtained by all image sensors 107 in the image sensor array 104 at the same magnification are stitched together to obtain a detection image corresponding to the entire surface of the wafer to be inspected.
[0058] In another embodiment, when performing inspection, several image sensors 107 in the image sensor array 104 can first be photographed using the same magnification (first magnification). When several first detection images are obtained, the signal processing circuit splices the several first detection images during image processing to obtain a detection image corresponding to the entire surface of the wafer to be inspected; then some image sensors 107 in the image sensor array 104 can first be photographed using a magnification (second magnification) to obtain a second detection image. Therefore, when performing defect detection, general defects can be detected at a smaller magnification, and at a larger magnification, a larger magnification and clearer image can be obtained for individual positions on the wafer to be detected to meet different detection requirements (such as feature size (CD) measurement and overlay error measurement). Specifically, the photoresist pattern used to measure the feature size (CD) can be detected and a specific feature size value can be obtained, and the photoresist pattern used to measure the overlay error can also be detected and a specific overlay error value can be obtained, or it can also be used to obtain a larger magnification and clearer image of a specific position for observation by process and equipment personnel. That is, the defect detection equipment of the present application can be used for rapid detection of general defects, and can also be used for rapid measurement of overlay errors and feature size values, thereby improving the detection efficiency of defects in the integrated circuit manufacturing process and the measurement efficiency of related parameters, and saving the time required for the entire integrated circuit process.
[0059] In a specific embodiment, before using the image sensor array 104 to capture and obtain the detection image corresponding to the entire surface of the wafer to be inspected, it is also necessary to align the image sensor array 104 (first wafer 105) with the wafer to be inspected 102 so that the position of the wafer to be inspected 102 can correspond to the position of the image sensor 107 on the image sensor array 104. Since the initial position of the wafer stage 101 is known, and the positions of the image sensors 107 on the image sensor array 104 are fixed, the position of the notch on the wafer 102 to be inspected is fixed. Therefore, before obtaining the detection image corresponding to the entire surface of the wafer to be inspected, the image sensor array 104 is used to obtain the image of the edge of the wafer to be inspected. The position of the notch on the wafer 102 to be inspected relative to the wafer stage 101 can be obtained through the corresponding image processing process. By adjusting the position of the wafer stage 101, the wafer 102 to be inspected on the wafer stage 101 can establish a positional connection with the image sensor 107 on the image sensor array 104, so that the position of the wafer 102 to be inspected can correspond to the position of the image sensor 107 on the image sensor array 104.
[0060] In one embodiment, the defect judgment module 108 includes a standard unit and a comparison unit. The standard unit stores a standard wafer image or a defect-free wafer image. The comparison unit is used to compare or match the detection image obtained by the image acquisition module with the standard wafer image or the defect-free wafer image, and then determine whether there are defects on the detection image and the location of the defects, thereby determining whether there are defects on the surface of the wafer to be inspected.
[0061] In another embodiment, the defect judgment module 108 may include an area division unit and a comparison unit, wherein the area division unit divides the detection image into a number of areas to be compared of the same size (when there are no defects, the graphics and patterns on the areas to be compared are the same or repeated), and the comparison unit is used to compare several adjacent areas to be compared to find out the locations of different points in the adjacent areas to be compared, which are the locations where defects exist.
[0062] In another embodiment, the defect judgment module 108 also performs learning during the defect judgment process to obtain the type of defect.
[0063] It should be noted that the defect judgment module 108 may use other methods or approaches to judge whether there are defects in the detection image.
[0064] The standard wafer image or the defect-free wafer image is an image obtained when there are no defects on the surface of the wafer to be inspected after a certain semiconductor process is performed.
[0065] In one embodiment, the defect judgment module 108 may also be integrated into the second wafer 106 through a semiconductor integration manufacturing process. In other embodiments, the defect judgment module 108 may be a separate chip or processor.
[0066] In one embodiment, please refer to Figure 3, the wafer defect detection device also includes an illumination light source 109, and the illumination light source 109 is used to illuminate the surface of the wafer 102 to be detected during the detection. In this embodiment, the illumination unit 109 is arranged on one side of the wafer stage 101 and the first wafer 105. When performing defect detection, the wafer stage 101 can be deflected so that the wafer 102 to be detected is in a tilted state, and at the same time, the illumination unit 109 illuminates the surface of the wafer 102 to be detected, and the image sensor array 104 in the image acquisition module 103 obtains a detection image with a dark field corresponding to the entire surface of the wafer to be detected by taking a single shot. In one embodiment, the illumination unit 109 can rotate around the wafer stage 101 and the first wafer 105. In another embodiment, the number of the illumination units 109 can be multiple (greater than or equal to 2), and the multiple illumination units 109 are evenly (at equal angles) distributed around the wafer stage 101 and the first wafer 105.
[0067] In other embodiments, the illumination light source can be directly set on one side of the image sensor 107 in the image sensor array 104. When shooting, the front side of the wafer 102 to be inspected is illuminated. The image sensor array 104 obtains a detection image with a bright field corresponding to the entire surface of the wafer to be inspected by shooting.
[0068] The wafer defect detection device of the present invention, since the image acquisition module 103 includes a first wafer 105, and an image sensor array 104 is formed in the first wafer 105, the positions between adjacent image sensors 107 in the image sensor array 104 are fixed, so when the image acquisition module 103 acquires the detection image of the entire surface of the wafer 102 to be detected through the image sensor array 104, the front and back left and right positions and the height positions between adjacent image sensors 107 do not need to be calibrated, which simplifies the operation process of the wafer defect detection device, and does not need to set up an additional auxiliary calibration structure, thereby simplifying the structure of the image acquisition module 103 and reducing the volume of the image acquisition module 103. In addition, since the entire wafer of the first wafer 105 is used to form the image sensor array 104, there is no need to set a cutting path like the existing integrated circuit manufacturing wafer, and the wafer needs to be divided into individual chips later, so there can be more space on the first wafer 105 for forming the image sensor 107 and the functional circuit connected to the image sensor 107.
[0069] Furthermore, the image acquisition module 103 obtains a detection image corresponding to the entire surface of the wafer 102 to be inspected by taking a shot through the image sensor array 104. Therefore, the wafer stage 101 does not need to perform scanning movement (move in the horizontal direction) when acquiring the detection image, and the wafer stage 101 described in the present application only needs to move in a small range or distance in the vertical direction and / or in the horizontal direction when aligning with the image sensor array 104. Therefore, the volume of the wafer stage 101 and the corresponding driving unit or driving device can be smaller, and the volume occupied by the image sensor array 104 is also smaller, thereby reducing the volume occupied by the entire wafer defect detection equipment and reducing energy consumption (small driving volume reduces energy consumption). In the present application, the detection image corresponding to the entire surface of the wafer 102 to be inspected is obtained by taking a picture once by the image acquisition module 103 through the image sensor array 104. The defect judgment module 108 judges whether there are defects on the surface of the wafer 102 to be inspected based on the detection image obtained by the image acquisition module 103. When performing defect detection, the time for acquiring the detection image is greatly reduced (one-time transient imaging), thereby improving the efficiency of defect detection and reducing costs (the cost of the image sensor array is greatly reduced compared to the optical scanning magnifier).
[0070] The above is only a preferred embodiment of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A wafer defect detection device, characterized in that: include: Wafer stage, used to fix the wafer to be inspected; An image acquisition module, wherein the image acquisition module comprises a first wafer, wherein the first wafer comprises a first surface and a second surface opposite to each other, wherein an image sensor array is formed on the first surface, wherein the image sensor array is formed in the first wafer by a semiconductor integrated manufacturing process, wherein the entire first wafer is used to form the image sensor array, wherein the image sensor array comprises a plurality of image sensors, wherein the plurality of image sensors are arranged in rows and columns in the first wafer, and wherein the image acquisition module obtains a detection image corresponding to the entire surface of the wafer to be detected by taking a single shot through the image sensor array; The image acquisition module further includes a second wafer, the second wafer is bonded to the second surface of the first wafer, and a plurality of signal processing circuits are formed in the second wafer, and the plurality of signal processing circuits are used to perform image processing on a plurality of images correspondingly obtained in the same shooting of a plurality of image sensors to obtain a detection image corresponding to the entire surface of the wafer to be detected; A defect judgment module is used to judge whether there are defects on the surface of the wafer to be inspected according to the inspection image obtained by the image acquisition module.
2. The wafer defect detection device according to claim 1, characterized in that: The image processing includes image splicing, and the image splicing includes an image preprocessing step, an image registration step, a change model establishment step, a same coordinate change step, and a fusion reconstruction step.
3. The wafer defect detection device according to claim 2, characterized in that: The image processing also includes one or more of image grayscale processing, image clarity processing, and synthesis processing of multiple defocused images.
4. The wafer defect detection device according to claim 1, characterized in that: The image sensor further includes a color filter layer and a plurality of micro lenses located on the color filter.
5. The wafer defect detection device according to claim 1 or 4, characterized in that: The image sensor array further includes a lens assembly array corresponding to a plurality of image sensors and disposed on the first surface of the first wafer.
6. The wafer defect detection device according to claim 1, characterized in that: The size of the image sensor array corresponds to the size of the wafer to be inspected.
7. The wafer defect detection device according to claim 1, characterized in that: The defect judgment module includes a standard unit and a comparison unit, wherein the standard unit stores a standard wafer image or a defect-free wafer image, and the comparison unit is used to compare or match the detection image obtained by the image acquisition module with the standard wafer image or the defect-free wafer image, and then judge whether there is a defect on the detection image and the location of the defect; The defect judgment module also performs learning during the defect judgment process to obtain the types of defects.
8. The wafer defect detection device according to claim 1, characterized in that: The wafer carrier can rotate, move vertically, move horizontally, and deflect.
