WO2023216643A1 - 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备 - Google Patents

一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备 Download PDF

Info

Publication number
WO2023216643A1
WO2023216643A1 PCT/CN2023/071167 CN2023071167W WO2023216643A1 WO 2023216643 A1 WO2023216643 A1 WO 2023216643A1 CN 2023071167 W CN2023071167 W CN 2023071167W WO 2023216643 A1 WO2023216643 A1 WO 2023216643A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
wafer
outline
parameters
cleaning
Prior art date
Application number
PCT/CN2023/071167
Other languages
English (en)
French (fr)
Inventor
盛思杰
Original Assignee
杭州众硅电子科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 杭州众硅电子科技有限公司 filed Critical 杭州众硅电子科技有限公司
Publication of WO2023216643A1 publication Critical patent/WO2023216643A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/001—Industrial image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/155—Segmentation; Edge detection involving morphological operators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/168—Segmentation; Edge detection involving transform domain methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/67005—Apparatus not specifically provided for elsewhere
    • H01L21/67011—Apparatus for manufacture or treatment
    • H01L21/67017—Apparatus for fluid treatment
    • H01L21/67028—Apparatus for fluid treatment for cleaning followed by drying, rinsing, stripping, blasting or the like
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/67005—Apparatus not specifically provided for elsewhere
    • H01L21/67242—Apparatus for monitoring, sorting or marking
    • H01L21/67259—Position monitoring, e.g. misposition detection or presence detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • G06T2207/20061—Hough transform
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30148—Semiconductor; IC; Wafer

Definitions

  • the invention belongs to the technical field of semiconductor processing equipment, and in particular relates to a wafer cleaning and drying module state detection method and device, and chemical mechanical planarization equipment having the above state detection device.
  • the wafer wet cleaning process effectively uses chemical solutions to clean impurities remaining on the wafer without damaging the surface characteristics and electrical properties of the wafer.
  • the wafer wet cleaning process is widely used in chemical mechanical planarization equipment, cleaning equipment and electroplating equipment.
  • cleaning and drying devices in the wafer wet cleaning process, such as devices that use spin rinse drying (SRD) to dry wafers, or devices that use IPA to clean and dry wafers.
  • SRD spin rinse drying
  • IPA to clean and dry wafers.
  • the wafer must be placed parallel to the bottom surface of the clamping jaw of the device for cleaning and drying the wafer. If the wafer cannot be placed parallel to the bottom surface of the clamping jaw as required, the clamping jaw cannot be fixed. Possibility of holding the wafer.
  • the wafer If the wafer is not well fixed, it will break away from the station during the drying process and damage the wafer and the device. Therefore, before working on the cleaning and drying module, it is necessary to check whether the wafer is successfully placed on the plane required by the cleaning and drying module and placed horizontally.
  • the current technology uses two sets of transmission laser sensors for wafer level detection and a laser ranging sensor for wafer presence detection. Using this solution will result in long debugging time, high cost, and complex mechanical structure. The detection feedback time is too long, the device is complex and unsightly, and other problems. In addition, current technology cannot detect abnormal conditions of the wafer clamp, which may cause the wafer to tilt or break when placing or removing the wafer under working conditions.
  • the present invention provides a wafer cleaning and drying module status detection method and device that can accurately monitor whether the wafer is tilted in real time by comparing the screen capture with the standard image outline, and a device with the above device flattening equipment.
  • a wafer cleaning and drying module status detection method which includes the following steps:
  • the parameters of the standard image outline are the center and radius of the circle;
  • the parameters of the standard image outline are minimum circumscribed polygon outline parameters.
  • the obtained standard image outline is a circular outline
  • the acquired standard image outline is a non-circular outline.
  • the circle contour detection uses the Hough transform circle detection method.
  • the minimum circumscribed polygon outline parameter is the number of sides or the length of each side or the included angle of each side.
  • the step of determining whether a wafer exists it is determined whether a wafer outline exists, and if a wafer outline exists, it is determined that the wafer exists.
  • the standard image preprocessing and/or the test image preprocessing method is smoothing processing of the image signal.
  • the smoothing processing of the image signal adopts Gaussian filtering method.
  • the standard image preprocessing and/or the test image preprocessing method is first grayscale processing, then binarization processing, and finally expansion processing.
  • the expansion processing The scattered contours after binary processing are connected together to form the complete contour of the wafer.
  • the wafer is in a stationary state or a rotating state.
  • Preprocess the target image of the marking position to obtain the target image outline of the marking position under the normal state of the wafer clamping device;
  • the parameters of the marked target image outline are the number of sides and the area.
  • the step of comparing the parameters of the image contour of the mark position to be tested with the parameters of the mark position target image first obtain the image contour with the same color as the mark position in the image to be measured, and select the image contour with the same number of edges as the mark position. image contour, and then determine whether the area of each image contour is within the set range. Select the image contour to be measured at the marked position through the above steps, and compare it with the area of the target image at the marked position.
  • Preprocess the target image obtain the target image outline of the mark position under the normal state of the wafer clamping device, and count and number the target image outlines of multiple mark positions;
  • the step of preprocessing the target image of the marked position and/or the preprocessing of the image to be measured of the marked position includes the step of extracting the set color and eliminating interfering colors.
  • the step of preprocessing the target image of the marked position and/or the preprocessing of the image to be measured of the marked position includes the step of eliminating interfering shapes in the image.
  • steps of preprocessing the target image and/or preprocessing the image to be tested for the mark bit include the following steps:
  • edge detection to delineate the edges of the mark bit image
  • Dilation processing is used to connect the relatively scattered contours of the image after the above processing to form a complete contour of the mark position.
  • the step of setting the color in the extracted image includes converting the original RGB image into HSV space, extracting the HSV value of the set color to make a mask, and performing a mask operation on the original RGB image to obtain a separated image.
  • the smoothing processing step of the image signal adopts Gaussian filtering method.
  • steps of placing the wafer on the wafer clamping device and taking out the wafer from the wafer clamping device are performed.
  • the invention also discloses a wafer cleaning and drying module status detection device, which includes:
  • the image acquisition unit is located above or on the side of the plane where the wafer is located, and is used to acquire images of the wafer and/or wafer clamping device in the cleaning and drying module during the wafer wet cleaning process, including the normal placement of the wafer. Standard image and/or target image of the marked position in the normal state of the wafer clamping device, and image to be tested;
  • the image processing unit is used to preprocess the standard image and/or the target image of the marked position and the image to be tested obtained by the image acquisition unit, and obtain the standard image outline and/or the target image outline of the marked position, and the outline of the image to be measured. ;
  • the judgment unit compares the image contour to be measured with the standard image contour and/or the target image contour of the mark position to determine whether the difference is within a set range, thereby determining the opening and closing state of the wafer clamping device.
  • Wafer holding jaws used to support, clamp and secure wafers
  • Wafer fixing jaw base used to support and fix the wafer
  • the base drive motor is used to drive the wafer to rotate
  • Cleaning liquid transfer pipeline is used to transfer cleaning liquid or other required liquids to the wafer.
  • the image acquisition unit is an image acquisition device equipped with a CCD sensor, or an image acquisition device equipped with a CMOS sensor, or an image acquisition device and a reflector equipped with a CCD sensor, or an image equipped with a CMOS sensor. Recording device and reflector, or photoelectric conversion device.
  • the invention also discloses a planarization equipment, which includes a grinding module, a cleaning module, an equipment front-end module, and a wafer transport mechanism for transporting wafers.
  • the cleaning module has the above-mentioned wafer cleaning and drying module. Group.
  • the beneficial effects of the present invention are: 1) The image acquisition unit can record the cleaning and drying conditions in the wafer wet cleaning process in real time, and the image processing unit can process the read images in real time, thus realizing the wafer wet cleaning process.
  • Figure 1 is a front view of the detection device of the present invention.
  • Figure 2 is a top view of the detection device of the present invention.
  • Figure 3 is a front view of the detection device of the present invention, with a reflector.
  • Figure 4 is a front view of the detection device of the present invention, in which the image processing unit and the judgment unit are integrated into one body.
  • Figure 5 is a schematic structural diagram of the image acquisition unit of the present invention located above the center of the wafer.
  • Figure 6 is a schematic diagram of the detection method of the present invention when acquiring a circular outline. At this time, the image acquisition unit is located directly above the wafer.
  • Figure 7 is a schematic diagram of the detection method of the present invention when acquiring a non-circular contour. At this time, the image acquisition unit is located directly above the wafer.
  • Figure 8 is a schematic diagram of the detection method of the present invention when acquiring a circular outline. At this time, the image acquisition unit is located obliquely above the wafer.
  • Figure 9 is a schematic diagram of the detection method of the present invention when acquiring a non-circular contour. At this time, the image acquisition unit is located obliquely above the wafer.
  • Figure 10 is a schematic diagram of the marking position of the wafer fixing clamp of the present invention in a normal state.
  • Figure 11 is a schematic diagram of the mark position when the wafer fixing clamp of the present invention is in an abnormal state.
  • Figure 12 is a schematic flow chart of wafer status detection according to the present invention.
  • FIG. 13 is a schematic flowchart of status detection of the wafer clamping device of the present invention.
  • Figure 14 is a schematic diagram of the application flow of the present invention.
  • a wafer cleaning and drying module status detection device includes an image acquisition unit 2, an image processing unit 4, and a judgment unit 5.
  • the image acquisition unit 2 is arranged above or on the side of the plane where the wafer 1 is located, and is used to acquire images of the wafer 1 and/or the wafer clamping device 3 in the cleaning and drying module during the wafer wet cleaning process. It includes a standard image of the wafer 1 placed normally and/or a target image of the mark position 34 of the wafer holding device 3 in a normal state, and an image to be tested.
  • the image acquisition unit 2 is on the upper plane of the wafer 1; when the wafer 1 is placed vertically, the image acquisition unit 2 is on the side plane of the wafer 1.
  • the image acquisition unit 2 can be directly above the wafer 1. At this time, the projection of the image acquisition unit 2 on the wafer 1 falls at the center of the wafer 1, as shown in Figure 5; it can also be located on the wafer 1. diagonally above, as shown in Figure 1.
  • the image acquisition unit 2 is an image acquisition device equipped with a CCD sensor; or an image acquisition device equipped with a CMOS sensor; or an image acquisition device equipped with a CCD sensor and a reflector 21.
  • the image acquisition device Obtaining the image reflected by the reflector 21 can reduce the interference of ambient light on the image acquisition device, and can reduce the wafer image deformation caused by the oblique position of the image acquisition device, as shown in Figure 3; alternatively, it is equipped with a CMOS sensor An image acquisition device and a reflector 21; or, a photoelectric conversion device.
  • the image processing unit 4 is used to preprocess the standard image and/or the target image of the marked position and the image to be tested acquired by the image acquisition unit 2, to obtain the standard image outline and/or the target image outline of the marked position, and the image to be measured. Image outline.
  • the judgment unit 5 compares the image contour to be measured with the standard image contour and/or the target image contour of the mark position to judge whether the difference is within the set range, thereby judging the opening and closing status of the wafer clamping device. If the difference is If the value exceeds the set range, the status is judged to be abnormal.
  • the above-mentioned image processing unit 4 and judgment unit 5 can also be integrated into one body, as shown in Figure 4 .
  • a light source may be provided that projects light toward the wafer and/or the wafer holding device.
  • the wafer clamping device 3 includes a wafer fixing jaw 31 , a wafer fixing jaw base 32 , and a base driving motor 33 .
  • the number of wafer fixing jaws 31 is three, which are used to support, clamp and fix the wafer 1; the wafer fixing jaw base 32 is used to support and fix the wafer 1; the base driving motor 33 is used to drive wafer 1 to rotate.
  • a cleaning liquid transmission pipeline 6 is also provided above the wafer clamping device 3 , which is used to transmit cleaning liquid or other required liquids to the wafer 1 . That is, the cleaning liquid transmission pipeline 6 transports cleaning liquid or other liquids to the surface of the wafer 1 .
  • the invention also discloses a chemical mechanical planarization equipment, which includes a grinding module, a cleaning module, an equipment front-end module, and a wafer transport mechanism for transporting wafers.
  • the cleaning module has the above-mentioned wafer wet method. Cleaning and drying module in cleaning process.
  • a method for detecting the status of a wafer cleaning and drying module including the following steps:
  • the abnormal status detection of the cleaning and drying module in the wafer wet cleaning process includes the detection of the presence or absence of the wafer and the detection of the wafer level. If there is no wafer or there is a wafer but it is tilted, it is determined that the cleaning and drying module is in an abnormal state during the wafer wet cleaning process.
  • an image contour detection method can be used, that is, by determining whether the wafer contour exists in the image contour to be measured. If there is no wafer contour, it is directly determined that wafer 1 does not exist. In other words, the step of determining whether the wafer 1 exists is to determine whether the wafer outline exists. If the wafer outline exists, it is determined that the wafer 1 exists.
  • existing judgment methods can also be used without limitation.
  • the above detection process is divided into two situations.
  • the first situation when the image acquisition unit 2 acquires a standard image of the normally placed wafer 1 directly above the wafer 1, the acquired standard image outline is a circular outline.
  • the second case when the image acquisition unit 2 acquires a standard image of the wafer 1 placed obliquely above the wafer 1, the acquired standard image outline is a non-circular outline.
  • the parameters of the above-mentioned standard image contour are the circle center and radius.
  • the center is located at coordinates (x1, y1) and its radius is R1.
  • the center is located at coordinates (x2, y2), and its radius value does not appear R1.
  • the circle outline is detected using the Hough transform circle detection method.
  • the standard image preprocessing and/or image preprocessing method to be tested is the smoothing of the image signal, specifically using the Gaussian filtering method.
  • the above-mentioned parameters of the standard image contour are the minimum circumscribed polygon contour parameters.
  • the minimum circumscribed polygon outline parameter is the number of sides or the length of each side or the angle between each side.
  • the standard image preprocessing and/or image preprocessing method to be tested is first grayscale processing, then binarization processing, and finally expansion processing. This expansion processing connects the relatively scattered contours after binarization processing to form a crystal. Complete outline of circle.
  • H and W are the side length parameters of the minimum circumscribed polygon of the wafer outline (it is a rectangle here, so the parameter description can be the length and width of the rectangle), and a is the minimum circumscribed polygon of the wafer outline. The angle between adjacent sides of a circumscribed polygon.
  • the abnormal state detection of the cleaning and drying module in the wafer wet cleaning process also includes the abnormal state detection of the wafer clamping device 3, that is, when the wafer 1 is placed Before the wafer clamping device 3 and before the step of removing the wafer 1 from the wafer clamping device 3, it is necessary to detect whether the wafer clamping device 3 is in a normal horizontal or vertical state.
  • the entire wafer clamping device 3 When the wafer clamping device 3 is placed normally, the entire wafer clamping device 3 is in a vertical state, and its upper plane is in a horizontal state.
  • the base drive motor 33 drives the wafer fixing jaw base 22, the wafer fixing jaw 31 and the wafer 1 to rotate, and the wafer clamping device 3 moves toward the wafer under the action of centrifugal force. The side is deflected at a certain angle to achieve the function of clamping and fixing the wafer.
  • the entire wafer clamping device 3 At this time, the entire wafer clamping device 3 is in an inclined state, and its upper plane is in an inclined state. If the wafer clamping device 3 is still tilted before the wafer is placed or before the wafer is taken out, it is determined that the wafer clamping device 3 is abnormal.
  • the deformation of the mark bit 34 in the screen of the image acquisition unit 2 can reflect the state of the wafer clamping device 3, which can be used before or after placing the wafer.
  • the state of the wafer clamping device 3 is detected before the wafer is taken out.
  • the marking position 34 is provided on the wafer fixing clamp 31 and is triangular in shape.
  • the color can be any color that is significantly different from the background color, as shown in FIG. 10 .
  • Preprocess the target image to obtain the target image outline of the mark position in the normal state of the wafer clamping device 3;
  • the parameters of the marked image outline are the number of sides and the area;
  • the parameters of the contour of the image to be measured at the marked position are the number of sides and the area;
  • the steps of comparing the parameters of the image contour of the mark position to be tested with the parameters of the target image of the mark position are as follows.
  • the image contour with the same color as the mark position 34 in the image to be tested is first obtained;
  • the second step select the image outline with the same number of sides as the mark position 34, in this embodiment, select the triangular image outline;
  • the third step is to determine whether the area of each image outline left after filtering through the above two steps is within the set range
  • the fourth step after the above three steps, select the outline of the image to be measured at the marked position, and compare it with the area of the target image at the marked position. If the areas are the same, it means that the wafer clamping device 3 is in a normal state. As shown in FIG. 11 , when the wafer clamping device 3 is in an abnormal state, although the outline of the image of the mark position to be measured is also a triangle, its area is different from the area of the triangle in FIG. 10 .
  • the obtained image contours contain image contours in multiple areas.
  • a batch of image contours are filtered out based on color differences, then a batch of image contours are filtered out based on shape differences, and then a batch of image contours are filtered out based on whether the area is within a certain range.
  • the contours of the image to be measured at the marked position are finally screened out, and compared with the area of the target image at the marked position.
  • the above steps are performed in the step of comparing the parameters of the contour of the image to be measured at the marked position with the parameters of the target image at the marked position.
  • the steps of preprocessing the target image and/or preprocessing the image to be measured for the mark bit interference of colors similar to the mark bit in the image may be directly eliminated.
  • the color comparison step can be omitted and the second to fourth steps can be performed directly.
  • the edge number comparison step can be omitted.
  • steps of preprocessing the target image and/or preprocessing the image to be measured for the marked bits especially the steps of eliminating interference in the image with colors similar to the marked bits, include:
  • Extract the set color from the image more specifically, this step converts the original RGB image into HSV space, extracts the set color HSV value to make a mask, and performs a mask operation on the original RGB image to obtain a separated image;
  • the image signal is smoothed; more specifically, it uses the Gaussian filtering method
  • edge detection to delineate the edges of the marker bit image
  • Expansion processing is used to connect the scattered contours of the image after the above processing to form a complete contour of the marked position.
  • each wafer fixing clamp 31 is provided with a marking bit 34 of a set color and shape, that is, the wafer clamping device 3 is provided with a plurality of marking bits of a set color and shape.
  • the detection steps at this time are:
  • Preprocess the target image obtain the target image outline of the mark position in the normal state of the wafer clamping device 3, and count and number the target image outlines of the multiple mark positions;
  • the parameters of the target image outline of the marked position corresponding to the count number are the number of sides and the area;
  • the opening and closing status of the wafer clamping device is determined. If there is an abnormality in one of them, it is determined that the wafer clamping device is abnormal. At this time, there is no need to perform all comparisons.
  • the steps of the state detection method of the cleaning and drying module in the wafer wet cleaning process of the present invention are described in detail below, which include detection of the presence or absence of the wafer, detection of the wafer level, and detection of the wafer clamping device. detection.
  • the image acquisition unit 2 acquires an image of the position of the wafer fixing jaw 31 in the normal state, which is recorded as M(i-1), and performs image processing on the image to obtain the target image outline of the mark position of the wafer clamping device 3 in the normal state.
  • MK(i-1) contains the number of mark bit edges o(i-1). , number p0 and area q(i-1) parameters.
  • the image acquisition unit 2 acquires the image of the wafer 1 in the normal state, which is recorded as N(i-1).
  • the image is processed to obtain the standard image outline of the wafer in the normal state.
  • the parameters for acquiring the standard image outline of the wafer are recorded as NK(i-1), if the image acquisition unit 2 is placed directly above the plane of wafer 1, NK(i-1) includes the position (x0, y0) and radius R0 of the center of wafer 1 in the image under normal conditions, If the image acquisition unit 2 is placed obliquely above the plane where the wafer 1 is located, NK(i-1) includes the number of sides a0 of the minimum circumscribed polygonal outline of the wafer outline under normal conditions, the length of each side b(i-1) and each The angle between the sides is c(i-1).
  • the image acquisition unit 2 acquires an image of the position of the wafer fixing clamp 31 under working conditions, denoted as M(i), performs image processing on the image, and acquires the mark position of the wafer clamping device 3 to be tested.
  • the number p1 and area q(i) parameters are compared with MK(i) and MK(i-1). If the difference exceeds the set threshold, the judgment unit 5 determines that the module wafer fixing clamp 31 Abnormal, the host computer judgment unit 5 will perform corresponding operations. If the difference is within the set threshold, it will be judged that the module wafer clamping device 3 is normal, and the work of placing the wafer 1 will begin.
  • the image acquisition unit 2 After placing the wafer 1, the image acquisition unit 2 acquires the image and determines whether the wafer exists by the image processing method. If the wafer does not exist, the determination unit 5 determines that the wafer in the module is abnormal, and the host computer determination unit 5 Make corresponding operations. If the wafer exists, record the image as N(i). Process the image to obtain the image outline of the wafer. The parameters for obtaining the wafer image outline are recorded as NK(i). If the image The acquisition unit 2 is placed directly above the plane where the wafer 1 is located. NK(i) includes the position (x1, y1) of the wafer center in the image and the radius R1.
  • NK(i) includes the number of sides a1 of the minimum circumscribed polygon contour of the wafer outline, the length of each side b(i) and the angle between each side c(i). Compare NK(i) with NK(i-1). If the difference exceeds the set threshold, the judgment unit 5 will judge that the wafer in the module is abnormal, and the host computer judgment unit 5 will perform corresponding operations. If the difference is within the set threshold, it is judged that the wafer in the module is normal, and the cleaning liquid transmission pipeline 6 transmits the cleaning liquid or liquid to the wafer 1 to start the wafer cleaning and drying work.
  • the image acquisition unit 2 acquires the image and determines whether there is a wafer through the image processing method. If the wafer does not exist, the determination unit 5 determines that the wafer in the module is abnormal, and the host computer determines whether the wafer exists. Unit 5 performs corresponding operations. If the wafer exists, the image is recorded as N(i+1), the image is processed to obtain the image outline of the wafer, and the parameters for obtaining the wafer image outline are recorded as NK(i +1), if the image acquisition unit 2 is placed directly above the plane where the wafer is located, NK(i+1) includes the position (x2, y2) and radius R2 of the wafer center in the image.
  • NK(i+1) includes the number of sides a2 of the minimum circumscribed polygon outline of the wafer outline, the length of each side b(i+1) and the angle between each side c(i+1). Compare NK(i+1) with NK(i-1). If the difference exceeds the set threshold, the judgment unit 5 will judge that the wafer in the module is abnormal, and the host computer judgment unit 5 will make a corresponding response. operation, if the difference is within the set threshold, it is judged that the wafer in the module is normal, and detection of the wafer clamping claw status begins.
  • the state of the wafer clamping device 3 is detected.
  • the image acquisition unit 2 acquires an image of the position of the wafer fixing jaw 31 under working conditions, which is marked as M(i+1).
  • the image is processed to obtain the wafer clamping device 3.
  • the image outline of the mark bit to be tested is counted and numbered and the parameters of the mark bit image outline corresponding to the count number are obtained, recorded as MK(i+1), MK(i+1) contains the mark Bit edge number o(i+1), number p2 and area q(i+1) parameters, compare MK(i+1) and MK(i-1), if the difference exceeds the set threshold, Then the judgment unit 5 judges that the module wafer clamping device 3 is abnormal, and the host computer judgment unit 5 performs corresponding operations. If the difference is within the set threshold, it is judged that the module wafer fixing clamp is 31 Normal, start taking out the wafer.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Hardware Design (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Manufacturing & Machinery (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Geometry (AREA)
  • Quality & Reliability (AREA)
  • Testing Or Measuring Of Semiconductors Or The Like (AREA)
  • Cleaning Or Drying Semiconductors (AREA)
  • Image Analysis (AREA)

Abstract

本发明公开了一种晶圆清洁干燥模组状态检测方法，包括以下步骤：获取晶圆正常放置的标准图像；对标准图像预处理，获取晶圆正常放置的标准图像轮廓；获取标准图像轮廓的参数；获取待测图像；对待测图像预处理，判断是否存在晶圆，若存在晶圆则获取待测图像轮廓；获取待测图像轮廓的参数；将待测图像轮廓的参数与标准图像轮廓的参数作比较，判断其差值是否在设定范围内，若差值超过设定阈值，则判断为晶圆倾斜。本发明还公开了一种晶圆清洁干燥模组状态检测装置。本发明还公开了一种平坦化设备。本发明实现了晶圆有无和晶圆是否倾斜的实时检测，检测准确率高，检测时间短。

Description

一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备 技术领域
本发明属于半导体加工设备技术领域，尤其是涉及一种晶圆清洁干燥模组状态检测方法、装置，及具有上述状态检测装置的化学机械平坦化设备。
背景技术
晶圆湿法清洗工艺是在不破坏晶圆表面特性及电特性的前提下，有效的使用化学溶液清洗残留在晶圆上的杂质。晶圆湿法清洗工艺广泛运用在化学机械平坦化设备、清洗设备及电镀设备中。现有晶圆湿法清洗工艺中的清洁干燥装置，例如使用旋转甩干方式干燥晶圆的装置(Spin Rinse Dry，SRD)、或以IPA清洁干燥晶圆的装置等。无论采用哪种清洁干燥方式，晶圆放入后都需与清洁干燥晶圆的装置的夹爪底面平行，如果未能成功按要求将晶圆放至与夹爪底面平行，夹爪存在不能固定住晶圆可能性，未很好固定的晶圆在干燥过程中将脱离工位而损坏晶圆和装置。因此，在清洁干燥模组工作前需要检测晶圆是否成功放置在清洁干燥模组要求的平面上并与水平放置。
当前技术为使用两组透过式激光传感器来进行晶圆水平检测，使用一个激光测距传感器来进行晶圆有无检测，使用此方案会产生调试时间过长，成本过高，机械结构复杂，检测反馈时间过长，装置复杂不美观等问题。此外，当前技术并不能检测晶圆夹爪的异常状态，会导致工况下放置或取出晶圆时晶圆的倾斜或晶圆的破碎等。
发明内容
为了克服现有技术的不足，本发明提供一种通过画面录取后与标准图像轮廓进行比对，可以实时准确监测晶圆是否倾斜的晶圆清洁干燥模组状态检测方法、装置及带有上述装置的平坦化设备。
本发明解决其技术问题所采用的技术方案是：一种晶圆清洁干燥模组状态检测方法，包括以下步骤：
获取晶圆正常放置的标准图像；
对标准图像预处理，获取晶圆正常放置的标准图像轮廓；
获取标准图像轮廓的参数；
获取待测图像；
对待测图像预处理，判断是否存在晶圆，若存在晶圆则获取待测图像轮廓；
获取待测图像轮廓的参数；
将待测图像轮廓的参数与标准图像轮廓的参数作比较，判断其差值是否在设定范围 内，若差值超过设定阈值，则判断为晶圆倾斜。
进一步的，当获取的标准图像轮廓为圆轮廓时，所述标准图像轮廓的参数为圆心和半径；
当获取的标准图像轮廓为非圆轮廓时，所述标准图像轮廓的参数为最小外切多边形轮廓参数。
进一步的，当在晶圆正上方获取晶圆正常放置的标准图像时，获取的标准图像轮廓为圆轮廓；
当在晶圆斜上方获取晶圆正常放置的标准图像时，获取的标准图像轮廓为非圆轮廓。
进一步的，圆轮廓检测采用霍夫变换圆检测方法。
进一步的，所述最小外切多边形轮廓参数为边数或各边长或各边夹角。
进一步的，所述判断是否存在晶圆步骤中，为判断是否存在晶圆轮廓，若存在晶圆轮廓则判定晶圆存在。
进一步的，当获取的标准图像轮廓为圆轮廓时，所述标准图像预处理和/或待测图像预处理方法为图像信号的平滑处理。
进一步的，所述图像信号的平滑处理采用高斯滤波方法。
进一步的，当获取的标准图像轮廓为非圆轮廓时，所述标准图像预处理和/或待测图像预处理方法为先灰度化处理，后二值化处理，最终膨胀处理，该膨胀处理将二值化处理后比较分散的轮廓连接到一起，以形成晶圆的完整轮廓。
进一步的，所述晶圆为静止状态或旋转状态。
进一步的，还包括以下步骤：
在晶圆夹持装置设置具有设定颜色和形状的标记位；
获取晶圆夹持装置正常状态下标记位的目标图像；
对标记位的目标图像预处理，获取晶圆夹持装置正常状态下的标记位目标图像轮廓；
获取标记位目标图像轮廓的参数；
获取晶圆夹持装置标记位的待测图像；
对标记位的待测图像预处理，得到标记位待测图像轮廓；
获取标记位待测图像轮廓的参数；
将标记位待测图像轮廓的参数与标记位目标图像的参数作比较，判断其差值是否在 设定范围内，从而判断晶圆夹持装置的开合状态。
进一步的，所述标记位目标图像轮廓的参数为边数和面积。
进一步的，所述将标记位待测图像轮廓的参数与标记位目标图像的参数作比较步骤中，先获取待测图像中与标记位的颜色相同的图像轮廓，选取与标记位有相同边数的图像轮廓，再判断各图像轮廓的面积是否在设定范围内，通过上述步骤选取标记位待测图像轮廓，将其与标记位目标图像的面积作比较。
进一步的，包括以下步骤，
在晶圆夹持装置设置具有设定颜色和形状的多个标记位；
获取晶圆夹持装置正常状态下多个标记位的目标图像；
对目标图像预处理，获取晶圆夹持装置正常状态下的标记位目标图像轮廓，并对多个标记位目标图像轮廓进行计数编号；
获取对应计数编号的标记位目标图像轮廓的参数；
获取晶圆夹持装置标记位的待测图像；
对标记位的待测图像预处理，得到标记位待测图像轮廓；
获取标记位待测图像轮廓的参数，该参数包括边数和面积；
对判断为边数相同且面积在设定范围内的标记位待测图像轮廓进行编号；
将上述编号的标记位待测图像轮廓与进行计数编号的标记位目标图像轮廓的面积进行分别比较；
若其中有一个异常即判断为晶圆夹持装置异常。
进一步的，所述对标记位的目标图像预处理和/或对标记位的待测图像预处理步骤中，包括提取设定颜色，排除干扰颜色步骤。
进一步的，所述对标记位的目标图像预处理和/或对标记位的待测图像预处理步骤中，包括排除图像中干扰形状的步骤。
进一步的，所述对目标图像预处理和/或对标记位的待测图像预处理步骤中，包括以下步骤，
提取图像中设定颜色；
灰度化；
图像信号的平滑处理；
边缘检测，以描绘标记位图像的边缘；
膨胀处理，用于将图像经上述处理后比较分散的轮廓连接到一起，以形成标记位的 完整轮廓。
进一步的，所述提取图像中设定颜色步骤包括，将原始RGB图像转化为HSV空间，提取设定颜色HSV值制作掩膜，对原始RGB图像进行掩膜操作得到分离图像。
进一步的，所述图像信号的平滑处理步骤采用高斯滤波方法。
进一步的，在晶圆放置于晶圆夹持装置和晶圆自晶圆夹持装置取出步骤之前进行。
本发明还公开了一种晶圆清洁干燥模组状态检测装置，包括：
图像获取单元，设于晶圆所在平面上方或侧方，用于获取晶圆湿法清洗工艺中清洁干燥模组中晶圆和/或晶圆夹持装置的图像，其包括晶圆正常放置的标准图像和/或晶圆夹持装置正常状态下标记位的目标图像，及待测图像；
图像处理单元，用于将图像获取单元获取的标准图像和/或标记位的目标图像，及待测图像进行预处理，得到标准图像轮廓和/或标记位的目标图像轮廓，及待测图像轮廓；
判断单元，将待测图像轮廓与标准图像轮廓和/或标记位的目标图像轮廓作比较，以判断差值是否在设定范围内，从而判断晶圆夹持装置的开合状态。
进一步的，还包括：
晶圆固定夹爪，用于支撑、夹持和固定晶圆；
晶圆固定夹爪底座，用于支撑和固定晶圆；
底座驱动电机，用于带动晶圆进行旋转；
清洗液传输管路，用于向晶圆传输清洗液或其它所需液体。
进一步的，所述图像获取单元为搭载CCD传感器的图像录取装置，或者，为搭载CMOS传感器的图像录取装置，或者，为搭载CCD传感器的图像录取装置及反光镜，或者，为搭载CMOS传感器的图像录取装置及反光镜，或者，为光电转化装置。
本发明还公开了一种平坦化设备，包括研磨模组，清洗模组，设备前端模组，及用于传输晶圆的晶圆运输机构，所述清洗模组具有上述的晶圆清洁干燥模组。
本发明的有益效果是，1)图像获取单元可以对晶圆湿法清洗工艺中清洁干燥工况画面进行实时录取，图像处理单元可以对所读取图像进行实时处理，因此实现了晶圆湿法清洗工艺中清洁干燥模组完整工况下晶圆有无和晶圆是否倾斜的实时检测；2)更加简洁和方便地检测清洁干燥模块中晶圆的有无和晶圆是否水平，可以高效地检测晶圆湿法清洗工艺中清洁干燥模组中晶圆的有无及晶圆是否水平，提高了清洁干燥模块中晶圆有无检测和晶圆水平检测的准确率和检测效率，降低晶圆因未放置水平而被甩出的损坏 概率；3)对晶圆夹持装置的异常状态检测，可有效避免因晶圆固定夹爪异常导致的机械手放置或取出晶圆产生的晶圆破碎；4)通过改进装置及装置的设置位置，避免了图像获取单元被清洁干燥模组工况下清洗液传输管路传输的液体干扰的风险，降低了误检测产生的可能性；5)有效减少整个检测装置的零件数量，降低调试时间与成本，降低由于机械结构或多个检测装置结合检测导致的故障的发生率，提高装置简洁美观程度；6)由于晶圆有无检测与晶圆水平检测的检测方法只考虑与预设晶圆图像尺寸和图像参数的对比，故无论清洁干燥模组设计为竖直还是水平放置晶圆，都适用本发明中的装置和方法；7)图像获取单元获取一张图像就可以实现晶圆有无和晶圆水平的检测，减少检测的时间成本，提高检测效率；8)有效利用了晶圆呈圆形，其可以比较快速抓取圆心和半径的特性，以及晶圆轮廓的颜色相对单一，可以将其快速区分于背景色的特点；9)重复利用了晶圆夹持装置在有效夹持晶圆和在进行清洁干燥工作时的运动特征。
附图说明
图1为本发明的检测装置的主视图。
图2为本发明的检测装置的俯视图。
图3为本发明的检测装置的主视图，此时带有反光镜。
图4为本发明的检测装置的主视图，此时图像处理单元和判断单元集成至一体。
图5为本发明的图像获取单元位于晶圆中心上方的结构示意图。
图6为本发明的检测方法在获取圆轮廓时的简示图，此时图像获取单元位于晶圆正上方。
图7为本发明的检测方法在获取非圆轮廓时的简示图，此时图像获取单元位于晶圆正上方。
图8为本发明的检测方法在获取圆轮廓时的简示图，此时图像获取单元位于晶圆斜上方。
图9为本发明的检测方法在获取非圆轮廓时的简示图，此时图像获取单元位于晶圆斜上方。
图10为本发明的晶圆固定夹爪处于正常状态下的标记位示意图。
图11为本发明的晶圆固定夹爪处于异常状态下的标记位示意图。
图12为本发明晶圆状态检测的流程示意图。
图13为本发明晶圆夹持装置状态检测的流程示意图。
图14为本发明的应用流程示意图。
具体实施方式
为了使本技术领域的人员更好的理解本发明方案，下面将结合本发明实施例中的附图，对发明实施例中的技术方案进行清楚、完整的描述，显然，所描述的实施例仅仅是本发明的一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
如图1-图2所示，一种晶圆清洁干燥模组状态检测装置，包括图像获取单元2，图像处理单元4，及判断单元5。
图像获取单元2，设置在晶圆1所在平面的上方或侧方，用于获取在晶圆湿法清洗工艺中清洗干燥模组中晶圆1和/或晶圆夹持装置3的图像，其包括晶圆1正常放置的标准图像和/或晶圆夹持装置3正常状态下标记位34的目标图像，及待测图像。
具体的，当晶圆1水平放置时，图像获取单元2在晶圆1的上方平面，当晶圆1竖直放置时，图像获取单元2在晶圆1的侧方平面。
更具体的，图像获取单元2可以在晶圆1的正上方，此时图像获取单元2在晶圆1的投影落在晶圆1的中心位置，如图5所示；也可以在晶圆1的斜上方，如图1所示。在本实施例中，图像获取单元2为搭载CCD传感器的图像录取装置；或者，为搭载CMOS传感器的图像录取装置；或者，为搭载CCD传感器的图像录取装置及反光镜21，此时图像录取装置获取经过反光镜21反射后的图像，可以减少环境光对图像录取装置获取画面的干扰，可以减少由于图像录取装置斜置产生的晶圆图像形变，如图3所示；或者，为搭载CMOS传感器的图像录取装置及反光镜21；或者，为光电转化装置。
图像处理单元4，用于将图像获取单元2获取的标准图像和/或标记位的目标图像，及待测图像进行预处理，得到标准图像轮廓和/或标记位的目标图像轮廓，及待测图像轮廓。
判断单元5，将待测图像轮廓与标准图像轮廓和/或标记位的目标图像轮廓作比较，以判断差值是否在设定范围内，从而判断晶圆夹持装置的开合状态，若差值超过设定范围，则判断为状态异常。
上述图像处理单元4和判断单元5也可以集成至一体，如图4所示。
在其他实施例中，也可以设置光源，其向晶圆和/或晶圆夹持装置投射光线。
具体的，晶圆夹持装置3包括晶圆固定夹爪31，晶圆固定夹爪底座32，及底座驱动电机33。在本实施例中，晶圆固定夹爪31的数量为三个，其用于支撑、夹持和固定晶圆1；晶圆固定夹爪底座32用于支撑和固定晶圆1；底座驱动电机33则用于带动晶 圆1进行旋转。
在晶圆夹持装置3的上方还设置有清洗液传输管路6，其用于向晶圆1传输清洗液或其它所需液体。即清洗液传输管路6向晶圆1表面输送清洗液或其它液体。
本发明还公开一种化学机械平坦化设备，包括研磨模组，清洗模组，设备前端模组，及用于传输晶圆的晶圆运输机构，所述清洗模组具有上述的晶圆湿法清洗工艺中清洁干燥模组。
一种晶圆清洁干燥模组状态检测方法，包括以下步骤：
获取晶圆1正常放置的标准图像；此处的正常放置指的是晶圆1完全水平放置或完全竖直放置，不会有倾斜；
对标准图像预处理，获取晶圆1正常放置的标准图像轮廓；
获取标准图像轮廓的参数；
获取待测图像；
对待测图像预处理，判断是否存在晶圆1，若存在晶圆1则获取待测图像轮廓；
获取待测图像轮廓的参数；
将待测图像轮廓的参数与标准图像轮廓的参数作比较，判断其差值是否在设定范围内，若差值超过设定阈值，则判断为晶圆倾斜。
换句话说，晶圆湿法清洗工艺中清洁干燥模组的异常状态检测包括晶圆有无的检测，及晶圆水平的检测。如果没有晶圆或者有晶圆但呈倾斜状态，则均判定晶圆湿法清洗工艺中清洁干燥模组处于异常状态。
具体的，对于晶圆1有无的检测，可以采用图像轮廓检测方法，即通过判定待测图像轮廓是否存在晶圆轮廓进行判定，若不存在晶圆轮廓则直接判定不存在晶圆1。换句话说，所述判断是否存在晶圆1的步骤中，为判断是否存在晶圆轮廓，若存在晶圆轮廓则判定晶圆1存在。当然也可以采用现有的判断方法，对其不作限定。
上述的检测过程分为两种情况，第一种情况，当图像获取单元2在晶圆1的正上方获取晶圆1正常放置的标准图像时，获取的标准图像轮廓为圆轮廓。第二种情况，当图像获取单元2在晶圆1的斜上方获取晶圆1正常放置的标准图像时，获取的标准图像轮廓为非圆轮廓。
第一种情况下，当获取的标准图像轮廓为圆轮廓时，上述的标准图像轮廓的参数为圆心和半径。如图6所示，当晶圆1正常水平放置时，其圆心位于坐标(x1，y1)，其半径为R1。如图7所示，当晶圆1倾斜时，其圆心位于坐标(x2，y2)，其半径值不会 出现R1。
在第一种情况下，圆轮廓的检测采用霍夫变换圆检测方法。标准图像预处理和/或待测图像预处理方法为图像信号的平滑处理，具体是采用高斯滤波方法。
第二种情况下，当获取的标准图像轮廓为非圆轮廓时，上述的标准图像轮廓的参数为最小外切多边形轮廓参数。
在第二种情况下，最小外切多边形轮廓参数为边数或者为各边长或者为各边夹角。标准图像预处理和/或待测图像预处理方法为先灰度化处理，后二值化处理，最终膨胀处理，该膨胀处理将二值化处理后比较分散的轮廓连接到一起，以形成晶圆的完整轮廓。
如图8、图9所示，H与W为晶圆轮廓最小外切多边形的边长参数(此处为长方形，所以参数描述可以为长方形的长和宽)，a为晶圆晶圆轮廓最小外切多边形相邻边之间的夹角。晶圆湿法清洗工艺中清洁干燥模组的异常状态检测除了包括晶圆有无的检测，晶圆水平的检测，还包括晶圆夹持装置3的异常状态检测，即在晶圆1放置到晶圆夹持装置3之前和晶圆1从晶圆夹持装置3上取出步骤之前，都要检测晶圆夹持装置3是否处于水平或竖直的正常状态。
当晶圆夹持装置3正常放置时，晶圆夹持装置3整体呈竖直状态，其上平面呈水平状态。在晶圆进行清洁干燥工作时，底座驱动电机33带动晶圆固定夹爪底座22、晶圆固定夹爪31及晶圆1进行旋转，晶圆夹持装置3在离心力的作用下，向晶圆侧偏转一定角度，达到晶圆夹持及固定作用，此时，晶圆夹持装置3整体呈倾斜状态，其上平面呈倾斜状态。若在晶圆放置前或晶圆取出前，晶圆夹持装置3仍呈倾斜状态，则判定为晶圆夹持装置3异常。
若在晶圆夹持装置3上平面添加标记位34，则标记位34在图像获取单元2画面中产生的形变即可体现晶圆夹持装置3的状态，即可用于在晶圆放置前或晶圆取出前检测晶圆夹持装置3的状态。此时需要在晶圆夹持装置3上设置具有设定颜色和形状的标记位34。在本实施例中，标记位34设置在晶圆固定夹爪31上，为三角形，颜色可以是任意与背景色差异大的颜色，如图10所示。
具体步骤为，
获取晶圆夹持装置3正常状态下标记位34的目标图像；
对目标图像预处理，获取晶圆夹持装置3正常状态下的标记位目标图像轮廓；
获取标记位图像轮廓的参数；具体的，该参数为边数和面积；
获取晶圆夹持装置标记位的待测图像；
对标记位的待测图像预处理，得到标记位待测图像轮廓；
获取标记位待测图像轮廓的参数；具体的，该参数为边数和面积；
将标记位待测图像轮廓的参数与标记位目标图像的参数作比较，判断其差值是否在设定范围内，从而判断晶圆夹持装置的开合状态，若差值超过设定范围，则判断为晶圆夹持装置异常。
具体的，将标记位待测图像轮廓的参数与标记位目标图像的参数作比较步骤展开来说，第一步，先获取待测图像中与标记位34的颜色相同的图像轮廓；第二步，选取与标记位34有相同边数的图像轮廓，在本实施例中为选取三角形图像轮廓；第三步，判断经过上述两个步骤筛选后留下的各图像轮廓的面积是否在设定范围内；第四步，经过上述三个步骤选取得到标记位待测图像轮廓，将其与标记位目标图像的面积作比较，如果面积相同，则说明晶圆夹持装置3处于正常状态。如图11所示，当晶圆夹持装置3处于异常状态时，其标记位待测图像轮廓虽然也为三角形，但其面积与图10中的三角形面积不相同。
换句话说，获得的图像轮廓包含有多个区域的图像轮廓，先通过颜色的区别筛除一批图像轮廓，再通过形状的区别筛除一批图像轮廓，接着通过面积是否在一定范围内筛除一批图像轮廓，最终筛选出标记位待测图像轮廓，将其与标记位目标图像的面积作比较。
上述步骤为在将标记位待测图像轮廓的参数与标记位目标图像的参数作比较步骤中进行。当然，在其他实施例中，还可以在对目标图像预处理和/或对标记位的待测图像预处理步骤中，直接先排除图像中与标记位相似颜色的干扰。此时，在将标记位待测图像轮廓的参数与标记位目标图像的参数作比较步骤中，就可以省略颜色比较这一步，直接进行第二步到第四步。当然，在其他实施例中，也可以在对目标图像预处理和/或对标记位的待测图像预处理步骤中，直接先排除图像中与标记位相似形状的干扰。此时，在将标记位待测图像轮廓的参数与标记位目标图像的参数作比较步骤中，就可以省略边数比较这一步。
上述对目标图像预处理和/或对标记位的待测图像预处理步骤中，特别是排除图像中与标记位相似颜色的干扰步骤中，包括：
提取图像中设定颜色；更具体的，该步骤将原始RGB图像转化为HSV空间，提取设定颜色HSV值制作掩膜，对原始RGB图像进行掩膜操作得到分离图像；
将带有设定颜色的图像灰度化；
接着进行图像信号的平滑处理；更具体的，其采用高斯滤波方法；
边缘检测，从而描绘标记位图像的边缘；
膨胀处理，用于将图像经过上述处理后比较分散的轮廓连接到一起，从而形成标记位的完整轮廓。
具体的，展开说一下在每个晶圆固定夹爪31上均设置具有设定颜色和形状的标记位34，即在晶圆夹持装置3设置具有设定颜色和形状的多个标记位。此时检测步骤为：
获取晶圆夹持装置3正常状态下多个标记位34的目标图像；
对目标图像预处理，获取晶圆夹持装置3正常状态下的标记位目标图像轮廓，并对多个标记位目标图像轮廓进行计数编号；
获取对应计数编号的标记位目标图像轮廓的参数；具体的，该参数为边数和面积；
获取晶圆夹持装置标记位的待测图像；
对标记位的待测图像预处理，得到标记位待测图像轮廓；
获取标记位待测图像轮廓的参数，该参数包括边数和面积；
对判断为边数相同且面积在设定范围内的标记位待测图像轮廓进行编号；
将上述编号的标记位待测图像轮廓与进行计数编号的标记位目标图像轮廓的面积进行分别比较；
通过对应计数编号的标记位轮廓面积比较，判断晶圆夹持装置开合状态，若其中有一个异常即判断为晶圆夹持装置异常，此时不需要进行全部的比较。
更具体的，下面详细描述本发明的用于晶圆湿法清洗工艺中清洁干燥模组的状态检测方法步骤，其包括晶圆有无的检测、晶圆水平的检测和晶圆夹持装置的检测。
图像获取单元2获取正常状态下晶圆固定夹爪31所在图像，记为M(i-1)，对该图像进行图像处理获取晶圆夹持装置3正常状态下的标记位目标图像轮廓，对多个标记位目标图像轮廓进行计数编号并获取对应计数编号的标记位目标图像轮廓的参数，记为MK(i-1)，MK(i-1)包含标记位边数o(i-1)、个数p0及面积q(i-1)参数。
图像获取单元2获取正常状态下晶圆1所在图像，记为N(i-1)，对该图像进行处理，获取晶圆正常状态下的标准图像轮廓，获取晶圆标准图像轮廓的参数记为NK(i-1)，若图像获取单元2放置于晶圆1所在平面正上方，NK(i-1)包含正常状态下晶圆1圆心在图像中的位置(x0,y0)及半径R0，若图像获取单元2放置于晶圆1所在平面斜上方，NK(i-1)则包含正常状态下晶圆轮廓最小外切多边形轮廓的边数a0，各边长b(i-1)和各边夹角c(i-1)。
在晶圆1放置之前，图像获取单元2获取工况下晶圆固定夹爪31所在图像，记为M(i)，对该图像进行图像处理，获取晶圆夹持装置3的待测标记位图像轮廓，对多个待测标记位图像轮廓进行计数编号并获取对应计数编号的标记位图像轮廓的参数，记为MK(i)，MK(i)包含标记位边数o(i)、个数p1及面积q(i)参数，将MK(i)与MK(i-1)进行比较，若其差值超过所设定阈值，则由判断单元5判断为模组晶圆固定夹爪31异常，由上位机判断单元5做出相应操作，若其差值在所设定阈值之内，则判断为模组晶圆夹持装置3正常，开始进行放置晶圆1工作。
在放置晶圆1之后，图像获取单元2获取图像，由图像处理方法判断是否存在晶圆，若晶圆不存在，则由判断单元5判断为模组中晶圆异常，由上位机判断单元5做出相应操作，若晶圆存在，则将该画面记为N(i)，对该图像进行处理，获取晶圆的图像轮廓，获取晶圆图像轮廓的参数记为NK(i)，若图像获取单元2放置于晶圆1所在平面正上方，NK(i)包含晶圆圆心在图像中的位置(x1,y1)及半径R1，若图像获取单元2放置于晶圆1所在平面斜上方，NK(i)则包含晶圆轮廓最小外切多边形轮廓的边数a1，各边长b(i)和各边夹角c(i)。将NK(i)与NK(i-1)进行比较，若其差值超过所设定阈值，则由判断单元5判断为模组中晶圆异常，由上位机判断单元5做出相应操作，若其差值在所设定阈值之内，则判断为模组中晶圆正常，清洗液传输管路6向晶圆1传输清洗液或液体，开始进行晶圆清洗干燥工作。
在晶圆清洗干燥工作结束后，图像获取单元2获取图像，由图像处理方法判断是否存在晶圆，若晶圆不存在，则由判断单元5判断为模组中晶圆异常，由上位机判断单元5做出相应操作，若晶圆存在，则将该画面记为N(i+1)，对该图像进行处理，获取晶圆的图像轮廓，获取晶圆图像轮廓的参数记为NK(i+1)，若图像获取单元2放置于晶圆所在平面正上方，NK(i+1)包含晶圆圆心在图像中的位置(x2,y2)及半径R2，若图像获取单元2放置于晶圆所在平面斜上方，NK(i+1)则包含晶圆轮廓最小外切多边形轮廓的边数a2，各边长b(i+1)和各边夹角c(i+1)。将NK(i+1)与NK(i-1)进行比较，若其差值超过所设定阈值，则由判断单元5判断为模组中晶圆异常，由上位机判断单元5做出相应操作，若其差值在所设定阈值之内，则判断为模组中晶圆正常，开始进行检测晶圆夹爪状态。
开始进行检测晶圆夹持装置3状态，图像获取单元2获取工况下晶圆固定夹爪31所在图像，记为M(i+1)，对该图像进行图像处理获取晶圆夹持装置3的待测标记位图像轮廓，对多个待测标记位图像轮廓进行计数编号并获取对应计数编号的标记位图像轮廓的参数，记为MK(i+1)，MK(i+1)包含标记位边数o(i+1)、个数p2及面积q(i+1)参数， 将MK(i+1)与MK(i-1)进行比较，若其差值超过所设定阈值，则由判断单元5判断为模组晶圆夹持装置3异常，由上位机判断单元5做出相应操作，若其差值在所设定阈值之内，则判断为模组晶圆固定夹爪31正常，开始进行取出晶圆工作。
待晶圆取出，单个晶圆的清洗干燥工况完成。
上述具体实施方式用来解释说明本发明，而不是对本发明进行限制，在本发明的精神和权利要求的保护范围内，对本发明作出的任何修改和改变，都落入本发明的保护范围。

Claims (24)

  1. 一种晶圆清洁干燥模组状态检测方法，其特征在于，包括以下步骤：
    获取晶圆正常放置的标准图像；
    对标准图像预处理，获取晶圆正常放置的标准图像轮廓；
    获取标准图像轮廓的参数；
    获取待测图像；
    对待测图像预处理，判断是否存在晶圆，若存在晶圆则获取待测图像轮廓；
    获取待测图像轮廓的参数；
    将待测图像轮廓的参数与标准图像轮廓的参数作比较，判断其差值是否在设定范围内，若差值超过设定阈值，则判断为晶圆倾斜。
  2. 根据权利要求1所述的晶圆清洁干燥模组状态检测方法，其特征在于：
    当获取的标准图像轮廓为圆轮廓时，所述标准图像轮廓的参数为圆心和半径；
    当获取的标准图像轮廓为非圆轮廓时，所述标准图像轮廓的参数为最小外切多边形轮廓参数。
  3. 根据权利要求2所述的晶圆清洁干燥模组状态检测方法，其特征在于：
    当在晶圆正上方获取晶圆正常放置的标准图像时，获取的标准图像轮廓为圆轮廓；
    当在晶圆斜上方获取晶圆正常放置的标准图像时，获取的标准图像轮廓为非圆轮廓。
  4. 根据权利要求2所述的晶圆清洁干燥模组状态检测方法，其特征在于：圆轮廓检测采用霍夫变换圆检测方法。
  5. 根据权利要求2所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述最小外切多边形轮廓参数为边数或各边长或各边夹角。
  6. 根据权利要求1所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述判断是否存在晶圆步骤中，为判断是否存在晶圆轮廓，若存在晶圆轮廓则判定晶圆存在。
  7. 根据权利要求2所述的晶圆清洁干燥模组状态检测方法，其特征在于：当获取的标准图像轮廓为圆轮廓时，所述标准图像预处理和/或待测图像预处理方法为图像信号的平滑处理。
  8. 根据权利要求7所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述图像信号的平滑处理采用高斯滤波方法。
  9. 根据权利要求2所述的晶圆清洁干燥模组状态检测方法，其特征在于：当获取的标准图像轮廓为非圆轮廓时，所述标准图像预处理和/或待测图像预处理方法为先灰度化处理，后二值化处理，最终膨胀处理，该膨胀处理将二值化处理后比较分散的轮廓连接到一起，以形成晶圆的完整轮廓。
  10. 根据权利要求1所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述晶圆为静止状态或旋转状态。
  11. 根据权利要求1所述的晶圆清洁干燥模组状态检测方法，其特征在于，还包括以下步骤：
    在晶圆夹持装置设置具有设定颜色和形状的标记位；
    获取晶圆夹持装置正常状态下标记位的目标图像；
    对标记位的目标图像预处理，获取晶圆夹持装置正常状态下的标记位目标图像轮廓；
    获取标记位目标图像轮廓的参数；
    获取晶圆夹持装置标记位的待测图像；
    对标记位的待测图像预处理，得到标记位待测图像轮廓；
    获取标记位待测图像轮廓的参数；
    将标记位待测图像轮廓的参数与标记位目标图像的参数作比较，判断其差值是否在设定范围内，从而判断晶圆夹持装置的开合状态。
  12. 根据权利要求11所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述标记位目标图像轮廓的参数为边数和面积。
  13. 根据权利要求12所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述将标记位待测图像轮廓的参数与标记位目标图像的参数作比较步骤中，先获取待测图像中与标记位的颜色相同的图像轮廓，选取与标记位有相同边数的图像轮廓，再判断各图像轮廓的面积是否在设定范围内，通过上述步骤选取标记位待测图像轮廓，将其与标记位目标图像的面积作比较。
  14. 根据权利要求11所述的晶圆清洁干燥模组状态检测方法，其特征在于，包括以下步骤，
    在晶圆夹持装置设置具有设定颜色和形状的多个标记位；
    获取晶圆夹持装置正常状态下多个标记位的目标图像；
    对目标图像预处理，获取晶圆夹持装置正常状态下的标记位目标图像轮廓，并对多个标记位目标图像轮廓进行计数编号；
    获取对应计数编号的标记位目标图像轮廓的参数；
    获取晶圆夹持装置标记位的待测图像；
    对标记位的待测图像预处理，得到标记位待测图像轮廓；
    获取标记位待测图像轮廓的参数，该参数包括边数和面积；
    对判断为边数相同且面积在设定范围内的标记位待测图像轮廓进行编号；
    将上述编号的标记位待测图像轮廓与进行计数编号的标记位目标图像轮廓的面积进行分别比较；
    若其中有一个异常即判断为晶圆夹持装置异常。
  15. 根据权利要求11所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述对标记位的目标图像预处理和/或对标记位的待测图像预处理步骤中，包括提取设定颜色，排除干扰颜色步骤。
  16. 根据权利要求15所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述对标记位的目标图像预处理和/或对标记位的待测图像预处理步骤中，包括排除图像中干扰形状的步骤。
  17. 根据权利要求15所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述对目标图像预处理和/或对标记位的待测图像预处理步骤中，包括以下步骤，
    提取图像中设定颜色；
    灰度化；
    图像信号的平滑处理；
    边缘检测，以描绘标记位图像的边缘；
    膨胀处理，用于将图像经上述处理后比较分散的轮廓连接到一起，以形成标记位的完整轮廓。
  18. 根据权利要求17所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述提取图像中设定颜色步骤包括，将原始RGB图像转化为HSV空间，提取设定颜色HSV值制作掩膜，对原始RGB图像进行掩膜操作得到分离图像。
  19. 根据权利要求17所述的晶圆清洁干燥模组状态检测方法，其特征在于：所述图像信号的平滑处理步骤采用高斯滤波方法。
  20. 根据权利要求10所述的晶圆清洁干燥模组状态检测方法，其特征在于：在晶圆放置于晶圆夹持装置和晶圆自晶圆夹持装置取出步骤之前进行。
  21. 一种晶圆清洁干燥模组状态检测装置，其特征在于包括：
    图像获取单元，设于晶圆所在平面上方或侧方，用于获取晶圆湿法清洗工艺中清洁干燥模组中晶圆和/或晶圆夹持装置的图像，其包括晶圆正常放置的标准图像和/或晶圆夹持装置正常状态下标记位的目标图像，及待测图像；
    图像处理单元，用于将图像获取单元获取的标准图像和/或标记位的目标图像，及待测图像进行预处理，得到标准图像轮廓和/或标记位的目标图像轮廓，及待测图像轮廓；
    判断单元，将待测图像轮廓与标准图像轮廓和/或标记位的目标图像轮廓作比较，以判断差值是否在设定范围内，从而判断晶圆夹持装置的开合状态。
  22. 根据权利要求21所述的晶圆清洁干燥模组状态检测装置，其特征在于，还包括：
    晶圆固定夹爪，用于支撑、夹持和固定晶圆；
    晶圆固定夹爪底座，用于支撑和固定晶圆；
    底座驱动电机，用于带动晶圆进行旋转；
    清洗液传输管路，用于向晶圆传输清洗液或其它所需液体。
  23. 根据权利要求21所述的晶圆清洁干燥模组状态检测装置，其特征在于：所述图像获取单元为搭载CCD传感器的图像录取装置，或者，为搭载CMOS传感器的图像录取装置，或者，为搭载CCD传感器的图像录取装置及反光镜，或者，为搭载CMOS传感器的图像录取装置及反光镜，或者，为光电转化装置。
  24. 一种平坦化设备，包括研磨模组，清洗模组，设备前端模组，及用于传输晶圆的晶圆运输机构，所述清洗模组具有如权利要求21-23中任一项所述的晶圆清洁干燥模组。
PCT/CN2023/071167 2022-05-11 2023-01-09 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备 WO2023216643A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210508423.9A CN114612474B (zh) 2022-05-11 2022-05-11 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备
CN202210508423.9 2022-05-11

Publications (1)

Publication Number Publication Date
WO2023216643A1 true WO2023216643A1 (zh) 2023-11-16

Family

ID=81870462

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/071167 WO2023216643A1 (zh) 2022-05-11 2023-01-09 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备

Country Status (3)

Country Link
CN (1) CN114612474B (zh)
TW (1) TW202345096A (zh)
WO (1) WO2023216643A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114612474B (zh) * 2022-05-11 2023-12-19 杭州众硅电子科技有限公司 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6405101B1 (en) * 1998-11-17 2002-06-11 Novellus Systems, Inc. Wafer centering system and method
JP2009145090A (ja) * 2007-12-12 2009-07-02 Kobelco Kaken:Kk 形状測定装置
CN102867765A (zh) * 2012-09-27 2013-01-09 盛美半导体设备（上海）有限公司 晶圆位置检测装置及检测方法
CN103681427A (zh) * 2013-12-09 2014-03-26 深圳市大族激光科技股份有限公司 一种基于视觉的晶圆旋转纠正与中心定位的方法
CN105185728A (zh) * 2015-06-17 2015-12-23 北京七星华创电子股份有限公司 一种基于图像的硅片分布状态识别方法及装置
JP2016155204A (ja) * 2015-02-26 2016-09-01 株式会社東京精密 研削加工装置
CN110146035A (zh) * 2019-04-12 2019-08-20 中民筑友科技投资有限公司 构件生产线的预埋件检测方法、装置、设备及系统
CN114199193A (zh) * 2020-09-18 2022-03-18 中国科学院微电子研究所 一种倾斜检测装置、方法和机械臂系统
CN114612474A (zh) * 2022-05-11 2022-06-10 杭州众硅电子科技有限公司 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06235697A (ja) * 1993-02-09 1994-08-23 Matsushita Electric Ind Co Ltd Ｉｃ検査装置
JP3774320B2 (ja) * 1998-01-14 2006-05-10 アルプス電気株式会社 ウエハの位置把握方法並びに露光方法及び露光装置
JP4326641B2 (ja) * 1999-11-05 2009-09-09 富士機械製造株式会社 装着装置，装着精度検出治具セットおよび装着精度検出方法
JP2013061468A (ja) * 2011-09-13 2013-04-04 Sharp Corp 画像形成装置
CN104400217A (zh) * 2014-10-17 2015-03-11 中国电子科技集团公司第二十九研究所 一种全自动激光焊接方法及装置
CN107274349B (zh) * 2017-06-05 2021-03-09 新疆大学 生物芯片荧光图像倾斜角度的确定方法及装置
CN109118476B (zh) * 2018-07-11 2020-11-24 华南理工大学广州学院 一种零部件边缘轮廓完整性检测方法及装置
CN109204378A (zh) * 2018-08-14 2019-01-15 北京交通大学 基于轮廓势能和轮廓向量的车轮非圆度检测方法及系统
CN211602965U (zh) * 2020-01-17 2020-09-29 广州祥亿玻璃制品有限公司 用于瓶体检测的图像采集识别系统
CN113379681B (zh) * 2021-05-20 2022-11-04 深圳技术大学 Led芯片倾斜角度的获取方法、系统、电子装置及存储介质
CN113752260B (zh) * 2021-09-07 2023-12-26 京东方科技集团股份有限公司 一种取料定位修正方法及装置

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6405101B1 (en) * 1998-11-17 2002-06-11 Novellus Systems, Inc. Wafer centering system and method
JP2009145090A (ja) * 2007-12-12 2009-07-02 Kobelco Kaken:Kk 形状測定装置
CN102867765A (zh) * 2012-09-27 2013-01-09 盛美半导体设备（上海）有限公司 晶圆位置检测装置及检测方法
CN103681427A (zh) * 2013-12-09 2014-03-26 深圳市大族激光科技股份有限公司 一种基于视觉的晶圆旋转纠正与中心定位的方法
JP2016155204A (ja) * 2015-02-26 2016-09-01 株式会社東京精密 研削加工装置
CN105185728A (zh) * 2015-06-17 2015-12-23 北京七星华创电子股份有限公司 一种基于图像的硅片分布状态识别方法及装置
CN110146035A (zh) * 2019-04-12 2019-08-20 中民筑友科技投资有限公司 构件生产线的预埋件检测方法、装置、设备及系统
CN114199193A (zh) * 2020-09-18 2022-03-18 中国科学院微电子研究所 一种倾斜检测装置、方法和机械臂系统
CN114612474A (zh) * 2022-05-11 2022-06-10 杭州众硅电子科技有限公司 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备

Also Published As

Publication number Publication date
CN114612474B (zh) 2023-12-19
CN114612474A (zh) 2022-06-10
TW202345096A (zh) 2023-11-16

Similar Documents

Publication Publication Date Title
JP3051279B2 (ja) バンプ外観検査方法およびバンプ外観検査装置
WO2023216643A1 (zh) 一种晶圆清洁干燥模组状态检测方法、装置及平坦化设备
TWI778078B (zh) 用於自動缺陷分類之方法及系統以及相關非暫時性電腦程式產品
US10417757B2 (en) Image inspection apparatus and image inspection method
CN107101598B (zh) 一种压电陶瓷银片同心度质量自动检测方法和装置
CN107290347B (zh) 蜂窝载体缺陷自动化检测方法
CN115020267B (zh) 一种半导体表面缺陷检测方法
CN109900719B (zh) 一种叶片表面刀纹的视觉检测方法
CN111307819B (zh) 晶圆边缘缺陷检测系统及方法
TWI770906B (zh) 晶圓表面缺陷檢測方法及其裝置
JP2010034133A (ja) 多結晶シリコンウエハのクラック検出装置
KR20160084212A (ko) 불량 검사 장치 및 방법
CN111855666B (zh) 一种轴承内圈侧周外观缺陷自动检测方法和系统
JP4823996B2 (ja) 輪郭検出方法及び輪郭検出装置
JP2010025597A (ja) 欠陥検出装置、欠陥検出方法およびプログラム
CN216411102U (zh) 一种硅片边缘检测系统
US11821847B2 (en) Wafer backside defect detection method and wafer backside defect detection apparatus
CN212585665U (zh) 一种内螺纹垂直度检测装置
JP4337581B2 (ja) 半導体ウエーハの両面研磨装置及び割れ検査方法
JP3632501B2 (ja) 結晶欠陥の検査方法及び検査装置
CN112712555A (zh) 一种茯苓表皮视觉定位方法
TWI609188B (zh) 檢測設備及其檢測方法
CN210180940U (zh) 检测硅晶圆缺陷的自动光学检测机构
CN217190987U (zh) 硅片尺寸检测筛选系统
TWI691715B (zh) 檢測矽晶圓缺陷的自動光學檢測機構及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23802420

Country of ref document: EP

Kind code of ref document: A1