WO2023006079A1

WO2023006079A1 - Wafer pre-alignment apparatus and wafer pre-alignment method

Info

Publication number: WO2023006079A1
Application number: PCT/CN2022/109018
Authority: WO
Inventors: 金攀; 肖蓉; 周许超; 刘涛; 潘炼东
Original assignee: 上海微电子装备（集团）股份有限公司
Priority date: 2021-07-29
Filing date: 2022-07-29
Publication date: 2023-02-02
Also published as: CN215299213U; CN117616548A

Abstract

Provided in the present invention are a wafer pre-alignment apparatus and a wafer pre-alignment method. The wafer pre-alignment apparatus comprises a support assembly, a camera, and a light source. The support assembly is used to support a wafer, an upper surface of the support assembly for placing the wafer being provided with at least three positioning markers located at the periphery of the wafer, the camera being used to capture an image of the wafer and the support assembly supporting the wafer, and the image comprising all of the positioning markers on the support assembly. The light source is used to provide an illumination beam such that an illumination light field is formed on the upper surface of the support assembly when an image is captured. The image captured by the camera can be used to obtain a center deviation amount and a rotation angle of the wafer without one revolution of the wafer, which can reduce mechanical complexity of the wafer pre-alignment apparatus, help reduce manufacturing costs of the wafer pre-alignment apparatus, and improve pre-alignment efficiency. In the wafer pre-alignment method provided in the present invention, the center deviation amount and the rotation angle of the wafer are obtained using the image captured by the camera.

Description

Wafer pre-alignment device and wafer pre-alignment method

technical field

The invention relates to the field of semiconductor equipment, in particular to a wafer pre-alignment device and a wafer pre-alignment method.

Background technique

Since the field of view of the lithography machine is very small, before the wafer is transferred to the lithography machine for exposure, the wafer must be pre-aligned so that the mark of the wafer transferred to the exposure table is within the field of view of the lithography machine. Inside. The wafer pre-alignment device is used to pre-align the wafer. The pre-alignment process generally makes the center (center) of the wafer within a certain range, while the notch of the wafer stays at a specified angle, that is, It includes two main processes of center positioning and gap orientation.

At present, when the wafer pre-alignment device detects the center deviation and rotation angle of the wafer, it needs to make the wafer rotate once to obtain the complete edge information of the wafer. Although this increases the data volume of the wafer edge information, it makes the Rotating the wafer once does not improve the accuracy of collecting wafer edge information, nor does it improve the accuracy of wafer pre-alignment, and also reduces the efficiency of wafer pre-alignment.

In addition, in the wafer pre-alignment device, the wafer is usually set on the chuck assembly. In order to make the wafer rotate a circle to collect edge information during pre-alignment, the wafer pre-alignment device needs to be provided with a driving assembly and can drive the chuck assembly, for example. A mechanism for lifting, rotating or translating, thus increasing the complexity of the mechanism in the wafer pre-alignment device and increasing the manufacturing cost of the wafer pre-alignment device.

Contents of the invention

The wafer pre-alignment device and the wafer pre-alignment method provided by the invention can reduce the manufacturing cost and improve the pre-alignment efficiency.

In order to achieve the above object, the present invention provides a wafer pre-alignment device. The wafer pre-alignment device includes a support assembly, a camera and a light source. The support assembly is used to support the wafer, and the upper surface of the support assembly on which the wafer is placed is provided with at least three positioning mark points located on the periphery of the wafer. The camera is used to take images of the wafer and the supporting assembly supporting the wafer, wherein the image includes all positioning marker points on the supporting assembly. The light source is used to provide an illumination light beam, so that an illumination light field is formed on the upper surface of the support component when the image is taken.

Optionally, the shape of each positioning mark point is a circle, an ellipse or a polygon.

Optionally, the distances from the at least three positioning marking points to the center of the upper surface of the support assembly are equal.

Optionally, the supporting component is a fork or a pre-alignment platform of a robot.

Optionally, the at least three positioning mark points are uniformly arranged on a circle centered on the center of the upper surface of the fork or the pre-alignment platform.

Optionally, the light source includes at least one surface light source, and the light emitting surface of the surface light source faces the support assembly.

Optionally, the surface light source is arranged opposite to the upper surface of the support assembly; or, the surface light source is arranged opposite to the lower surface of the support assembly and part of the light-emitting surface of the surface light source exceeds the surface of the support assembly. lower surface.

Optionally, the light source includes two surface light sources, and the two surface light sources are respectively arranged opposite to the upper surface and the lower surface of the support component.

Optionally, several vacuum suction cups are arranged on the upper surface of the support assembly.

Optionally, the camera is a CCD camera or a CMOS camera.

The present invention provides a wafer pre-alignment method, the wafer pre-alignment method comprising:

A support assembly is provided, the support assembly is used to support the wafer, and the upper surface of the support assembly on which the wafer is placed is provided with at least three positioning mark points located on the periphery of the wafer;

a camera taking an image of the wafer and the support assembly supporting the wafer, the image including at least three of the positioning markers on the support assembly;

converting the image to a top plan view based on the actual physical locations of the centers of at least three locator markers in the image on the support assembly and the positions of the centers of each of the locator markers in the image;

determining the position of the center of the wafer in the top plan view, and determining the positional offset of the center of the wafer from the center of the support assembly; and

Determining an apex position of the notch of the wafer in the top plan view, and determining a rotation angle of the wafer based on the center of the wafer and the apex position of the notch in the top plan view.

Optionally, the method for converting the image into a top plan view includes: performing automatic threshold binarization processing and connected domain area screening on the image, so as to determine where the center of the positioning marker point on the support component is located. The position in the image; based on the actual physical position of the center of the at least three positioning marker points on the support assembly and the position of the center of each of the positioning marker points in the image, calculate and obtain a mapping transformation matrix and performing mapping transformation on the image according to the mapping transformation matrix to form a top plan view of the wafer and the support assembly.

Optionally, the center of the top plan view coincides with the center of the support assembly in the top plan view; or, there is a set distance between the center of the top view plan view and the center of the support assembly in the top view plan view.

Optionally, the method for determining the position of the wafer center in the top plan view and determining the position deviation between the wafer center and the center of the support assembly includes: performing automatic threshold binarization on the top view plan view processing, and then filter out the region of the wafer in the top view plan through the area of the connected domain; use the Ganny algorithm to extract wafer edge information in the top view plan, and perform circle fitting according to the wafer edge information to obtain the wafer center The position in the top plan view; and based on the position of the wafer center in the top view plan, obtain the number of deviation pixels between the wafer center and the center of the support assembly in the top view plan, and then according to the pixels in the top view plan The size calculates the positional deviation between the center of the wafer and the center of the support assembly.

Optionally, the method for determining the apex position of the notch of the wafer in the top plan view includes: calculating the distance from the edge point of the wafer to the center of the wafer in the top plan view, and filtering out a candidate region of the gap; and calculating the similarity between the candidate region of the gap and the standard template of the gap, and determining the position of the apex of the gap at the position with the highest similarity.

Optionally, the method for determining the apex position of the notch of the wafer in the top plan view includes: cutting several small pictures from the edge of the wafer in the top plan view; using a feature extraction algorithm to extract the feature; use machine learning algorithm to classify the features of each of the small pictures and analyze whether there is a gap in each of the small pictures, and use the small picture with the gap as the candidate area of the gap; combine the candidate area of the gap with the gap The standard template uses a geometric template matching algorithm to locate the gap, and outputs the position and confidence of the gap on each of the small images; and judges the final position of the gap according to the confidence corresponding to each of the small images, and outputs the The position is transformed into the top plan view, and the position of the apex of the notch in the top plan view is determined.

Optionally, the method for determining the position of the vertex of the notch of the wafer in the top plan view includes: intercepting several small pictures at the edge of the wafer in the top view plan view; Figures are directly input to obtain the position and confidence of the gap on each of the small pictures; and judge the final position of the gap by the confidence corresponding to each of the small pictures, and convert the position of the gap to the position of the gap. In the top plan view, determine the apex position of the gap in the top plan view.

Optionally, the method for determining the vertex position of the wafer notch in the top plan view includes: using a deep learning target detection algorithm, inputting the top view plan view to obtain the position of the notch in the top view plan view, and then determining the position of the notch in the top view plan view. The apex position of the notch in the top plan view;

Alternatively, the method for determining the position of the apex of the notch of the wafer in the top plan view includes: using a deep learning target detection algorithm to input the image to obtain the position of the notch in the image; The position is transformed into the top plan view, the position of the notch in the top view plan is obtained, and then the apex position of the notch in the top view plan is determined.

Optionally, the method for determining the apex position of the wafer notch in the top plan view includes: performing polar coordinate transformation on the top view plan view, and stretching the donut diagram of the wafer edge in the top view plan view as Histogram; based on geometric or pixel grayscale features, search for a region matching the gap standard template in the histogram as a candidate region of the gap, and output the position and confidence of the gap on the candidate region; according to each Judging the final position of the notch based on the confidence level corresponding to the candidate area, and converting the position of the notch into the top plan view; and determining the apex position of the notch in the top view plan view.

Optionally, the method for determining the apex position of the notch of the wafer in the top plan view includes: performing a search in the top view plan view, and determining the region matching the notch standard template in the top view plan view as the notch region, Then determine the position of the apex of the gap in the top plan view;

Alternatively, the method for determining the position of the vertex of the notch of the wafer in the top plan view includes: searching in the image, and determining the region in the image that matches the notch standard template as the notch region; The position of the notch area in the center is transformed into the top view plan view, the position of the notch in the top view plan view is obtained, and then the apex position of the notch in the top view plan view is determined.

In the wafer pre-alignment device and wafer pre-alignment method of the present invention, the upper surface of the support assembly for placing the wafer is provided with at least three positioning mark points located on the periphery of the wafer, and the camera photographs the wafer and the support The image formed by the support assembly of the wafer includes all the positioning mark points on the support assembly, and the image can be used to obtain the position deviation between the center of the wafer and the center of the upper surface of the support assembly (which can be referred to as the center deviation of the wafer) And to obtain the rotation angle of the wafer, that is to say, the present invention sets at least three positioning mark points on the upper surface of the support assembly where the wafer is placed, and obtains the center deviation and rotation angle of the wafer using a single image taken by the camera The wafer does not need to be rotated once, which can reduce the complexity of the mechanism of the wafer pre-alignment device, help to reduce the manufacturing cost of the wafer pre-alignment device and improve the pre-alignment efficiency.

Description of drawings

FIG. 1 is a schematic structural diagram of a wafer pre-alignment device according to an embodiment of the present invention.

FIG. 2 is a schematic plan view of a fork in a wafer pre-alignment device according to an embodiment of the present invention.

FIG. 3 is a schematic diagram of the structure of the surface light source and the lower surface of the fork in the wafer pre-alignment device according to an embodiment of the present invention.

FIG. 4 is a schematic structural diagram of two surface light sources provided in a wafer pre-alignment device according to an embodiment of the present invention.

FIG. 5 is a schematic structural diagram of a wafer pre-alignment device according to another embodiment of the present invention.

FIG. 6 is a schematic plan view of a pre-alignment stage in a wafer pre-alignment device according to an embodiment of the present invention.

FIG. 7 is a flowchart of a wafer pre-alignment method according to an embodiment of the present invention.

Explanation of reference numerals: 100a-fork; 100b-pre-alignment stage; 101-positioning mark point; 102-vacuum chuck; 200-wafer; 300-surface light source; 400-camera.

Detailed ways

The wafer pre-alignment device proposed by the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer from the following description. It should be noted that the drawings are all in very simplified form and use imprecise scales, and are only used to facilitate and clearly assist the purpose of illustrating the embodiments of the present invention.

Embodiment one

In order to reduce manufacturing costs and improve pre-alignment efficiency, this embodiment provides a wafer pre-alignment device. FIG. 1 is a schematic structural diagram of a wafer pre-alignment device according to an embodiment of the present invention. FIG. 2 is a schematic plan view of a fork in a wafer pre-alignment device according to an embodiment of the present invention. As shown in FIG. 1 and FIG. 2 , the wafer pre-alignment device includes a support assembly (such as a fork 100 a of a robot arm), a light source (such as a surface light source 300 ) and a camera 400 . The support assembly is used to support the wafer 200 , and the upper surface of the support assembly for placing the wafer 200 is provided with at least three positioning mark points 101 located on the periphery of the wafer 200 . The camera 400 is used to take images of the wafer 200 and the supporting assembly supporting the wafer 200 , wherein the image includes all positioning marker points 101 on the supporting assembly. The light source is used to provide an illumination light beam, so that an illumination light field is formed on the upper surface of the wafer 200 and the support assembly when the image is captured.

In this embodiment, the supporting component may be a fork 100a of a manipulator. The wafer pre-alignment device of this embodiment will be described below by taking the fork 100a whose supporting component is a manipulator as an example.

Since three points can define a plane, in this embodiment, the number of positioning mark points 101 on the upper surface of the fork 100a is set to be at least three. As shown in FIG. 2 , the shape of the positioning marker point 101 may be a circle. But not limited thereto, the positioning marker point 101 can also be in other shapes such as ellipse or polygon (such as rectangle, triangle), as long as the center of the positioning marker point 101 can be easily found in the image captured by the camera 400 .

In order to improve the contrast of the positioning mark point 101 in the image captured by the camera, in this embodiment, the reflectivity of the surface of the positioning mark point 101 can be greater than that of other areas on the upper surface of the sheet fork 100a where no positioning mark point is set, that is, in the When illuminated by a light source, the positioning mark point 101 is brighter than other areas on the upper surface of the fork 100a. In this embodiment, the positioning mark point 101 can be formed by digging out a groove corresponding to the pattern on the upper surface of the fork 100a and filling the groove with a material with high reflectivity.

In this embodiment, the distances from the at least three positioning mark points 101 to the center of the upper surface of the fork 100a (that is, the geometric center of the upper surface of the fork 100a) can be equal, so as to pass through all the positions in the image captured by the camera. The center of the positioning mark point 101 determines the center of the upper surface of the fork 100a. For example, at least three positioning mark points 101 are uniformly arranged on a circle centered on the center of the upper surface of the fork 100a.

As shown in FIG. 2, the fork 100a may have several (for example three) support arms, the upper surfaces of the several support arms are coplanar, and the support arms protrude below the wafer 200 to support the wafer 200. Wafer 200. A vacuum chuck 102 may be provided on each of the support arms, and the vacuum chuck 102 may be used to absorb and fix the wafer 200 . A connecting piece is provided between the support arms of the fork 100a, and the connecting piece can be located at the periphery of the wafer 200 and surrounds the wafer 200, and the at least three positioning mark points 101 can be specifically arranged on the connecting piece. The upper surface of the component, so that the wafer 200 will not block the positioning markers 101 .

In this embodiment, the light source may include at least one surface light source 300, and the light emitting surface of the surface light source 300 may be arranged facing the fork 100a. The surface light source 300 is, for example, a flat light source. But not limited thereto, in other embodiments, the light source can also be other light sources such as point light sources, as long as the camera 400 can form an illumination light field on the upper surface of the fork 100a when the camera 400 captures an image.

As shown in Figure 1, the light source may include a surface light source 300, the surface light source 300 may be arranged opposite to the upper surface of the sheet fork 100a, and the light emitting surface of the surface light source 300 and the upper surface of the sheet fork 100a may have a The included angle is set so that the surface light source 300 is irradiating the upper surface of the wafer 200 and the upper surface of the fork 100a with positive light, and a space can be left for installing the camera 400 .

FIG. 3 is a schematic diagram of the structure of the surface light source and the lower surface of the fork in the wafer pre-alignment device according to an embodiment of the present invention. As shown in FIG. 3 , the surface light source 300 may be arranged opposite to the lower surface of the fork 100a, the light emitting surface of the surface light source 300 and the lower surface of the fork 100a may have a set angle, and the Part of the light-emitting surface of the surface light source 300 can exceed the lower surface of the fork 100a, so that the surface light source 300 can backlight (i.e. not illuminate) the upper surface of the wafer 200 and the fork 100a, However, the brightness of the upper surface of the wafer 200 and the fork 100a is relatively dark, which can reduce the impact of the graphics (patterns) on the upper surface of the wafer 200 on the shooting effect of the camera 400 (that is, the impact on the image obtained by the camera 400). ) to improve the stability of the wafer pre-alignment device.

FIG. 4 is a schematic structural diagram of two surface light sources provided in a wafer pre-alignment device according to an embodiment of the present invention. In order to ensure that the upper surface of the wafer 200 and the fork 100a has a certain brightness, and reduce the influence of the surface pattern of the wafer 200, as shown in FIG. 4, the light source of the wafer pre-alignment device may include two surface light sources 300, the two surface light sources 300 can be arranged opposite to the upper surface and the lower surface of the sheet fork 100a respectively, that is, one of the two surface light sources 300 is arranged opposite to the upper surface of the sheet fork 100a, and the two surface light sources 300 The other one can be set opposite to the lower surface of the fork 100a to form a positive and negative complementary configuration.

In this embodiment, the camera 400 may be a charge coupled device camera (Charge Coupled Device Camera, CCD camera) or a complementary metal oxide semiconductor (Complementary Metal-Oxide-Semiconductor, CMOS) image sensor, a complementary metal oxide semiconductor image sensor It can be referred to as a CMOS camera for short. Specifically, the camera 400 may be an area array CCD camera, which is used to take images of the wafer 200 and the upper surface of the fork 100a. In this embodiment, all the positioning mark points 101 are located within the shooting range of the camera 400, so that the image captured by the camera 400 includes the image of the wafer 200 and the images of all the positioning mark points 101, and is captured by the camera 400 The positional deviation between the center of the wafer 200 and the center of the upper surface of the fork 100 a (abbreviated as the center deviation of the wafer 200 ) and the rotation angle of the wafer 200 can be obtained from the image.

The wafer pre-alignment method provided in the present application uses the images captured by the camera 400 to obtain the center deviation and the rotation angle of the wafer 200 . In this embodiment, the so-called orientation measurement may refer to the measurement of the dwell angle of the notch on the wafer 200 . The wafer pre-alignment method includes:

S1, providing a support assembly, the support assembly is used to support the wafer, and the upper surface of the support assembly on which the wafer is placed is provided with at least three positioning mark points located on the periphery of the wafer;

S2. The camera captures an image of the wafer and the support assembly supporting the wafer, where the image includes at least three positioning marker points on the support assembly;

S3, based on the actual physical positions of the centers of at least three positioning marker points in the image on the support assembly and the positions of the centers of each of the positioning marker points in the image, converting the image into a top view floor plan;

S4. Determine the position of the wafer center in the top plan view, and determine the positional deviation between the wafer center and the center of the support assembly; and

S5. Determine the apex position of the notch of the wafer in the top plan view, and determine the rotation angle of the wafer based on the center of the wafer and the apex position of the notch in the top plan view.

Hereinafter, the wafer pre-alignment method will be described by taking the supporting component as an example.

1 to 2, a fork 100a is provided, the fork 100a is used to support the wafer 200, and the upper surface of the fork 100a on which the wafer 200 is placed is provided with at least three positioning positions located at the periphery of the wafer 200. Mark point 101.

FIG. 7 is a flowchart of a wafer pre-alignment method according to an embodiment of the present invention. The wafer pre-alignment method of the present application will be described below with reference to FIG. 7 .

Referring to FIG. 7, the fork 100a supports and fixes the wafer 200 and drives the wafer 200 to move to a set position (the set position can be specifically set according to actual conditions), and a camera 400 is used to pair the fork 100a and the wafer 200. An image is taken, the image includes images of all positioning markers 101 arranged on the upper surface of the fork 100 a and an image of the upper surface of the wafer 200 .

It should be noted that since the shooting range of the camera 400 is limited, if the camera 400 vertically shoots the wafer 200 and the wafer fork 100a, it is difficult to photograph the complete surface of the wafer 200 and the periphery of the wafer 200 All the positioning mark points 101, for this purpose, the camera 400 is usually taken obliquely, so that the image taken by the camera 400 is not a standard top plan view of the wafer 200 and the fork 100a, and the center of the upper surface of the fork 100a in the image taken by the camera 400 Does not coincide with the center of the image. In order to obtain the center deviation and rotation angle of the wafer 200 , in this embodiment, the image captured by the camera 400 is converted into a top plan view of the wafer 200 and the wafer fork 100 a.

In this embodiment, based on the actual physical position of the center of the at least three positioning mark points 101 on the sheet fork 100a and the position of the center of each of the positioning mark points 101 in the image captured by the camera 400, the camera 400 takes a picture The image of is converted to a top plan view.

Specifically, the method for converting the image captured by the camera 400 into a top-down plan view includes: performing automatic threshold binarization processing and connected domain area screening on the image captured by the camera 400, so as to determine the position of the positioning marker point 101 on the fork 100a. The position of the center in the image taken by the camera 400; combined with the actual physical position of the center of the at least three positioning mark points 101 on the sheet fork 100a and the position of the center of each positioning mark point 101 in the image taken by the camera 400, A mapping transformation matrix is obtained by calculation; the image captured by the camera 400 is mapped and transformed according to the mapping transformation matrix to form a top plan view of the wafer 200 and the fork 100a.

There may be a set distance between the center of the top plan view and the center of the upper surface of the fork 100a in the top plan view (may be referred to as the center of the fork). For the convenience of calculation, the center of the fork in the top plan view coincides with the center of the top plan view.

It should be noted that, both the automatic threshold binarization processing and connected domain area screening in this embodiment may adopt binarization processing means and connected domain area screening means known in the art.

Before the image captured by the camera 400 is subjected to automatic threshold binarization processing and connected domain area screening, the position parameters (ie coordinates) of the positioning marker point 101 can be read to generate a mask (Mask) to reduce the size of the positioning marker point 101. detection range and improve detection efficiency. After determining the position of the center of the positioning marker point 101 on the fork 100 a in the image captured by the camera 400 , the position parameter of the positioning marker point can also be updated for use in subsequent reading of the position parameter of the positioning marker point 101 .

After obtaining the top plan view, determining the position of the center of the wafer in the top plan view, and determining the position deviation of the center of the wafer from the center of the support assembly.

Specifically, firstly, an automatic threshold binarization process is performed on the top view plan view, and then the area of the wafer 200 in the top view plan view is screened out by the area of the connected domain.

Then, the Ganny algorithm is used to extract the wafer edge information in the top view plan view, and the circle fitting is performed according to the wafer edge information to obtain the position of the wafer center in the top view plan view.

Next, based on the position of the wafer center in the top plan view, the number of deviation pixels (pixels) between the wafer center and the fork center (ie the center of the top plan view) in the top view plan view is obtained, and then calculated according to the pixel size of the top view plan view The position deviation between the center of the wafer and the center of the wafer fork (that is, the center deviation of the wafer) is obtained. For example, the center of the wafer and the center of the chip fork deviate by 100 pixels (pixels) in the top view, and the pixel size of the top view is 0.2 mm (that is, the physical distance corresponding to one pixel is 0.2 mm). At this time, the wafer The center deviation is 20mm.

The position of the wafer center in the top view plan obtained by the circle fitting according to the wafer edge information is rough positioning. After the rough positioning, the projection measurement algorithm can also be used to detect the edge points of the wafer to accurately locate the wafer center. Position in top plan view.

After determining the center of the wafer and obtaining the center deviation of the wafer, determine the apex position of the notch of the wafer in the top plan view, and determine the wafer center and the apex position of the notch based on the top plan view. The rotation angle of the circle.

Specifically, in some embodiments, when determining the apex position of the wafer notch in the top plan view, the distance from the edge point of the wafer to the center of the wafer in the top plan view can be calculated, because the distance between the boundary of the notch and the center of the wafer The distance is smaller than the radius of the wafer 200, and the candidate area of the notch can be screened out according to the known wafer radius and notch depth information. Then, calculate the similarity between the candidate region of the gap and the standard template of the gap, and the position with the highest similarity is the apex position of the gap.

In some embodiments, the method for determining the position of the apex of the notch of the wafer in the top plan view may include: cutting several small pictures from the edge of the wafer in the top plan view; using a feature extraction algorithm (for example, the HOG algorithm) to extract the features; then use machine learning algorithms (supervised and unsupervised learning) to classify the features of each small image and analyze whether there is a gap in each small image, and then use the small image with a gap as a candidate area for the gap; then, from the historical wafer image Select one as the gap standard template, use the geometric template matching algorithm to locate the gap between the small pictures with gaps (can be single or multiple) and the gap standard template, and output the position and confidence of the gap on each small picture; then according to each small picture The corresponding confidence level judges the final position of the notch, and converts the position of the notch into the top view plan view, and then determines the apex position of the notch in the top view plan view. The features may be LBP features, HARR features, HOG features or other effective features that can represent gap information. The machine learning algorithm can be an SVM supervised learning algorithm or an unsupervised learning algorithm.

The method for cutting several small pictures at the edge of the wafer in the top plan view includes: according to the center position of the wafer, cut several small pictures at the edge of the wafer in the top plan view, and the interval of the screenshots is, for example, 0.5°.

In some embodiments, the method for determining the vertex position of the notch of the wafer in the top plan view may include: intercepting several small pictures at the edge of the wafer in the top view plan view; and then using the deep learning target detection algorithm to directly input the small pictures taken Get the position and confidence of the gap on each small picture. The gap may appear in multiple small pictures. Finally, the final position of the gap can be judged by the confidence corresponding to each small picture, and the position of the gap can be converted to the top view plan view , determine the vertex position of the notch in the top plan view.

The deep learning target detection algorithm can be a classic target detection network such as yolo or SSD, or other convolutional neural networks.

In some embodiments, the method for determining the position of the vertex of the notch in the wafer in the top view may include: using a deep learning object detection algorithm, inputting the top view to obtain the position of the notch in the top view, and then determining the position of the notch in the top view. The apex position of the notch in the above top plan view.

In some embodiments, the method for determining the position of the apex of the notch of the wafer in the top plan view may include: using a deep learning target detection algorithm, inputting the image captured by the camera 400 to obtain the position of the notch in the image captured by the camera 400; The position of the notch in the image captured by the camera 400 is converted into the top view plan view, the position of the notch in the top view plan view is obtained, and then the apex position of the notch in the top view plan view is determined.

In some embodiments, the method for determining the position of the vertex of the notch of the wafer in the top plan view may include: performing polar coordinate transformation on the top plan view, and stretching the donut diagram of the edge of the wafer in the top plan view into a rectangular diagram; and , select one from the historical wafer image as the notch standard template; then based on the geometric or pixel grayscale features, search for the region matching the notch standard template in the histogram as the candidate region of the notch, and output the notch on the candidate region Position and confidence level: judge the final position of the gap according to the confidence level corresponding to each of the candidate regions, and convert the position of the gap into the top view plan view, and then determine the vertex position of the gap in the top view plan view.

In some embodiments, the method for determining the apex position of the notch of the wafer in the top plan view may include: searching in the top view plan view, and determining the region matching the notch standard template in the top view plan view as the notch region, Then determine the apex position of the gap in the top plan view.

In some embodiments, the method for determining the apex position of the notch of the wafer in the top plan view includes: searching in the image captured by the camera 400, and determining the region in the image captured by the camera 400 that matches the notch standard template as the notch region ; Convert the position of the notch area in the image captured by the camera 400 to the top plan view, obtain the position of the notch in the top view plan, and then determine the apex position of the notch in the top view plan.

After determining the apex position of the notch, use the center position of the wafer and the apex position of the notch in the plan view to obtain the tilt angle of the line connecting the center of the wafer and the apex of the notch, and then determine the rotation angle of the wafer (this rotation angle is, for example, equal to the The angle of inclination of the line connecting the center of the circle and the vertex of the notch). Wherein, the rotation angle of the wafer 200 can be calibrated by the apex position of the notch.

In some embodiments, the deep learning object detection algorithm can also be used to directly perform gap detection and positioning on the image captured by the camera 400 or the top plan view.

It should be noted that, since the image captured by the camera 400 is obtained, the image is fixed, and then the center deviation and the rotation angle of the wafer 200 are obtained by using the image captured by the camera 400, and the positioning repeatability and measurement repeatability of the wafer are obtained. relatively good.

In this embodiment, after obtaining the center deviation and rotation angle of the wafer 200, the wafer pre-alignment device can perform pre-alignment correction on the wafer 200, so that the The marking of the wafer is within the field of view of the photolithography machine.

In the wafer pre-alignment device and wafer pre-alignment method of the present application, the upper surface of the support assembly (for example, fork 100a) on which the wafer 200 is placed is provided with at least three positioning mark points 101 located on the periphery of the wafer 200, and The image formed by the camera shooting the wafer 200 and the supporting assembly supporting the wafer 200 includes all the positioning mark points 101 on the supporting assembly, and the image can be used to obtain the positional deviation between the center of the wafer and the center of the upper surface of the supporting assembly and obtain the wafer That is to say, the wafer pre-alignment device of this embodiment sets at least three positioning mark points 101 on the upper surface of the support assembly where the wafer 200 is placed, and uses a single image taken by the camera to obtain the orientation of the wafer. The center deviation and the rotation angle do not require the wafer to rotate once, which can reduce the complexity of the mechanism of the wafer pre-alignment device, help to reduce the manufacturing cost of the wafer pre-alignment device and improve the pre-alignment efficiency.

Embodiment two

This embodiment provides a wafer pre-alignment device. The difference from Embodiment 1 is that the supporting component in the wafer pre-alignment device of this embodiment is a pre-alignment stage 100b; the similarities between this embodiment and Embodiment 1 can be referred to Embodiment 1, and will not be repeated here. repeat.

FIG. 5 is a schematic structural diagram of a wafer pre-alignment device according to another embodiment of the present invention. As shown in FIG. 5 , the wafer pre-alignment device of this embodiment includes a support assembly, a light source (such as a surface light source 300 ) and a camera 400 , wherein the support assembly may be a pre-alignment stage 100 b.

FIG. 6 is a schematic plan view of a pre-alignment stage in a wafer pre-alignment device according to an embodiment of the present invention. As shown in FIG. 6, the wafer 200 can be placed on the upper surface of the pre-alignment table 100b, and at least three positioning mark points 101 are arranged on the upper surface of the pre-alignment table 100b and located on the upper surface of the wafer 200. peripheral, and the at least three positioning mark points 101 are located within the shooting range of the camera 400 , the image formed by the camera 400 shooting the wafer 200 and the pre-alignment stage 100 b supporting the wafer 200 includes all the positioning mark points 101 .

In order to facilitate obtaining the center of the upper surface of the pre-alignment stage 100b, the at least three positioning mark points 101 can be evenly arranged at the center of the upper surface of the pre-alignment stage 100b (that is, the center of the upper surface of the pre-alignment stage 100b Geometric center) on the circumference of the circle center.

Several (for example, three) vacuum chucks 102 may also be provided on the upper surface of the pre-alignment stage 100 b, and the vacuum chucks 102 may be used to absorb and fix the wafer 200 .

In this embodiment, the image formed by the camera 400 on the wafer 200 and the pre-alignment stage 100 b includes an image of the wafer 200 and images of the at least three positioning markers 101 . Using the image captured by the camera 400, the method for obtaining the position deviation between the center of the wafer 200 and the center of the upper surface of the pre-alignment stage 100b and the rotation angle of the wafer 200 is the same as or similar to that of Embodiment 1. , which will not be repeated here.

The above description is only a description of the preferred embodiments of the present invention, and is not any limitation to the scope of rights of the present invention. Any person skilled in the art can use the methods and technical contents disclosed above to analyze the present invention without departing from the spirit and scope of the present invention. Possible changes and modifications are made in the technical solution. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention, which do not depart from the content of the technical solution of the present invention, all belong to the technical solution of the present invention. protected range.

Claims

A wafer pre-alignment device, characterized in that, comprising:

A support assembly for supporting the wafer, the upper surface of the support assembly on which the wafer is placed is provided with at least three positioning mark points located on the periphery of the wafer;

a camera for taking an image of the wafer and the support assembly supporting the wafer, wherein the image includes all of the positioning markers on the support assembly; and

The light source is used to provide an illumination light beam so that an illumination light field is formed on the upper surface of the support component when the image is taken.
The wafer pre-alignment device according to claim 1, wherein the shape of each of the positioning mark points is a circle, an ellipse or a polygon.
The wafer pre-alignment device according to claim 1, wherein the distances from the at least three positioning mark points to the center of the upper surface of the support assembly are equal.
The wafer pre-alignment device according to any one of claims 1 to 3, wherein the support component is a fork of a robot or a pre-alignment table.
The wafer pre-alignment device according to claim 4, wherein the at least three positioning mark points are uniformly arranged on a circle centered on the center of the upper surface of the fork or the pre-alignment table.
The wafer pre-alignment device according to claim 1, wherein the light source comprises at least one surface light source, and the light emitting surface of the surface light source faces the support assembly.
The wafer pre-alignment device according to claim 6, wherein the surface light source is arranged opposite to the upper surface of the support assembly; or, the surface light source is arranged opposite to the lower surface of the support assembly and A part of the light emitting surface of the surface light source exceeds the lower surface of the supporting component.
The wafer pre-alignment device according to claim 6, wherein the light source comprises two surface light sources, and the two surface light sources are respectively arranged opposite to the upper surface and the lower surface of the support assembly.
The wafer pre-alignment device according to claim 1, wherein a plurality of vacuum chucks are arranged on the upper surface of the support assembly.
The wafer pre-alignment device according to claim 1, wherein the camera is a CCD camera or a CMOS camera.
A wafer pre-alignment method, characterized in that, comprising:

A support assembly is provided, the support assembly is used to support the wafer, and the upper surface of the support assembly on which the wafer is placed is provided with at least three positioning mark points located on the periphery of the wafer;

a camera taking an image of the wafer and the support assembly supporting the wafer, the image including at least three of the positioning markers on the support assembly;

converting the image to a top plan view based on the actual physical locations of the centers of at least three locator markers in the image on the support assembly and the positions of the centers of each of the locator markers in the image;

determining the position of the center of the wafer in the top plan view, and determining the offset of the center of the wafer from the center of the support assembly; and

An apex position of the notch of the wafer in the top plan view is determined, and a rotation angle of the wafer is determined based on the center of the wafer and the apex position of the notch in the top plan view.
The wafer pre-alignment method according to claim 11, wherein the method for converting the image into a top plan view comprises:

Carrying out automatic threshold value binarization processing and connected domain area screening on the image to determine the position of the center of the positioning marker point on the support component in the image;

calculating and obtaining a mapping transformation matrix based on the actual physical positions of the centers of the at least three positioning markers on the support assembly and the positions of the centers of each of the positioning markers in the image; and

performing mapping transformation on the image according to the mapping transformation matrix to form a top plan view of the wafer and the support assembly.
The wafer pre-alignment method according to claim 11, wherein the center of the top plan view coincides with the center of the support assembly in the top plan view; or, the center of the top plan view coincides with the support assembly in the top view plan view. Component centers have a set spacing.
The wafer pre-alignment method according to claim 11, wherein said determining the position of the center of the wafer in the top plan view, and determining the position deviation between the center of the wafer and the center of the support assembly Methods include:

Carrying out automatic threshold binarization processing on the top view plan view, and then filtering out the area of the wafer in the top view plan view through the connected domain area;

Using the Ganny algorithm to extract wafer edge information in the top view plan view, performing circle fitting according to the wafer edge information, and obtaining the position of the wafer center in the top view plan view; and

Based on the position of the wafer center in the top plan view, the number of deviation pixels between the wafer center and the center of the support assembly in the top view plan is obtained, and then the difference between the wafer center and the center of the support assembly is calculated according to the pixel size of the top view plan view. The positional deviation of the center of the support assembly described above.
The wafer pre-alignment method according to claim 14, wherein the method for determining the position of the apex of the notch of the wafer in the top plan view comprises:

Calculating the distance from the edge point of the wafer to the center of the wafer in the top plan view, and selecting the candidate area of the notch according to the information of the radius of the wafer and the depth of the notch; and

The similarity between the candidate region of the gap and the standard template of the gap is calculated, and the highest similarity is determined as the apex position of the gap.
The wafer pre-alignment method according to claim 14, wherein the method for determining the position of the apex of the notch of the wafer in the top plan view comprises:

Several small pictures are cut from the edge of the wafer in the top plan view;

Using a feature extraction algorithm to extract the features on each of the small pictures;

Using a machine learning algorithm to classify the features of each of the small pictures and analyze whether there is a gap in each of the small pictures, and use the small picture with the gap as a candidate area for the gap;

Using the geometric template matching algorithm to locate the gap between the candidate region of the gap and the gap standard template, and output the position and confidence of the gap on each of the small pictures;

Judging the final position of the notch according to the confidence corresponding to each of the small images, and converting the position of the notch into the top view plan view, and determining the apex position of the notch in the top view plan view.
The wafer pre-alignment method according to claim 14, wherein the method for determining the position of the apex of the notch of the wafer in the top plan view comprises:

Several small pictures are cut from the edge of the wafer in the top plan view;

Using a deep learning target detection algorithm, directly input the intercepted small images to obtain the position and confidence of the gap on each of the small images; and

The final position of the gap is judged by the confidence corresponding to each of the small pictures, and the position of the gap is converted into the top plan view to determine the apex position of the gap in the top plan view.
The wafer pre-alignment method according to claim 14, wherein the method for determining the position of the apex of the notch in the top plan view comprises: using a deep learning target detection algorithm to input the top view plan view , obtaining the position of the notch in the top plan view, and then determining the apex position of the notch in the top plan view;

Alternatively, the method for determining the position of the apex of the notch of the wafer in the top plan view includes: using a deep learning target detection algorithm to input the image to obtain the position of the notch in the image; The position is transformed into the top plan view, the position of the notch in the top view plan is obtained, and then the apex position of the notch in the top view plan is determined.
The wafer pre-alignment method according to claim 14, wherein the method for determining the position of the apex of the notch of the wafer in the top plan view comprises:

performing polar coordinate transformation on the top plan view, and stretching the ring diagram of the wafer edge in the top view plan view into a rectangular map;

Based on geometric or pixel grayscale features, searching for a region matching the gap standard template in the histogram as a candidate region of the gap, and outputting the position and confidence of the gap on the candidate region;

judging the final position of the notch according to the confidence corresponding to each of the candidate regions, and converting the position of the notch into the top plan view; and

A vertex position of the notch in the top plan view is determined.
The wafer pre-alignment method according to claim 14, wherein the method for determining the position of the apex of the wafer notch in the top plan view comprises: searching in the top view plan view, and The area matching the gap standard template in the plan view is determined as the gap area, and then the apex position of the gap in the top plan view is determined;

Alternatively, the method for determining the position of the vertex of the notch of the wafer in the top plan view includes: searching in the image, and determining the region in the image that matches the notch standard template as the notch region; The position of the notch area in the center is transformed into the top view plan view, the position of the notch in the top view plan view is obtained, and then the apex position of the notch in the top view plan view is determined.