A pre-positioning system and method for ultra-thin wafers

Abstract

The application belongs to the technical field of semiconductor devices and integrated circuits, and specifically discloses a pre-positioning system for an ultrathin wafer, which comprises a vacuum adsorption wafer supporting table, an optical imaging module, an image processing module and a motion control module; the vacuum adsorption wafer supporting table is used for loading and adsorbing and fixing a wafer; the optical imaging module comprises a first camera and a second camera; the first camera is used for overall image acquisition to obtain a wafer overall image; and the second camera is used for separate imaging on a wafer edge cutting part to obtain a wafer edge cutting part image. The pre-positioning system for the ultrathin wafer can realize non-damage, high-precision and high-efficiency pre-positioning of the ultrathin wafer through the cooperative operation of the modules, and can provide reliable guarantee for a subsequent edge cutting process, so as to improve the chip yield and production efficiency.

Description

Technical Field

[0001] This invention relates to the field of semiconductor devices and integrated circuit technology, and in particular to a pre-positioning system and method for ultrathin wafers. Background Technology

[0002] In semiconductor manufacturing, wafer dicing is a critical process. Its core purpose is to remove defective areas at the wafer edges and adjust the wafer's outer diameter to provide a substrate that meets precision requirements for subsequent processes such as photolithography and packaging. As a high-value semiconductor material, the precise positioning of the wafer's edge contour directly determines the dicing accuracy, which in turn affects chip yield and production efficiency.

[0003] The traditional positioning method involves first mechanically clamping the wafer to align its center with a rotary table for centering. Then, the rotary table rotates the wafer a full circle, using sensors at the wafer's edge to determine the cutting edge position. The wafer is then moved to this fixed position via the Z-axis, and a second mechanical clamping move it to the final location. Because this positioning method requires mechanical clamping, it is prone to causing irreversible damage such as stress damage and edge chipping to ultra-thin wafers, and it cannot be applied to ultra-thin wafers or wafers with minor edge damage.

[0004] To address the physical damage issues associated with mechanical positioning, visual positioning technology, with its advantages of non-contact operation, high precision, and fast response, is gradually becoming a core technology for ultra-thin wafer edge positioning. Currently, most wafer visual positioning solutions on the market employ a single-camera imaging architecture, whose imaging accuracy cannot meet the high-precision positioning requirements of wafer edge cutting.

[0005] Furthermore, in existing visual positioning solutions, the single imaging and positioning process cannot simultaneously meet the dual requirements of overall wafer centering and precise positioning of the diced area, resulting in a tradeoff between positioning accuracy and efficiency, and making it difficult to adapt to the comprehensive requirements of stability, non-destructiveness, and high precision in the positioning process of ultra-thin wafers.

[0006] In summary, existing pre-positioning systems for ultrathin wafers have limitations in meeting the comprehensive requirements of stability, non-destructiveness, and high precision during the positioning process of ultrathin wafers. Summary of the Invention

[0007] This invention provides a pre-positioning system for ultrathin wafers, which can solve the problem that existing pre-positioning systems for ultrathin wafers are difficult to adapt to the comprehensive requirements of stability, non-destructiveness, and high precision in the positioning process of ultrathin wafers.

[0008] In a first aspect, the present invention provides a pre-positioning system for ultrathin wafers, comprising: A vacuum adsorption wafer support stage, which is used to support and adsorb wafers for fixation; An optical imaging module includes a first camera and a second camera. The first camera performs overall image acquisition to obtain an overall image of the wafer, and the second camera is used to perform separate imaging of the wafer's diced edge to obtain an image of the wafer's diced edge. The image processing module, connected to the optical imaging module, is used to receive an overall image of the wafer and calculate the center position and flat edge rotation angle of the wafer. It is also used to receive an image of the wafer's cut edge and calculate the cut edge position and cut edge angle. The motion control module is connected to the image processing module and the vacuum adsorption support stage respectively. It is used to control the vacuum adsorption support stage to move to the first reference position according to the center position and the flat edge rotation angle. It is also used to control the vacuum adsorption support stage to move to the second reference position according to the cutting edge position and the cutting edge angle. The motion control module is also used to send a positioning ready signal to the slitting device after the wafer is positioned to the second reference position.

[0009] This invention provides a pre-positioning system for ultrathin wafers, which, compared to existing technologies, has, but is not limited to, the following advantages: In this pre-positioning system for ultra-thin wafers, the vacuum adsorption stage serves as the foundation for supporting and fixing the wafer, enabling small-area contact fixation of the ultra-thin wafer and avoiding stress damage and edge chipping caused by physical clamping. It is also suitable for wafers with slight burrs or damage on the edges. The optical imaging module is the core foundation for ensuring positioning accuracy. It adopts a dual-camera collaborative imaging architecture to address the technical shortcomings of existing single-camera solutions, achieving the goal of moving from overall preliminary positioning to precise local positioning. The image processing module, as the core computing unit of the system, realizes the conversion from image data to positioning parameters through a step-by-step algorithm process of preprocessing, edge detection, feature extraction, and positioning calculation. The motion control module is the execution component that realizes precise wafer displacement. It is connected to the image processing module and the vacuum adsorption stage, and also interacts with the edge trimming device to form a complete positioning-edge trimming collaborative process.

[0010] This invention provides a pre-positioning system for ultrathin wafers, employing a dual-camera imaging, vacuum adsorption wafer support, and step-by-step positioning control scheme. Compared to existing technologies, this system offers significant advantages: First, vacuum adsorption completely solves the positioning damage problem of ultrathin wafers and wafers with edge breakage, improving system adaptability. Second, the dual cameras work collaboratively, achieving overall centering through the first camera and precise local positioning through the second camera, balancing positioning efficiency and accuracy. Compared to a large field-of-view, high-resolution single-camera solution, this significantly reduces equipment procurement costs and installation space requirements, adapting to the overall size requirements of proximity lithography machines. Third, through closed-loop linkage of image processing and motion control, combined with anti-interference algorithm design, the system ensures a significantly improved positioning success rate in cleanroom environments, even when faced with interference such as wafer reflection and minor contamination. It also supports rapid adaptation to wafers of various materials and specifications, enhancing system versatility. The present invention provides a pre-positioning system for ultra-thin wafers. Through the coordinated operation of various modules, it can achieve pre-positioning of ultra-thin wafers without damage, with high precision and high efficiency, providing a reliable guarantee for subsequent edge cutting processes, thereby improving chip yield and production efficiency.

[0011] Furthermore, the optical imaging module also includes a backlight source, which is used to illuminate from the back of the wafer to highlight the wafer edge contour.

[0012] Furthermore, the pre-positioning system for ultrathin wafers also includes a bottom support, on which the motion control module is mounted.

[0013] Furthermore, the motion control module includes an X-axis drive mechanism, a Y-axis drive mechanism located at the output end of the X-axis drive mechanism, and an R-axis drive mechanism located at the output end of the Y-axis drive mechanism, wherein the X-axis drive mechanism is located on the bottom support. The vacuum adsorption support stage is located at the output end of the R-direction drive mechanism; The X-axis driving mechanism is used to drive the Y-axis driving mechanism to move along the first direction, the Y-axis driving mechanism is used to drive the R-axis driving mechanism to move along the second direction, and the R-axis driving mechanism is used to drive the vacuum adsorption substrate stage to rotate.

[0014] Furthermore, the first direction is perpendicular to the second direction.

[0015] Furthermore, an optical support is provided on the bottom support, and the optical imaging module is located on the optical support.

[0016] Furthermore, both the first camera and the second camera are industrial-grade monochrome CMOS area scan cameras.

[0017] Furthermore, the distance between the first reference position and the preset reference position is no greater than 300um.

[0018] Furthermore, the distance between the second reference position and the preset reference position is no greater than 60 μm.

[0019] Secondly, the present invention also provides a pre-positioning method for a pre-positioning system for ultrathin wafers, using a pre-positioning system for ultrathin wafers provided in the first aspect of the present invention, the method comprising the following steps: S1: The robotic arm places the wafer on the vacuum adsorption wafer support stage, which adsorbs and fixes the wafer. S2: The first camera captures an overall image of the wafer, obtains the overall edge contour image of the wafer, and transmits it to the image processing module; S3: The image processing module processes the overall edge contour image, extracts the wafer edge curve, calculates the wafer's center position and flat edge rotation angle, and sends the calculation results to the motion control module; S4: The motion control module drives the vacuum adsorption plate holder to move to the first reference position according to the center position and the flat edge rotation angle; S5: The second camera separately images the wafer dicing edge, acquires the shape, position, and angle of the dicing edge, and transmits it to the image processing module; S6: The image processing module performs depth processing on the image of the cut edge shape position and angle, calculates the cut edge position and cut edge angle of the wafer, and sends the calculation results to the motion control module; S7: The motion control module drives the vacuum adsorption support stage to move to the second reference position according to the cutting edge position and cutting edge angle; S8: The motion control module sends a positioning ready signal to the edge cutting device. After the edge cutting is completed, the edge cutting device sends a reset signal to the motion control module. After receiving the reset signal from the edge cutting device, the motion control module drives the vacuum adsorption support stage back to the initial position, completing the single pre-positioning process. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a pre-positioning system for ultrathin wafers provided by the present invention; Figure 2 The main structural view of a pre-positioning system for ultrathin wafers provided by the present invention; Figure 3 A schematic diagram of the installation structure of the motion control module and the bottom support of a pre-positioning system for ultra-thin wafers provided by the present invention; Figure 4This is a schematic diagram of the motion control module of a pre-positioning system for ultrathin wafers provided by the present invention.

[0021] Explanation of reference numerals in the attached figures: 1. Optical imaging module; 2. Motion control module; 100. Vacuum adsorption stage; 101. First camera; 102. Second camera; 103. Backlight; 200. Bottom support; 201. X-axis drive mechanism; 202. Y-axis drive mechanism; 203. R-axis drive mechanism; 204. Optical support. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings showing multiple embodiments according to this application. It should be understood that the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments described in this application without creative effort will fall within the scope of protection of this application.

[0023] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used in the description of this application is for the purpose of describing specific embodiments only and is not intended to limit this application; the terms "comprising," "including," "having," "containing," etc., in the description, claims, and accompanying drawings of this application are open-ended terms. Therefore, "comprising," "including," or "having" refers to, for example, a method or apparatus having one or more steps or elements, but is not limited to having only these one or more elements. The terms "first," "second," etc., in the description, claims, or accompanying drawings of this application are used to distinguish different objects, not to describe a specific order or hierarchy. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0024] In the description of this invention, it should be understood that the terms "upper", "lower", "left", "right", "front", "rear", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0025] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "attachment" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0026] It should be emphasized that when the term "comprising / including" is used in this specification, it is used to explicitly indicate the presence of the stated feature, integer, step, or component, but does not exclude the presence or addition of one or more other features, integers, steps, parts, or groups of features, integers, steps, or parts.

[0027] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, B and / or C can represent: B existing alone, B and C existing simultaneously, or C existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0028] like Figures 1 to 4 As shown, an embodiment of the present invention provides a pre-positioning system for ultrathin wafers, comprising: Vacuum adsorption wafer support stage 100 is used to support and adsorb wafers for fixing. Optical imaging module 1 includes a first camera 101 and a second camera 102. The first camera 101 performs overall image acquisition to obtain an overall image of the wafer, and the second camera 102 is used to perform separate imaging of the wafer diced edge to obtain an image of the wafer diced edge. The image processing module, connected to the optical imaging module 1, is used to receive an overall image of the wafer and calculate the center position and the rotation angle of the flat edge of the wafer. It is also used to receive an image of the wafer's cut edge and calculate the cut edge position and the cut edge angle of the wafer. The motion control module 2 is connected to the image processing module and the vacuum adsorption support stage 100 respectively. It is used to control the vacuum adsorption support stage 100 to move to the first reference position according to the center position and the flat edge rotation angle. It is also used to control the vacuum adsorption support stage 100 to move to the second reference position according to the cutting edge position and the cutting edge angle. The motion control module 2 is also used to send a positioning ready signal to the edge cutting device after the wafer is positioned to the second reference position.

[0029] Specifically, the vacuum adsorption stage 100 serves as the base for supporting and fixing the wafer, enabling small-area contact fixation of ultra-thin wafers. This avoids stress damage and edge chipping caused by physical clamping, and is also suitable for wafers with slight burrs or breakage on the edges. Driven by the motion control module 2, the stage can perform translational and rotational movements. Its adsorption function is achieved through vacuum negative pressure, which can stably fix ultra-thin wafers of different specifications. Furthermore, the adsorption force can be dynamically adjusted according to the wafer material and thickness to prevent excessive adsorption force from causing wafer deformation. In addition, the stage's structural design is adapted to the installation requirements of a hollow parallel backlight, providing space for clear imaging by the optical imaging module 1 and ensuring that the wafer edge contours can be effectively captured.

[0030] The optical imaging module 1 is the core foundation for ensuring positioning accuracy. It employs a dual-camera collaborative imaging architecture to specifically address the technical shortcomings of existing single-camera solutions, achieving the goal of moving from overall preliminary positioning to precise local positioning. The first camera 101 is an industrial-grade monochrome CMOS area array camera with a resolution ≥ 5 million pixels (2600×2128), pixel size ≤ 2.5μm, and is equipped with an FA process-grade lens. Its working distance is adjustable from 50-200mm, and its field of view is adjustable from 50-250mm. It features global shutter technology to avoid image ghosting during wafer movement and quickly acquire images of the entire wafer edge contour. The camera's wide field of view design can completely cover wafers ranging from 2 to 6 inches, providing comprehensive image data support for subsequent calculations of the center position and the flat edge rotation angle.

[0031] The second camera 102 also employs an industrial-grade monochrome CMOS area array camera, maintaining consistency with the first camera selection to simplify system adaptation. Its core function is to perform local high-definition imaging of the wafer's diced edge, focusing on the flat edge or notch features of the diced area to obtain precise information on the diced edge's shape, position, and angle. Compared to the first camera, the second camera achieves a small field of view focusing through lens parameter adjustment, highlighting the detailed features of the diced edge and effectively resisting interference factors such as wafer surface reflections and minor stains. This provides high-quality local images for the depth calculation of the image processing module, ensuring the accuracy of the diced edge's position and angle calculations. Furthermore, the optical imaging module 1 can be paired with a hollow parallel backlight. Through light homogenization and filtering, the emitted light is made nearly parallel, illuminating the wafer in the opposite direction of the lens optical axis. This eliminates interference from surface reflections on imaging and enhances the grayscale difference between the wafer edge and the background, further improving image clarity.

[0032] The image processing module, as the core computing unit of the system, realizes the transformation from image data to positioning parameters through a step-by-step algorithm process of preprocessing, edge detection, feature extraction, and localization calculation. It establishes data transmission paths with the first camera 101 and the second camera 102 respectively, receives the acquired image information in real time, and performs targeted processing. For the overall wafer image transmitted by the first camera 101, the image processing module can first perform preprocessing operations such as distortion correction, grayscale enhancement, noise filtering, and ROI region extraction to eliminate interference caused by lens distortion, image noise, and uneven light source. Then, the wafer edge curve is extracted by multi-threshold edge detection and sub-pixel positioning technology. The effective edge points are fitted with a circle using the iterative weighted least squares method to calculate the wafer's center coordinates (X0, Y0), i.e., the center position, and the flat edge rotation angle. At the same time, the calculation results are verified to ensure that the center position error is within a reasonable range, providing accurate parameters for preliminary positioning.

[0033] For the wafer edge image transmitted by the second camera 102, the image processing module performs in-depth optimization based on the above preprocessing operations. It removes edge noise and repairs broken edges through morphological operations, accurately extracts the flat edge or notch features of the edge, and calculates the coordinates of the edge position and the edge angle θ using a preset mask reference position. This determines the distance and rotation angle that the wafer needs to be further translated, providing parameter support for precise positioning. This module has strong anti-interference capabilities and can filter out the influence of minor defects such as slight chipping at the wafer edge through threshold filtering, ensuring the stability of the positioning calculation results. It also supports rapid parameter switching to adapt to the positioning needs of wafers with different materials and manufacturing processes.

[0034] The motion control module 2 is the actuator for precise wafer displacement. It connects to the image processing module and the vacuum adsorption stage 100, and also interacts with the dicing equipment to form a complete positioning-dicing coordinated process. This module can use an EtherCAT bus controller and can be equipped with a servo motor and a linear encoder. Closed-loop control ensures the accuracy and speed of position adjustment. The linear encoder can acquire the actual position of the wafer in real time and compare it with the target position. The deviation is corrected by a PID algorithm to ensure that the adjustment error is ≤ ±1μm.

[0035] In the positioning process, motion control module 2 first receives the center position and flat edge rotation angle parameters output by the image processing module, and drives the vacuum adsorption wafer stage 100 to translate along the XY direction and rotate around the R axis, moving the wafer to the first reference position, so that the deviation between the wafer center and the preset reference position is controlled within ±300μm, completing the initial centering. Then, it receives the dicing position and dicing angle parameters, and drives the wafer stage to perform secondary fine-tuning, moving the wafer to the second reference position, reducing the deviation between the wafer center and the preset reference position to within ±60μm, achieving precise positioning. When the wafer reaches the second reference position, motion control module 2 sends a positioning ready signal to the dicing equipment, triggering the dicing process. After dicing is completed, motion control module 2 receives a reset signal from the dicing equipment, and drives the vacuum adsorption wafer stage 100 back to the initial position, preparing for the positioning process of the next wafer. Through the above process, the single wafer positioning cycle can be ensured to be ≤6s, meeting the high-speed production rhythm of ≥180 wafers per hour.

[0036] The innovation of this application lies in its use of a dual-camera imaging, vacuum adsorption wafer support, and step-by-step positioning control scheme, which has significant advantages over existing technologies: First, vacuum adsorption completely solves the positioning damage problem of ultra-thin wafers and wafers with broken edges, improving system adaptability; Second, the dual cameras work together, achieving overall centering through the first camera and precise local positioning through the second camera, balancing positioning efficiency and accuracy. Compared with a large field of view and high-resolution single-camera solution, it significantly reduces equipment procurement costs and installation space requirements, adapting to the overall size requirements of proximity lithography machines; Third, through closed-loop linkage of image processing and motion control, combined with anti-interference algorithm design, it ensures a positioning success rate of ≥99.9% in cleanroom environments, even when faced with interference such as wafer reflection and minor stains. It also supports rapid adaptation to wafers of multiple materials and specifications, improving system versatility.

[0037] The working principle of this application embodiment is as follows: A robotic arm places a 2-6 inch ultrathin wafer onto a vacuum adsorption stage 100. The stage uses vacuum negative pressure to adsorb and fix the wafer, avoiding physical contact damage. Then, a first camera 101 captures an overall image of the wafer, transmitting the overall edge contour image to an image processing module. The image processing module performs preprocessing, edge extraction, and circle fitting calculations to obtain the center position and flat edge rotation angle, sending this information to a motion control module 2. The motion control module 2 drives the stage to move the wafer to a first reference position, completing initial centering. Next, a second camera 102 images the wafer's diced edge separately, transmitting the local image to the image processing module. Depth processing calculates the diced edge position and angle, feeding this information back to the motion control module 2. The motion control module 2 drives the stage for secondary fine-tuning, moving the wafer to a second reference position for precise positioning. Finally, the motion control module 2 sends a positioning ready signal to the dicing device. After dicing is complete, it receives a reset signal, driving the stage to reset, completing a single pre-positioning process.

[0038] This pre-positioning system for ultra-thin wafers aims to solve the damage problem caused by traditional mechanical clamping and positioning, as well as the technical bottleneck of existing single-camera vision positioning solutions that struggle to balance accuracy and cost-effectiveness. It is suitable for edge-cutting pre-positioning scenarios of 2-6 inch ultra-thin wafers and is compatible with wafers of different materials and processes, including silicon-based, silicon carbide, and gallium nitride, meeting the environmental requirements of semiconductor cleanrooms (Class 100). Through the coordinated operation of its modules, it achieves damage-free, high-precision, and high-efficiency pre-positioning of ultra-thin wafers, providing reliable assurance for subsequent edge-cutting processes, thereby improving chip yield and production efficiency.

[0039] like Figures 1 to 4 As shown, in some embodiments of the present invention, the optical imaging module 1 further includes a backlight 103, which is used to illuminate from the back of the wafer to highlight the wafer edge contour.

[0040] Specifically, the backlight 103 is a key auxiliary component in the optical imaging module 1 that ensures imaging quality. It adopts a hollow parallel backlight design to adapt to the structural layout of the vacuum adsorption stage 100. Its hollow area can provide sufficient space for the XYR three-axis movement of the stage and the wafer, avoiding motion interference, while ensuring that the light source can achieve full coverage illumination from the back of the wafer.

[0041] More specifically, the backlight 103 is composed of LED beads, a diffuser, and a parallel film: the LED beads are arranged in a reasonable manner, and the diffuser and parallel film structure are used to make the light emitted by the beads uniform after being processed by the diffuser, and then filtered by the parallel film to form parallel light that is close to the ideal state. This light illuminates the wafer in the opposite direction to the optical axis of the first camera 101 and the second camera 102, eliminating the interference of wafer surface reflection on imaging from the source. It is especially suitable for the imaging needs of wafers made of semi-transparent or highly reflective materials such as silicon-based and silicon carbide.

[0042] In detail, the core function of the backlight 103 is to enhance the grayscale difference between the wafer edge and the background, making the edge contours clearly discernible and laying the foundation for accurate calculations by the image processing module. Its brightness supports stepless adjustment from 0-100%, dynamically adapting to the wafer material, thickness, and surface condition. Even with semi-transparent wafers, significant contrast can be achieved through brightness adjustment, ensuring effective recognition of edge features. Simultaneously, the backlight 103 utilizes a highly stable LED light source, offering uniform luminous intensity and a long lifespan. It can withstand the harsh environment of semiconductor cleanrooms (Class 100), resisting the effects of slight dust and temperature / humidity fluctuations on light source performance, ensuring consistent imaging over long-term operation.

[0043] When the first camera 101 captures the overall image, the backlight 103 provides uniform parallel light illumination to ensure that the overall edge contour of the wafer is fully presented and to avoid edge omissions due to uneven illumination. When the second camera 102 captures a local area, the strong light is focused on the chamfered area to highlight the flat edge or notch feature details. Combined with the high-resolution imaging of the camera, the image processing module can accurately extract the chamfered position and angle information.

[0044] like Figures 1 to 4 As shown, in some embodiments of the present invention, the pre-positioning system for ultrathin wafers further includes a bottom support 200, and the motion control module 2 is disposed on the bottom support 200.

[0045] Specifically, the bottom bracket 200 is the fundamental load-bearing component of the entire pre-positioning system, providing a stable mounting platform and support foundation for the motion control module 2 and other related components. It can be manufactured from a single piece of high-performance aluminum alloy, possessing sufficient structural rigidity and load-bearing strength to stably support the weight of the motion control module 2 and the inertial forces generated during movement, while effectively reducing the overall equipment weight to meet the size and weight requirements of the proximity lithography machine. In practical applications, the structural shape of the bottom bracket 200 can be adapted to the installation dimensions of the motion control module 2 and the overall equipment layout requirements. It typically adopts a frame or flat plate structure, with pre-drilled mounting holes at the bottom to facilitate fixing the entire pre-positioning system to the designated workstation of the lithography machine, achieving precise docking with the entire machine.

[0046] In detail, mounting the motion control module 2 on the bottom bracket 200 aims to improve the accuracy and stability of motion control through a stable foundation, thus resolving positioning deviations caused by unstable mounting references during movement. Specifically, the bottom bracket 200 ensures the flatness and levelness of the mounting surface through precise machining. Furthermore, the bottom bracket 200 also serves to distribute stress and provide shock absorption, effectively absorbing vibrations generated during equipment operation and reducing the impact of external vibrations on the precision components of the motion control module 2.

[0047] like Figures 1 to 4 As shown, in some embodiments of the present invention, the motion control module 2 includes an X-axis drive mechanism 201, a Y-axis drive mechanism 202 disposed at the output end of the X-axis drive mechanism 201, and an R-axis drive mechanism 203 disposed at the output end of the Y-axis drive mechanism 202. The X-axis drive mechanism 201 is disposed on the bottom support 200. The vacuum adsorption plate support stage 100 is located at the output end of the R-direction drive mechanism 203; The X-axis drive mechanism 201 is used to drive the Y-axis drive mechanism 202 to move along the first direction, the Y-axis drive mechanism 202 is used to drive the R-axis drive mechanism 203 to move along the second direction, and the R-axis drive mechanism 203 is used to drive the vacuum adsorption substrate stage 100 to rotate.

[0048] Specifically, the X-axis drive mechanism 201, Y-axis drive mechanism 202, and R-axis drive mechanism 203 are the core execution components of the motion control module 2 for realizing the XYR three-axis motion of the wafer. The X-axis drive mechanism 201, as the basic drive unit, can be composed of a precision ball screw slide and a servo motor. Through the precise positioning and installation of the bottom bracket 200, its motion trajectory is ensured to be completely consistent with the preset first direction, providing a benchmark for overall displacement adjustment. The Y-axis drive mechanism 202 is rigidly connected to the output end of the X-axis drive mechanism 201, and its structural selection is adapted to the X-axis drive mechanism 201. Its motion direction (second direction) is perpendicular to the first direction, which can drive the subsequent R-axis drive mechanism 203 and the vacuum adsorption wafer stage 100 to achieve lateral displacement adjustment. The R-axis drive mechanism 203 adopts a high-precision rotary platform with a torque motor design, which has the advantages of good low-speed stability and high positioning accuracy. Its output end is fixedly connected to the vacuum adsorption wafer stage 100, which can drive the wafer stage to achieve 360° stepless rotation, meeting the adjustment requirements of the wafer flat edge rotation angle and shaving angle.

[0049] More specifically, in the attached diagram, the first direction refers to the longitudinal direction, that is, the front-to-back position, and the second direction refers to the transverse direction, that is, the left-to-right position.

[0050] In traditional positioning methods, there are only two axes, Z and R. Pre-alignment is achieved through two clamping processes: first, the wafer is mechanically clamped once to align the wafer's center with the rotary table for centering; then, the rotary table rotates the wafer once, and the edge position is determined by a sensor at the wafer's edge. The Z-axis is then moved to the fixed edge position, and the wafer is mechanically clamped a second time to move it to the fixed position.

[0051] In this application, a visual algorithm module (optical imaging module + image processing module) + motion XYR axis replaces the mechanical clamping structure, realizing linkage positioning of four axes (XYZR), completely eliminating the traditional two-stage mechanical clamping, fundamentally avoiding damage to ultra-thin wafers caused by mechanical clamping, and greatly improving the system's adaptability to wafers in different states.

[0052] In detail, the combined design of this three-stage drive mechanism achieves precise control of wafer position and angle through modular division of labor, while ensuring the independence and coordination of each movement direction. Specifically, the X-axis drive mechanism 201, based on the stable support of the bottom bracket 200, drives the ball screw through a servo motor to move the Y-axis drive mechanism 202 smoothly along the first direction, realizing the wafer's position calibration in the front-back direction; the Y-axis drive mechanism 202 simultaneously responds to control commands, driving the R-axis drive mechanism 203 and the wafer stage to move along the second direction, completing the fine adjustment of the wafer's displacement in the left-right direction. The coordinated action of the two can quickly align the wafer center with the first reference position. The R-axis drive mechanism 203, based on the angle parameters output by the image processing module, drives the vacuum adsorption wafer stage 100 to rotate precisely, realizing the calibration of the wafer's flat edge rotation angle and dicing angle, ensuring that the dicing position matches the preset reference.

[0053] The series design of the three-stage drive mechanism described above can also optimize the flexibility and precision control of motion control: In the initial centering stage, the X and Y drive mechanisms 202 respond quickly to commands, driving the wafer to achieve a large range of displacement, controlling the center deviation within ±300μm; In the fine positioning stage, the three work together to fine-tune, with the X and Y drive mechanisms 202 achieving micron-level displacement compensation and the R drive mechanism 203 completing precise angle calibration, ultimately reducing the overall deviation to within ±60μm.

[0054] like Figures 1 to 4 As shown, in some embodiments of the present invention, the first direction is perpendicular to the second direction.

[0055] In the attached diagram, the first direction refers to the longitudinal direction, i.e., the front-to-back position, and the second direction refers to the transverse direction, i.e., the left-to-right position. The first and second directions are arranged perpendicularly at 90°, forming a standard two-dimensional Cartesian coordinate system motion plane. This vertical design simplifies the motion control logic to the greatest extent, ensuring that the wafer's transverse and longitudinal displacement adjustments do not interfere with each other. In actual operation, the X-axis drive mechanism 201 and the Y-axis drive mechanism 202 can independently and precisely control the wafer's movement in the two vertical directions, greatly improving positioning accuracy and efficiency. Simultaneously, this vertical layout also helps optimize the overall structure of the equipment, reducing space occupation and making the entire pre-positioning system more compact and rational. It better adapts to the overall layout of proximity lithography machines, meeting the requirements of semiconductor production for equipment miniaturization and integration. Moreover, the vertical design is mechanically more stable, better able to withstand various forces generated during wafer movement, ensuring smooth and reliable movement, and further improving the overall performance of the system.

[0056] In detail, the core of the perpendicular design in the first and second directions is to simplify positioning calculations and improve the accuracy and efficiency of displacement adjustment through an orthogonal motion architecture, solving problems such as motion coupling and complex positioning algorithms caused by non-orthogonal directions. Specifically, the orthogonal layout ensures that the X-axis and Y-axis drive mechanisms do not interfere with each other, each independently completing the displacement adjustment of its corresponding axis. The image processing module can output the center offset ΔX and ΔY, which can then be directly converted into motion commands for the two drive mechanisms without complex coordinate conversions, significantly reducing the computational load of the control algorithm and achieving a positioning cycle of ≤6s per wafer. At the same time, the perpendicular relationship ensures that the wafer can be quickly reached from any position in the XY plane through the coordinated action of the two drive mechanisms, avoiding redundant positioning actions caused by directional deviations and further improving positioning efficiency.

[0057] like Figures 1 to 4 As shown, in some embodiments of the present invention, an optical support 204 is provided on the bottom support 200, and the optical imaging module 1 is disposed on the optical support 204. Specifically, the optical imaging module 1 is disposed above the motion control module 2.

[0058] Specifically, the optical bracket 204 serves as the dedicated mounting platform for the optical imaging module 1, and its design must balance structural stability with the requirements of the imaging optical path. This bracket is made of lightweight aluminum alloy, effectively reducing its weight while ensuring sufficient rigidity, thus avoiding additional load on the bottom bracket 200. The top of the bracket body has an adjustable mounting slot, allowing for height adjustment of the first camera 101 and the second camera 102 in the Z-axis direction (i.e., the vertical direction) via a micrometer head. The adjustment accuracy can reach ±0.01mm, ensuring that the optical axes of the two cameras are perpendicular to the surface of the vacuum adsorption stage 100.

[0059] In terms of optical path layout, the bottom of the bracket and the bottom bracket 200 are precisely aligned, and the connection between the bottom of the bracket and the bottom bracket 200 can be ensured by tightening screws. In particular, a tilt adjustment mechanism can be provided at the mounting position of the second camera 102, which can realize the angle fine adjustment of the camera lens within ±5° through worm gear transmission, to compensate for the image field tilt caused by assembly errors of the optical system, and ensure that the image in the cropped area is always at the optimal focal plane.

[0060] In detail, this design places the optical imaging module 1 directly above the motion control module 2, allowing the parallel light emitted by the backlight 103 to directly penetrate the hollow area of ​​the vacuum adsorption stage 100, and after reflection from the wafer edge, enter the camera lens to form a clear bright-field imaging effect. Simultaneously, the optical support 204 maintains a safe distance from the motion control module 2 to avoid mechanical interference during X / Y / R axis movement, ensuring stable imaging quality within the maximum travel range of the stage.

[0061] like Figures 1 to 4 As shown, in some embodiments of the present invention, the first camera 101 and the second camera 102 are both industrial-grade monochrome CMOS area array cameras.

[0062] Specifically, the industrial-grade monochrome CMOS area scan camera is the core component adapted to the high-precision imaging requirements of this system. The first camera 101 and the second camera 102 are selected with the same model, both using domestically produced GMAX chips, ensuring consistent and compatible imaging performance. Both cameras are equipped with ≥5 megapixel (2600×2128) resolution, ≤2.5μm pixel size, and ≥20fps frame rate. Combined with global shutter technology, they effectively avoid image blurring during wafer movement, ensuring the clarity and integrity of dynamically acquired images and providing high-quality data input for subsequent image processing. The monochrome CMOS area scan camera eliminates the need for color filtering and conversion processing, resulting in faster imaging response, reduced color interference, and enhanced grayscale differences between wafer edges and the background, making it more suitable for edge contour recognition scenarios on semiconductor wafers.

[0063] In detail, the two similar industrial-grade monochrome CMOS area array cameras work together. The first camera, 101, leverages its wide field-of-view imaging advantage to quickly acquire images of the overall edge contour of the wafer. With its high resolution and small pixel size, it accurately captures the subtle features of the wafer edge. The second camera, 102, focuses on the local area of ​​the cut edge. Through consistent imaging performance between the two cameras, it ensures the consistency of grayscale response between the local and overall images, reducing the complexity of parameter calibration during image processing and improving the efficiency of positioning calculations. Simultaneously, the industrial-grade materials and CMOS area array structure give the cameras strong environmental adaptability, allowing them to stably adapt to the temperature, humidity, and dust-free requirements of semiconductor cleanrooms (Class 100), and preventing performance degradation over long-term operation.

[0064] In some embodiments of the present invention, the distance between the first reference position and the preset reference position is no greater than 300 μm.

[0065] Specifically, the preset reference position is the initial wafer centering target position preset by the system. This preset reference position corresponds to the initial alignment reference of the dicing equipment. The first reference position is the actual position reached by the wafer after initial centering. The distance between the two is controlled within the range of ≤300μm, which is an optimized threshold that balances the efficiency of initial centering and the accuracy of subsequent fine positioning. This distance threshold is achieved through feedback from the motion control module 2 and precise calculation by the image processing module. Relying on the micron-level adjustment capability of the X and Y axis drive mechanisms, and combined with the real-time position feedback of the grating ruler, it ensures that the position deviation of the wafer after initial centering does not exceed the preset range, reserving reasonable fine-tuning space for subsequent fine positioning processes.

[0066] In detail, if the spacing threshold is too large, it will increase the amount of fine-tuning required in the subsequent precision positioning stage, prolonging the single-wafer positioning cycle and failing to meet the high-speed production demand of ≥180 wafers per hour. If the spacing threshold is too small, it will significantly increase the difficulty of calculation and adjustment for initial centering, increase the computational load on the equipment, and make it susceptible to wafer surface interference and imaging errors, leading to an increased centering failure rate. Controlling the spacing to ≤300μm allows for efficient completion of initial centering through the overall image acquisition of the first camera 101 and the rapid response of the drive mechanism, while keeping the deviation within the range that the precision positioning module can quickly compensate for. This ensures that subsequent fine-tuning through the coordinated fine-tuning of the X, Y, and R drive mechanisms can quickly reduce the deviation to the precision positioning standard of ≤60μm.

[0067] In some embodiments of the present invention, the distance between the second reference position and the preset reference position is no greater than 60 μm.

[0068] Specifically, the second reference position is the final target position reached by the wafer after fine positioning, and the preset reference position is the initial centering target position of the wafer preset by the system. This preset reference position corresponds to the initial alignment reference of the dicing equipment, and the distance between the two is controlled within the range of ≤60μm, which is the core threshold to meet the dicing accuracy requirements of ultra-thin wafers. This threshold is achieved by the step-by-step positioning from the initial centering to the fine positioning of the system. Based on the aforementioned first reference position deviation of ≤300μm, through the micron-level fine adjustment of the X and Y axis drive mechanisms and the precise angle calibration of the R axis drive mechanism, combined with the closed-loop feedback of the grating ruler and the correction of the PID algorithm, the final position deviation is locked within the preset range, providing high-precision substrate guarantee for subsequent dicing processes.

[0069] In detail, a 60μm threshold is the optimal choice for balancing edge-cutting accuracy and positioning efficiency: if the threshold is too large, the wafer edge-cutting position deviation will exceed the process standard, directly affecting chip yield; if the threshold is too small, repeated fine-tuning and calibration are required, significantly extending the single-wafer positioning cycle and failing to meet the high-speed production requirements of ≥180 wafers per hour. By controlling the spacing to ≤60μm, the edge-cutting features can be accurately captured and adjustment parameters output through the local high-definition imaging of the second camera 102 and the depth calculation of the image processing module, while the high-speed response capability of the motion control module can quickly complete position and angle compensation, achieving a balance between accuracy and efficiency.

[0070] This invention also provides a pre-positioning method for a pre-positioning system of ultra-thin wafers. It employs a pre-positioning system for ultra-thin wafers provided in the first aspect of this invention. Specifically, it uses a vacuum adsorption stage 100 to support the wafer, employs dual-camera imaging, and combines an image processing module and a motion control module 2 to achieve a process from initial centering to precise positioning. Ultimately, it achieves non-destructive, high-precision, and high-efficiency pre-positioning of ultra-thin wafers. This method is adaptable to the positioning requirements of 2-6 inch ultra-thin wafers of different materials and processes, meeting the needs of semiconductor cleanroom (Class 100) environments and high-speed production rates of ≥180 wafers per hour. The core principle of this method is: dual cameras acquire overall and partial images of the wafer; the image processing module extracts precise positioning parameters; the motion control module drives the vacuum adsorption stage to complete a stepped position calibration; and finally, the wafer is precisely aligned to a preset reference, providing a reliable substrate for the edge-cutting process. Simultaneously, signal linkage between modules achieves automated closed-loop process control.

[0071] Specifically, the pre-positioning method for the pre-positioning system of ultrathin wafers includes the following steps: S1: The robotic arm places the wafer on the vacuum adsorption stage 100, and the vacuum adsorption stage 100 adsorbs and fixes the wafer. In step S1, the robotic arm smoothly places the ultrathin wafer to be positioned onto the bearing surface of the vacuum adsorption stage 100 according to a preset trajectory. During placement, hard contact with the wafer surface and edges is avoided to prevent secondary damage to the ultrathin wafer or areas with slight burrs on the edges. After the wafer is placed in place, the vacuum adsorption stage 100 activates its vacuum adsorption function, using negative pressure to tightly adsorb and fix the wafer. The adsorption force can be dynamically adjusted according to the wafer material (silicon-based, silicon carbide, gallium nitride, etc.) and thickness to ensure that the wafer is firmly fixed without any wafer deformation.

[0072] In this step, relying on the small-area contact and fixing characteristics of the vacuum adsorption stage 100 to the wafer, the risk of damage to ultra-thin wafers and wafers with broken edges by physical contact is avoided from the source. At the same time, it provides a stable bearing foundation for subsequent image acquisition and position adjustment, ensuring that the wafer does not shift or shake during the positioning process.

[0073] S2: The first camera 101 captures an overall image of the wafer, obtains the overall edge contour image of the wafer, and transmits it to the image processing module; The specific process of step S2 is as follows: After the wafer is fixed, the optical imaging module 1 is activated. The optical bracket 204 on the bottom support 200 provides a stable mounting reference for the first camera 101, ensuring that the first camera 101 is located at a preset height above the motion control module 2 and maintains a certain working distance from the wafer. At the same time, the backlight 103 is activated to illuminate from the back of the wafer, emitting uniform parallel light through the hollow parallel structure to eliminate reflection interference on the wafer surface and enhance the grayscale difference between the edges and the background. The first camera 101 (an industrial-grade monochrome CMOS area scan camera) activates the global shutter to capture an overall image of the wafer, quickly capturing an image of the entire edge contour covering the full size of the wafer. After acquisition, the image data is immediately transmitted to the image processing module.

[0074] In this step, the first camera 101, with its high resolution (≥5 million pixels) and small pixel size, combined with the uniform light effect of the backlight, can accurately capture the overall edge contour and flat edge features of the wafer. The global shutter technology effectively avoids image ghosting caused by slight wafer displacement during image acquisition, providing high-definition and complete image data support for subsequent overall centering calculations.

[0075] S3: The image processing module processes the overall edge contour image, extracts the wafer edge curve, calculates the wafer's center position and flat edge rotation angle, and sends the calculation results to the motion control module 2; The specific process of step S3 is as follows: After receiving the overall edge contour image transmitted by the first camera 101, the image processing module initiates a step-by-step processing flow: First, lens distortion can be eliminated through distortion correction algorithm, and the conversion between image coordinates and device coordinate system is completed by combining the intrinsic and extrinsic parameter matrices obtained from hand-eye calibration; then, edge contrast can be enhanced through adaptive histogram equalization, image noise can be filtered by Gaussian filtering, and a preset ROI region can be extracted to reduce invalid calculations; next, dynamic threshold edge detection based on the Canny algorithm framework is used, combined with sub-pixel positioning technology to extract the wafer edge curve, and broken edges are repaired and noise is removed through morphological operations; finally, the effective edge points are fitted with a circle using iterative weighted least squares method to calculate the wafer's center position (X0, Y0), radius R, and flat edge rotation angle, and the rationality of the calculation results is verified simultaneously to ensure that the deviation is within an acceptable range before sending the calculation results to the motion control module 2.

[0076] In this step, multiple algorithms are used to process the wafer together to eliminate imaging interference and errors, accurately extract the overall geometric features of the wafer, and provide reliable positioning parameters for the initial centering. The iterative weighted least squares method can effectively reduce the impact of slight defects at the wafer edge on the fitting results and ensure the stability of parameter calculation.

[0077] S4: The motion control module 2 drives the vacuum adsorption plate holder 100 to move to the first reference position according to the center position and the flat edge rotation angle. The specific process of step S4 is as follows: After receiving the center position and flat edge rotation angle parameters output by the image processing module, the motion control module 2 drives the X-axis drive mechanism 201, Y-axis drive mechanism 202, and R-axis drive mechanism 203 on the bottom support 200 to work together. The X-axis and Y-axis drive mechanisms move along the vertical direction (orthogonal coordinate system), causing the vacuum adsorption wafer stage 100 to translate, quickly bringing the wafer center close to the preset reference. The R-axis drive mechanism drives the wafer stage to rotate, calibrating the flat edge rotation angle, and finally moving the wafer to the first reference position, ensuring that the distance between the first reference position and the preset reference position is no greater than 300μm. Throughout the adjustment process, the grating ruler collects the actual position of the wafer stage in real time, and the deviation is corrected through a PID algorithm, forming a closed-loop control.

[0078] In this step, relying on the closed-loop control architecture of three-axis linkage, the wafer position and angle can be quickly adjusted. The 300μm threshold design takes into account both the initial centering efficiency and the fine-tuning space for subsequent precise positioning, avoiding the extension of the overall positioning cycle due to excessive initial deviation.

[0079] S5: The second camera 102 images the wafer dicing edge separately, acquires the shape, position and angle of the dicing edge, and transmits it to the image processing module; The specific process of step S5 is as follows: After the wafer reaches the first reference position, the second camera 102 (an industrial-grade monochrome CMOS area array camera of the same model as the first camera) starts working. Relying on the precise positioning of the optical bracket 204, it focuses on the wafer's diced edge area for local high-definition imaging. The backlight 103 remains operational, with strong light focused on the diced edge to further highlight the details of the flat edge or notch features. The second camera 102 acquires images of the diced edge shape, position, and angle through a global shutter, ensuring that the image clearly presents the diced edge features. After acquisition, the image data is transmitted to the image processing module.

[0080] In this step, selecting the same model of dual cameras ensures consistent imaging performance. The small field-of-view focusing design of the second camera can improve the imaging accuracy of the cut edge area. Combined with the directional illumination of the backlight, it can effectively resist interference such as surface stains and slight reflections, providing high-quality local images for accurate positioning.

[0081] S6: The image processing module performs depth processing on the image of the cut edge shape, position and angle, calculates the cut edge position and angle of the wafer, and sends the calculation results to the motion control module 2; The specific process of step S6 is as follows: After receiving the wafer edge image transmitted by the second camera 102, the image processing module performs depth optimization based on the previous preprocessing, focusing on enhancing the feature extraction of the edge area: by using precise threshold filtering to remove interference from defective areas such as edge burrs and cracks, and using quadratic polynomial interpolation to achieve sub-pixel-level positioning of edge features, combined with the preset mask reference position, the wafer edge position coordinates and edge angle θ are calculated, the distance and rotation angle of the wafer that need to be further translated are determined, and after verifying the parameter error, the data is sent to the motion control module 2.

[0082] In this step, targeted algorithm optimization is performed on the local features of the cutting edge to improve the accuracy of the cutting edge parameter calculation, providing a micron-level adjustment basis for subsequent fine positioning and ensuring that the cutting edge position is precisely matched with the cutting edge equipment reference.

[0083] S7: The motion control module 2 drives the vacuum adsorption support stage 100 to move to the second reference position according to the cutting edge position and cutting edge angle; The specific process of step S7 is as follows: After receiving the cutting edge position and angle parameters, the motion control module 2 drives the three-axis drive mechanism to initiate a fine-tuning mode. Compared to the initial centering stage, this reduces the adjustment speed to improve stability. The X and Y axis drive mechanisms perform micron-level displacement compensation, while the R axis drive mechanism performs angle fine-tuning, precisely moving the vacuum adsorption stage 100 to ultimately position the wafer at the second reference position, ensuring that the distance between the second reference position and the preset reference position is no greater than 60μm. During the adjustment process, deviations can be corrected in real time through closed-loop feedback to ensure that the positioning accuracy meets the standards.

[0084] In this step, the 60μm threshold design meets the high precision requirements of ultra-thin wafer dicing, and the deviation is accurately compensated through low-speed fine-tuning and closed-loop control.

[0085] S8: Motion control module 2 sends a positioning ready signal to the edge cutting device. After the edge cutting is completed, the edge cutting device sends a reset signal to motion control module 2. After receiving the reset signal from the edge cutting device, motion control module 2 drives the vacuum adsorption support stage 100 back to the initial position, completing the single pre-positioning process. The specific process of step S8 is as follows: After the wafer reaches the second reference position, the motion control module 2 sends a positioning ready signal to the dicing equipment, informing it that the dicing process can begin. After completing the dicing operation, the dicing equipment sends a reset signal to the motion control module 2. Upon receiving the reset signal, the motion control module 2 drives the three-axis drive mechanism to return the vacuum adsorption stage 100 to its initial position. Simultaneously, the vacuum adsorption stage 100 releases the vacuum, waiting for the robotic arm to remove the diced wafer, completing a single pre-positioning process. Subsequently, the system automatically resets, preparing for the positioning operation of the next wafer.

[0086] In this step, the motion control module and the edge cutting equipment are linked to realize an automated closed-loop process of positioning-edge cutting-reset, which does not require manual intervention and ensures production continuity. At the same time, the reset design provides a unified initial reference for the next positioning, avoiding the accumulation of reference deviation.

[0087] In some embodiments of the present invention, step S3 further includes: After calculating the wafer's center position (X0, Y0), radius R, and flat-edge rotation angle, the positioning result verification process is executed: The calculated center position and radius R are compared with the preset wafer specifications. If the error exceeds the preset threshold (e.g., center point position error > 20μm), the repositioning process is triggered, and the image processing and parameter calculation are re-performed in this step. If the positioning error exceeds the standard three times in a row, an alarm signal will be output immediately, the positioning process will be suspended and a fault investigation prompt will be given to ensure the reliability of the positioning results. After the verification is passed, the calculation results of the center position and the rotation angle of the flat side are sent to the motion control module 2.

[0088] In summary, this pre-positioning method, through the synergistic effect of various modules, not only solves the problem of damage to ultra-thin wafers caused by traditional mechanical positioning, but also overcomes the bottleneck of the imbalance between accuracy and economy in single-camera visual positioning. It achieves a balance of positioning accuracy, efficiency, compatibility and stability, and is fully adapted to the pre-positioning requirements of edge trimming for semiconductor ultra-thin wafers.

[0089] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A pre-positioning system for ultrathin wafers, characterized in that, include: Vacuum adsorption wafer support stage (100), the vacuum adsorption wafer support stage (100) is used to support and adsorb and fix wafers; An optical imaging module (1) is provided, comprising a first camera (101) and a second camera (102). The first camera (101) performs overall image acquisition to obtain an overall image of the wafer, and the second camera (102) is used to perform separate imaging of the wafer diced edge portion to obtain an image of the wafer diced edge portion. The image processing module is connected to the optical imaging module (1) and is used to receive the overall image of the wafer and calculate the center position and the flat edge rotation angle of the wafer. It is also used to receive the image of the wafer's cut edge and calculate the cut edge position and the cut edge angle of the wafer. The motion control module (2) is connected to the image processing module and the vacuum adsorption support stage (100) respectively. It is used to control the vacuum adsorption support stage (100) to move to the first reference position according to the center position and the flat edge rotation angle. It is also used to control the vacuum adsorption support stage (100) to move to the second reference position according to the cutting edge position and the cutting edge angle. The motion control module (2) is also used to send a positioning ready signal to the edge cutting device after the wafer is positioned to the second reference position.

2. The pre-positioning system for ultrathin wafers according to claim 1, characterized in that, The optical imaging module (1) also includes a backlight (103) for illuminating the back of the wafer to highlight the wafer edge contour.

3. The pre-positioning system for ultrathin wafers according to claim 1, characterized in that, It also includes a bottom bracket (200), on which the motion control module (2) is mounted.

4. The pre-positioning system for ultrathin wafers according to claim 3, characterized in that, The motion control module (2) includes an X-axis drive mechanism (201), a Y-axis drive mechanism (202) located at the output end of the X-axis drive mechanism (201), and an R-axis drive mechanism (203) located at the output end of the Y-axis drive mechanism (202). The X-axis drive mechanism (201) is located on the bottom support (200). The vacuum adsorption support stage (100) is located at the output end of the R-direction drive mechanism (203); The X-axis driving mechanism (201) is used to drive the Y-axis driving mechanism (202) to move along the first direction, the Y-axis driving mechanism (202) is used to drive the R-axis driving mechanism (203) to move along the second direction, and the R-axis driving mechanism (203) is used to drive the vacuum adsorption support stage (100) to rotate.

5. A pre-positioning system for ultrathin wafers according to claim 4, characterized in that, The first direction is perpendicular to the second direction.

6. The pre-positioning system for ultrathin wafers according to claim 3, characterized in that, An optical bracket (204) is provided on the bottom bracket (200), and the optical imaging module (1) is provided on the optical bracket (204).

7. The pre-positioning system for ultrathin wafers according to claim 1, characterized in that, Both the first camera (101) and the second camera (102) are industrial-grade monochrome CMOS area array cameras.

8. The pre-positioning system for ultrathin wafers according to claim 1, characterized in that, The distance between the first reference position and the preset reference position is no greater than 300um.

9. A pre-positioning system for ultrathin wafers according to claim 1, characterized in that, The distance between the second reference position and the preset reference position is no greater than 60 μm.

10. A pre-positioning method for a pre-positioning system for ultrathin wafers according to any one of claims 1-9, characterized in that, The method includes the following steps: S1: The robotic arm places the wafer on the vacuum adsorption stage (100), which adsorbs and fixes the wafer; S2: The first camera (101) captures an overall image of the wafer, obtains the overall edge contour image of the wafer, and transmits it to the image processing module; S3: The image processing module processes the overall edge contour image, extracts the wafer edge curve, calculates the center position and flat edge rotation angle of the wafer, and sends the calculation results to the motion control module (2). S4: The motion control module (2) drives the vacuum adsorption plate holder (100) to move to the first reference position according to the center position and the flat edge rotation angle; S5: The second camera (102) separately images the wafer dicing edge, acquires the shape, position and angle of the dicing edge image and transmits it to the image processing module; S6: The image processing module performs depth processing on the image of the cut edge shape position and angle, calculates the cut edge position and cut edge angle of the wafer, and sends the calculation results to the motion control module (2). S7: The motion control module (2) drives the vacuum adsorption support stage (100) to move to the second reference position according to the cutting edge position and cutting edge angle; S8: The motion control module (2) sends a positioning ready signal to the edge cutting device. After the edge cutting is completed, the edge cutting device sends a reset signal to the motion control module (2). After receiving the reset signal from the edge cutting device, the motion control module (2) drives the vacuum adsorption support stage (100) back to the initial position to complete the single pre-positioning process.

Images