Multi-vision bonding alignment method and system, computer equipment and medium

Abstract

The invention relates to a multi-vision bonding alignment method and system, computer equipment and a medium, and relates to the technical field of semiconductor manufacturing. The method comprises the following steps: determining a reference position of a first carrying table and an image of a third marking graph of a marking image area on the first carrying table, then controlling the first carrying table to move, determining the position of a second carrying table, aligning the positions of corresponding wafers on the two carrying tables, and obtaining the image of the third marking graph at the current position through a window area of the second carrying table, and comparing the image of the third mark pattern at the current position with the image of the third mark pattern at the reference position to obtain an alignment deviation between the reference position and the current position, so as to control the first carrying table to move to eliminate the alignment deviation. Compared with the prior art, mechanical positioning errors caused by the two processes are solved, the final alignment precision of the two wafers is improved, the image fitting process is omitted, and the alignment process is simpler.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of semiconductor manufacturing, in particular to a multi-vision bonding alignment method, system, computer device and medium. BACKGROUND

[0002] Wafer to wafer bonding, also known as wafer direct bonding, refers to a process of bonding two wafers together. In the field of semiconductor technology, wafer direct bonding technology can be used to realize wafer to wafer three-dimensional integration. That is, two or more wafers with the same or different functions are bonded. With the development of the times, the requirement for bonding alignment accuracy between two wafers is also getting higher and higher. When bonding fails, it usually results in the scrapping of two or more wafers, so the cost loss when bonding fails is extremely high. The reason for the bonding failure may be that the wafer alignment accuracy is not enough.

[0003] In the current bonding alignment technology, it is difficult to accurately align two wafers due to line-of-sight obstruction. Therefore, in one technology, IR penetration technology is used to achieve the alignment of two wafers. However, because the infrared wavelength is too long, the image resolution is poor, and therefore the wafer alignment accuracy is not enough. In another technology, a corresponding mark pattern is arranged on each of the two bonding target objects' support platforms (upper and lower loading platforms). The mark pattern arranged on the upper loading platform is arranged near the wafer adsorbed by the upper loading platform, and can be used to calibrate the position of the wafer on the upper loading platform. The mark pattern arranged on the lower loading platform is arranged on a glass window, and also near the wafer adsorbed by the lower loading platform, and can also be used to calibrate the position of the wafer on the loading platform. After the upper and lower mark patterns are calibrated to obtain a calibration image, accurate alignment can be achieved by controlling the two loading platforms to reach the calibration image during alignment. However, this alignment method still has the problem of low alignment accuracy. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a multi-vision bonding alignment method, system, computer device and medium, which can reduce mechanical positioning error, improve alignment accuracy, and has the characteristics of simple alignment process.

[0005] In a first aspect, a multi-vision bonding alignment method is provided in an embodiment, comprising: determining a reference position of a first loading platform, and obtaining an image of a third mark pattern in a mark pattern region on the first loading platform; controlling the first loading platform to move and determining the position of a second loading platform, so that when the first loading platform is at the reference position, the position of a wafer on the second loading platform is aligned with the position of a wafer on the first loading platform; wherein the mark pattern region on the first carrier and the viewing window region on the second loading platform are arranged on the same side of the corresponding wafer; The method comprises the following steps: controlling the first carrier to restore the reference position to obtain a current position, and acquiring an image of the third mark pattern at the current position through the window area of the second carrier; comparing the image of the third mark pattern at the current position with the image of the third mark pattern at the reference position to obtain an alignment deviation between the reference position and the current position, so as to control the first carrier to move to eliminate the alignment deviation.

[0006] In a second aspect, a multi-vision bonding alignment system is provided in an embodiment, comprising: The first carrier is provided with a mark pattern area, and the third mark pattern is arranged on the mark pattern area. The second carrier is provided with a window area, and the third mark pattern on the first carrier can be acquired through the window area; the third mark pattern and the window area are arranged on the same side of the wafer fixing area. The first vision subsystems are used to observe the positions of the wafers on the first carrier and the second carrier in sequence, so that at least two pairs of corresponding mark patterns of the two wafers are located at the visual centers of the corresponding first vision subsystems, and the reference position of the first carrier is determined based on the positions of the visual centers; and the second vision subsystems are used to observe the third mark pattern before bonding, so as to determine the alignment deviation between the current position and the reference position of the first carrier, and control the first carrier to move to eliminate the alignment deviation.

[0007] In a third aspect, a computer device is provided in an embodiment, comprising at least one computer storage medium and at least one processor, the at least one computer storage medium is stored with a control program of the multi-vision bonding alignment method, and the at least one processor is used to execute the control program stored on the at least one computer storage medium.

[0008] In a fourth aspect, a computer readable storage medium is provided in an embodiment, wherein the medium is stored with a program, and the program can be loaded and executed by a processor to execute the multi-vision bonding alignment method.

[0009] The multi-vision bonding alignment method has the following advantages: In the alignment process, the alignment of the wafer is realized by the alignment of the mark patterns on the two wafers first, so that the mechanical error caused by the relative mechanical movement in the wafer adsorption process can be solved; the alignment of the upper and lower stages is realized by the third mark pattern in the first stage mark pattern area, so that the mechanical error caused by the reciprocating movement of the stage can be solved, thereby reducing the mechanical positioning error caused by the above two processes and improving the final alignment accuracy of the two wafers. In addition, the whole alignment process does not need to be image-fitted, so the alignment process is simpler. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is a schematic view of a multi-vision bonding system structure carrying a wafer according to an embodiment of the present application; Figure 2 is a sectional view of a lower stage fixed with a wafer according to an embodiment of the present application; Figure 3 is a top view of a lower stage carrying a wafer according to an embodiment of the present application; Figure 4 is a top view of a lower stage carrying a wafer according to an embodiment of the present application; Figure 5 is a top view of a lower stage carrying a wafer according to an embodiment of the present application; Figure 6 is a sectional view of an upper stage fixed with a wafer according to an embodiment of the present application; Figure 7 is a flowchart of a multi-vision bonding method according to an embodiment of the present application; Figure 8 is a schematic view of the alignment of the first mark image and the vision center of the first vision subsystem according to an embodiment of the present application; Figure 9 is a schematic view of the alignment of the third mark image and the vision center of the second vision subsystem according to an embodiment of the present application; Figure 10 is a schematic view of the determination of the position of the second stage according to an embodiment of the present application; Figure 11 is a schematic view of the alignment of the first stage and the second stage according to an embodiment of the present application.

[0011] In the diagram, 01 is a first stage, 02 is a second stage, 03 is a first vision subsystem, 031 is a first image acquisition device, 032 is a second image acquisition device, 041 is a third image acquisition device, 05 is a first window area, 06 is a third mark pattern, 07 is a second window area, 08 is a first wafer, 09 is a second wafer, and 10 is a first mark pattern. Detailed Implementation

[0012] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0013] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0014] The serial numbers assigned to components in this document, such as "first" and "second," are used only to distinguish the described objects and have no sequential or technical meaning. The terms "connection" and "linkage" used in this application, unless otherwise specified, include both direct and indirect connections (linkages).

[0015] To facilitate the explanation of the inventive concept of this application, the wafer bonding alignment technology will be briefly described below.

[0016] In a current wafer bonding alignment technology, corresponding marker patterns (upper and lower marker patterns) are set on the upper and lower loading stages. First, the position of the wafer on the upper loading stage is calibrated based on the upper marker pattern, and the position of the wafer on the lower loading stage is calibrated based on the lower marker pattern. Once the wafer's position on the stage is calibrated, the stage carries the wafer for vertical alignment. During the vertical alignment process, a composite image of the upper and lower marker patterns is obtained through an image acquisition device, and then vertical alignment is performed based on the calibration patterns.

[0017] The applicant discovered in their research that, in the aforementioned scheme, on the one hand, although the wafers on the upper and lower stages have been positioned on the stages using corresponding marker patterns, mechanical errors due to relative mechanical movement still exist during the wafer adsorption and fixation process, resulting in low alignment accuracy. On the other hand, the alignment process also requires obtaining a fitting image using the two marker patterns on the upper and lower stages, and determining the alignment error through the fitting image, a complex process.

[0018] In view of this, this application provides a multi-vision bonding alignment method, system, computer device, and medium. During the alignment process, the wafers are first aligned by aligning the marking patterns on the two wafers, which solves the mechanical error introduced by the relative mechanical movement of the wafers during the adsorption process in the prior art, which relies on the marking patterns on the stage to determine the wafer position. Then, the alignment of the upper and lower stages is achieved by aligning the marking patterns on one of the wafer stages. This solves the mechanical error caused by the reciprocating motion of the stage, thereby reducing the mechanical positioning error caused by the above two processes and improving the final alignment accuracy of the two wafers. On the other hand, since there is no need to fit an image, the alignment process is simpler.

[0019] For ease of understanding, the multi-vision bonding alignment system will be explained below.

[0020] This application provides a multi-vision bonding system in its embodiments. Please refer to [link / reference]. Figure 1 It includes a first platform 01, a second platform 02, a first vision subsystem 03, and a second vision subsystem.

[0021] Those skilled in the art will understand that since wafer bonding is often based on two platforms, upper and lower, the embodiments of this application will use upper and lower platforms as examples for illustration.

[0022] It should be noted that the first stage 01 in this application can be either an upload stage or a download stage, without affecting the scope of protection of this application. For a download stage, the wafers on it are adsorbed and fixed above the download stage; for an upload stage, the wafers on it are adsorbed and fixed below the upload stage. For ease of understanding, the following explanation uses the first stage 01 as a download stage as an example. Correspondingly, the second stage 02 is an upload stage. However, those skilled in the art will understand that the solution of this application is also applicable to the case where the first stage 01 is an upload stage and the second stage 02 is a download stage, all of which are within the scope of protection of this application. Those skilled in the art will understand that when the first stage 01 is an upload stage, it is similar to... Figure 1 Conversely, the third image acquisition device 041 is located below the first stage 01 and the second stage 02, and its visual direction is upward.

[0023] Please refer toFigure 2 , Figure 3 , Figure 4 and Figure 5 For the first platform 01, a marking graphic area 05 is provided on it, and a third marking graphic 06 is provided on the marking graphic area 05. In one embodiment, please refer to... Figure 3 Two graphic centers can be defined for the third marker graphic 06: a first graphic center 061 and a second graphic center 062. In one embodiment, please refer to... Figure 4 Two third marker graphics 06 are provided on the marker graphic area 05. In one embodiment, please refer to... Figure 5 There are two marked graphic regions 05, and each marked graphic region 05 is set with a third marked graphic, so that two third marked graphics 06 can be obtained.

[0024] Those skilled in the art will understand that there can be more than one graphic center or third mark graphic 06, and all of them are within the scope of protection of this application without departing from the inventive concept of this application.

[0025] For marking graphic areas, it can be like this: Figure 4 and Figure 5 The transparent viewing area shown can also be as follows: Figure 3 The existing area on the first stage 01 shown can be used to create marker graphics directly on the existing area. For the two marker graphic areas, one can be a transparent window area and the other can be the existing area on the first stage 01.

[0026] For the second platform 02, please refer to... Figure 6 The viewing area 07 of the second stage includes a penetrating viewing area and / or a viewing area provided with a light-transmitting material. To allow observation of the third mark pattern on the first carrier 01 through the viewing area 07 of the second stage, the viewing area 07 of the second stage can be one or two. When the second stage has two viewing areas 07, both can be penetrating viewing areas (i.e., through-holes), both can be viewing areas provided with a light-transmitting material, or one can be a penetrating viewing area and the other a viewing area provided with a light-transmitting material. The penetrating viewing area includes the area corresponding to the portion of the second carrier 02 that only retains the portion carrying the second wafer 09.

[0027] Those skilled in the art will understand that the viewing area 07 and the marking pattern area 05 of the second stage are both located on the same side of the wafer fixing area so that the third image acquisition device 041 can acquire an image of the third marking pattern 06 through the viewing area 07 of the second stage during alignment.

[0028] In some embodiments, the transparent window area may be made of glass, but other transparent materials may also be used.

[0029] The second vision subsystem includes at least two subsystems, each of which includes a third image acquisition device 041. This allows the third image acquisition device 041 to acquire the corresponding third mark pattern 06, observe the third mark pattern 06 before bonding, and determine the alignment deviation between the current position of the first stage 01 and the reference position, so as to control the movement of the first stage 01 to eliminate the alignment deviation.

[0030] The first vision subsystem 03 includes at least two subsystems, each including a first image acquisition device 031 and a second image acquisition device 032 with opposite image acquisition directions and coaxial optical axes. These at least two first vision subsystems are used to sequentially observe the positions of wafers (first wafer 08) on the first stage 01 and wafers (second wafer 09) on the second stage 02, such that at least two pairs of corresponding marker patterns on the two wafers are located at the visual center of the corresponding first vision subsystem 03, and the reference position of the first stage 031 can be determined by the visual center position.

[0031] In one embodiment, the first image acquisition device 031 and the second image acquisition device 032 of the first vision subsystem 03 are fixedly connected by a robotic arm. Typically, a C-shaped robotic arm can be used to fix the relative positions of the first image acquisition device 031 and the second image acquisition device 032. Alternatively, a synchronous linkage control method can be used without mechanical fixed connection.

[0032] Based on the aforementioned wafer direct bonding alignment system, since at least two first vision subsystems 03 are used, when determining the positions of the first wafer 08 on the first stage 01 and the second wafer 09 on the second stage 02, at least two pairs of corresponding marker patterns on the two wafers are located at the visual center of the corresponding first vision subsystem. Furthermore, by aligning the center of the marker pattern with the visual center, the reference position of the first stage 01 and the position of the second stage 02 can be determined. When both stages are in their determined positions, the first wafer 08 and the second wafer 09 are aligned. Based on this characteristic, even if there are placement errors when the wafers are adsorbed and fixed onto the stages, these errors can be eliminated using this alignment method. This solves the mechanical errors introduced by the relative mechanical movement of the wafers during adsorption and fixing, which are present in the prior art when relying on marker patterns on the stages to determine wafer positions. Furthermore, the mechanical errors caused by the reciprocating motion of the stages can be eliminated by using the marker pattern image of the marked pattern area.

[0033] The inventive concept of this application will be further explained below in conjunction with the visual bonding alignment method.

[0034] This application provides a multi-view bonding alignment method, which can be implemented based on the multi-view bonding alignment system of any of the above embodiments. Please refer to... Figure 7 Alignment methods include: Step S10: Determine the reference position of the first platform and acquire an image of the third marker graphic in the marked graphic area on the first platform.

[0035] Please refer to Figure 8 The first stage 01 is controlled to move relative to at least two first vision subsystems 03, such that the first stage 01 is located between the first image acquisition device 031 and the second image acquisition device 032 included in each first vision subsystem 03, and the centers of at least two first marking patterns 10 on the first wafer 08 on the first stage are respectively located at the visual center of the corresponding first vision subsystem 03, and the alignment position of the visual center of the first stage is used as the determined reference position.

[0036] For ease of understanding, the following embodiments use two first visual subsystems, each corresponding to two first marker graphics and two second marker graphics, and two second visual subsystems, each corresponding to two third marker graphics, as examples for illustration.

[0037] Since the first stage 01 moves relative to the two first vision subsystems 03, we can first control the first stage 01 to move relative to one of the first vision subsystems 03, thereby moving the first stage 01 and / or one of the first vision subsystems 03 so that the visual center of the first vision subsystem 03 is aligned with the center of a first mark pattern 10 on the first wafer 08 on the first stage 01. Next, the other first vision subsystem is moved so that the visual center of the other first vision subsystem is aligned with the center of another first mark pattern on the first wafer 08 on the first stage 01. Please refer to... Figure 8 As those skilled in the art will understand, due to the symmetrical occlusion relationship between the two first visual subsystems in this schematic diagram, only one of the first visual subsystems can be seen in the diagram. Therefore, the current position of the first stage 01 can be used as a definite reference position and recorded.

[0038] Next, it is necessary to determine the position of the second stage 02 when the two wafers are aligned. Due to the occlusion relationship between the two wafers, it is necessary to control the movement of the first stage 01 so that it moves out of the visual range of the first vision subsystem. However, when the first stage 01 returns to the reference position, mechanical movement errors exist, and due to the obstruction of the line of sight, the two first vision subsystems cannot observe the two first and second mark patterns, thus the mechanical movement errors cannot be resolved. Therefore, it is necessary to use the third mark pattern 06 on the first stage 01 to solve this problem.

[0039] Therefore, after determining the reference position of the first stage 01, before controlling it to leave the visual range of the first vision subsystem, please refer to... Figure 9 The first stage 01 and the two first vision subsystems 03 remain in their fixed positions, while the two second vision subsystems are moved so that the first stage 01 is located within the visual area of ​​the third image acquisition device 041 included in each second vision subsystem, and the centers of at least two third marker graphics 06 are located at the visual center of the corresponding second vision subsystem (i.e., at the visual center of the third image acquisition device 041 included in the second vision subsystem). Then, the images of the corresponding third marker graphics can be acquired by the third image acquisition device to indirectly record the position of the first stage 01.

[0040] In other embodiments, the reference position of the first stage 01 can also be determined in other ways. For example, a preset position can be directly used as the reference position, and then each first vision subsystem and each second vision subsystem can be moved to determine the positions of all first and second vision subsystems to help obtain the alignment position of the second stage 02 with the first stage 01 and to acquire the image of the third marker graphic. When the first stage 01 is in the reference position, the positional relationship between it and the first vision subsystem 03 and the second vision subsystem can be referenced. Figure 9 As will be understood by those skilled in the art, whether the first vision subsystem is controlled to move and align first or the second vision subsystem is controlled to move and align first does not affect the solution to the technical problem of this application, and both are within the protection scope of this application.

[0041] Step S20: Control the first stage to move and determine the position of the second stage, so that when the first stage is in the reference position, the position of the wafer on the second stage is aligned with the position of the wafer on the first stage. The marking pattern area on the first stage and the viewing window area on the second stage are located on the same side of the corresponding wafer.

[0042] As one embodiment of this application, step S20 may include: Step S201: Control the first platform to move so that the first platform leaves the visual area of ​​all first vision subsystems.

[0043] Please refer to Figure 1 The presence of the first stage 01 will block the observation of the second stage by the first vision subsystem 03. Therefore, after the position of the first vision subsystem is fixed, the first stage 01 is controlled to leave the visual area of ​​the first vision subsystem in order to prepare for the second stage 02 to enter the visual area of ​​the first vision subsystem.

[0044] Those skilled in the art will understand that the marking patterns on the wafer are observed through the first image acquisition module 031 for the second stage 02 to be entered. Therefore, the presence of the first stage 01 will obstruct the observation of the marking patterns on the second wafer 09. Thus, the movement of the first stage 01 here needs to leave the visual area of ​​the first vision subsystem 03, but whether it moves out of the visual area of ​​the second vision subsystem is not limited, and both are within the protection scope of this application.

[0045] Step S202: Control the second stage to move so that it enters the visual area of ​​all first vision subsystems, and the centers of at least two second marker patterns on the second wafer on the second stage are respectively located at the visual centers of the corresponding first vision subsystems, and determine the current position of the second stage as the position of the second stage. Here, at least two first marker patterns correspond one-to-one with at least two second marker patterns, and their positions on the wafers are also corresponding.

[0046] Please refer to Figure 10 In step S202, while the first vision subsystem 03 and the second vision subsystem remain stationary, the second stage 02 is controlled to move so that the two second mark patterns are respectively located at the visual centers of the two corresponding second vision subsystems. In this way, it can be ensured that the position of the second wafer 09 at this time is aligned vertically with the position of the first wafer 08 when the first stage 01 is in the reference position.

[0047] In addition, the applicant's research found that if there is only one third marker pattern, due to the distance between the third marker pattern and the wafer, and the large size of the wafer itself, there is a very slight error between the images of the third marker pattern obtained before and after, and the alignment error between the two wafers is already large. However, it is difficult to obtain this image error through the image acquisition device. Therefore, there is still a large alignment accuracy problem.

[0048] Therefore, this application employs two or more pairs of third marker graphics for more precise comparison and alignment.

[0049] Step S30: Control the first stage to restore the reference position to achieve alignment.

[0050] Step S30 may specifically include: controlling the first stage to recover the current position according to the determined reference position, and acquiring an image of the third mark pattern at the current position through the viewing area of ​​the second stage, comparing the image of the third mark pattern at the current position with the image of the third mark pattern at the reference position to obtain the alignment deviation between the reference position and the current position, so as to control the first stage to move and eliminate the alignment deviation.

[0051] As a specific embodiment of this application, controlling the first stage 01 to recover the current position according to a determined reference position, and acquiring an image of the third marker graphic at the current position through the viewing area of ​​the second stage, may include: Step S100: Control the first stage to move so that the first stage enters the visual areas of all the first vision subsystems and the second vision subsystems, and restores the current position according to the determined reference position.

[0052] After the first stage 01 returns to its reference position, it can be aligned with the second stage 02, whose alignment position has been determined. However, due to mechanical motion errors, the position that the first stage 01 believes it has returned to may deviate from the previous reference position. Therefore, it is necessary to know the deviation between the current position obtained after restoration and the original reference position.

[0053] Step S200: Control each third image acquisition device to acquire the image of the current third marker graphic through the viewing area of ​​the second platform.

[0054] By comparing the image of the third marker at the current position with the image of the third marker at the reference position, the deviation between the current position and the calibration position can be accurately obtained, thereby controlling the movement of the first platform 01 to achieve the desired result. Figure 11 The final alignment is shown.

[0055] Those skilled in the art will understand that each step of moving the first platform to correct the deviation can obtain the current position of the first platform, thereby obtaining a new image of the third marker graphic for comparison, which is also within the scope of protection of this application.

[0056] In addition, during the relative vertical movement of the first and second platforms, a third marker graphic can be used to observe whether alignment deviation has occurred.

[0057] Those skilled in the art will understand that the solutions in the above embodiments are also applicable to situations where the first platform is an uploading platform and the second platform is a downloading platform, and are also within the protection scope of this application.

[0058] Based on the above alignment method, by aligning the marking patterns on the upper and lower wafers, the mechanical motion error introduced during the wafer adsorption process in determining the wafer position by relying on the marking patterns on the stage is resolved. Furthermore, by using the third marking pattern in the marking pattern area of ​​the first stage, the mechanical motion error during the repeated positioning process of the first stage is resolved. This reduces the mechanical positioning error caused by the above two processes and improves the final alignment accuracy of the two wafers. In addition, since the entire alignment process does not require image fitting, it is simpler.

[0059] One embodiment of this application provides a computer device, including at least one computer storage medium and at least one processor. The at least one computer storage medium stores a control program for any of the wafer direct bonding alignment methods described in the above embodiments, and the at least one processor is used to execute the control program stored on the at least one computer storage medium.

[0060] One embodiment of this application provides a computer-readable storage medium storing a program, the stored program including methods that can be loaded by a processor and processed in any of the above embodiments.

[0061] Those skilled in the art will understand that all or part of the functions of the various methods in the above embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to achieve the above functions. For example, the program can be stored in the memory of a device, and when the program in the memory is executed by the processor, all or part of the above functions can be achieved. In addition, when all or part of the functions in the above embodiments are implemented by computer programs, the program can also be stored in a server, another computer, disk, optical disk, flash drive, or external hard drive, etc., and can be downloaded or copied to the memory of a local device, or the system of the local device can be updated. When the program in the memory is executed by the processor, all or part of the functions in the above embodiments can be achieved.

[0062] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A multi-vision bond alignment method, characterized by, The method comprises the following steps: determining a reference position of the first carrier and obtaining an image of a third mark pattern in a mark pattern region on the first carrier; controlling the first carrier to move and determining a position of the second carrier so that when the first carrier is at the reference position, the position of the wafer on the second carrier is aligned with the position of the wafer on the first carrier; wherein the mark pattern region on the first carrier and the window region on the second carrier are arranged on the same side of the corresponding wafer; controlling the first carrier to return to the reference position to achieve alignment, comprising: controlling the first carrier to return to the determined reference position to obtain a current position, and obtaining an image of the third mark pattern in the current position through the window region of the second carrier, comparing the image of the third mark pattern in the current position with the image of the third mark pattern in the reference position to obtain an alignment deviation between the reference position and the current position, so as to control the first carrier to move to eliminate the alignment deviation.

2. The multi-vision bond alignment method of claim 1, wherein, The method comprises the following steps: controlling the first carrier to move and determining a position of the second carrier so that when the first carrier is at the reference position, the position of the wafer on the second carrier is aligned with the position of the wafer on the first carrier; wherein the mark pattern region on the first carrier and the window region on the second carrier are arranged on the same side of the corresponding wafer; controlling the first carrier to return to the reference position to achieve alignment, comprising: controlling the first carrier to return to the determined reference position to obtain a current position, and obtaining an image of the third mark pattern in the current position through the window region of the second carrier, comparing the image of the third mark pattern in the current position with the image of the third mark pattern in the reference position to obtain an alignment deviation between the reference position and the current position, so as to control the first carrier to move to eliminate the alignment deviation.

3. The multi-vision key bonding alignment method of claim 1, wherein, The method comprises the following steps: controlling the first carrier to move and determining a position of the second carrier so that when the first carrier is at the reference position, the position of the wafer on the second carrier is aligned with the position of the wafer on the first carrier; wherein the mark pattern region on the first carrier and the window region on the second carrier are arranged on the same side of the corresponding wafer; controlling the first carrier to return to the reference position to achieve alignment, comprising: controlling the first carrier to return to the determined reference position to obtain a current position, and obtaining an image of the third mark pattern in the current position through the window region of the second carrier, comparing the image of the third mark pattern in the current position with the image of the third mark pattern in the reference position to obtain an alignment deviation between the reference position and the current position, so as to control the first carrier to move to eliminate the alignment deviation. The method comprises the following steps:

4. The multi-vision bonding alignment method according to claim 2 or 3, wherein, controlling the first carrier to move and determining a position of the second carrier so that when the first carrier is at the reference position, the position of the wafer on the second carrier is aligned with the position of the wafer on the first carrier; wherein the mark pattern region on the first carrier and the window region on the second carrier are arranged on the same side of the corresponding wafer; controlling the first carrier to return to the reference position to achieve alignment, comprising: controlling the first carrier to return to the determined reference position to obtain a current position, and obtaining an image of the third mark pattern in the current position through the window region of the second carrier, comparing the image of the third mark pattern in the current position with the image of the third mark pattern in the reference position to obtain an alignment deviation between the reference position and the current position, so as to control the first carrier to move to eliminate the alignment deviation. controlling the first carrier to move so that the first carrier leaves the visual field of all the first vision subsystems; controlling the second carrier to move so that the second carrier enters the visual field of all the first vision subsystems and so that the centers of the second mark patterns on the second wafer on the second carrier are located at the visual centers of the corresponding first vision subsystems respectively, and determining the current position of the second carrier as the position of the second carrier; wherein the first mark patterns and the second mark patterns correspond to each other in one-to-one correspondence and in position on the wafer.

5. The multi-vision bond alignment method according to any one of claims 1 to 3, wherein The mark pattern region includes a transparent window region opened on the first carrier and / or a region originally present on the first carrier.

6. The multi-vision bond alignment method of claim 5, wherein, The control of the first carrier to restore to the determined reference position to obtain the current position and to obtain the image of the third mark pattern at the current position through the window region of the second carrier includes: controlling the first carrier to move so that the first carrier enters the visual field of all the first vision subsystems and the second vision subsystems and restores to a current position according to the determined reference position; controlling each third image acquisition device to obtain the image of the third mark pattern at the current position through the window region of the second carrier; The comparison of the image of the third mark pattern at the current position with the image of the third mark pattern at the reference position to obtain the alignment deviation between the reference position and the current position includes: corresponding comparison of the image of the third mark pattern at the current position with the image of the third mark pattern at the reference position to obtain the alignment deviation between the reference position and the current position.

7. The multi-vision bond alignment method of claim 1, wherein, The window region of the second carrier includes a transparent window region and / or a window region provided with a light-transmitting material.

8. The multi-vision bond alignment method of claim 1, wherein, The first carrier and the second carrier are upper and lower carriers, and the first carrier is an upper carrier or a lower carrier.

9. A multi-vision bonding alignment system, characterized by, It includes: The first carrier is provided with a mark pattern region, and the mark pattern region is provided with a third mark pattern; The second carrier is provided with a window region, and the image of the third mark pattern on the first carrier can be obtained through the window region; the third mark pattern and the window region are both arranged on the same side of the wafer fixing region; Each first vision subsystem includes a first image acquisition device and a second image acquisition device with opposite image acquisition directions and coaxial optical axes; the at least two first vision subsystems are used to observe the positions of the wafer on the first carrier and the wafer on the second carrier in sequence, so that at least two pairs of corresponding mark patterns of the two wafers are both located at the visual centers of the corresponding first vision subsystems, and the reference position of the first carrier is determined by the visual center position; and Each second vision subsystem includes a third image acquisition device; The at least two second vision subsystems are used to observe the third mark pattern before bonding to determine the alignment deviation between the current position and the reference position of the first carrier, so as to control the movement of the first carrier to eliminate the alignment deviation.

10. The multi-vision bond alignment system of claim 9, wherein, The first carrier and the second carrier are upper and lower carriers, and the first carrier is an upper carrier or a lower carrier.

11. The multi-vision bond alignment system of claim 9, wherein, The mark pattern region includes a transparent window region opened on the first carrier and / or a region originally present on the first carrier.

12. The multi-vision bond alignment system of claim 9, wherein, The window region of the second carrier includes a transparent window region and / or a window region provided with a light-transmitting material.

13. The multi-vision bond alignment system of claim 9, wherein, The first image acquisition device and the second image acquisition device are fixedly connected by a mechanical arm, so that the relative positions of the first image acquisition device and the second image acquisition device are fixed. 14.A computer device, comprising at least one computer storage medium and at least one processor, wherein the at least one computer storage medium stores a control program of the multi-vision key alignment method according to any one of claims 1 to 8, and the at least one processor is configured to execute the control program stored in the at least one computer storage medium.

15. A computer-readable storage medium, characterized in that, The medium stores a program, which can be loaded and executed by the processor to perform the multi-vision key alignment method according to any one of claims 1 to 8.

Images