TDI camera focus position calibration method and wafer inspection apparatus

Abstract

The application provides a TDI camera focusing position calibration method and wafer detection equipment, and belongs to the technical field of semiconductor optical detection. The TDI camera focusing position calibration method comprises generating a focusing matrix, aligning Trombone coordinates, registering the focusing matrix, and generating an optimal focusing position. The optimal focusing position is solved by a Grid Search algorithm. The application provides a TDI camera optimal focusing position calibration method for wafer bright field defect detection. The optimal focusing position is calculated by registering the focusing matrix generated by two-dimensional search, the influence of system error is reduced, the calibration accuracy of the TDI camera is improved, and the application in the field of wafer optical detection is facilitated.

Description

Technical Field

[0001] This invention belongs to semiconductor optical inspection technology, specifically relating to a TDI camera focus position calibration method and wafer inspection equipment. Background Technology

[0002] Bright field inspection (BFI) is a type of inspection equipment used in semiconductor manufacturing. It utilizes a high-resolution TDI camera to detect surface defects on wafers after photolithography. It can quickly and accurately detect surface defects at the 10+ nanometer level and is a very important inspection device in the semiconductor manufacturing field.

[0003] TDI (Time Delayed Integration) cameras are a common type of industrial camera used in BFI wafer inspection systems. They utilize multiple exposures of the same moving target through multi-stage photosensitive units, employing line-by-line charge superposition to improve the image's signal-to-noise ratio and sensitivity. They offer advantages such as fast response speed and wide dynamic range. The focusing process of a TDI camera can be controlled by object distance and image distance, which indirectly correspond to the Z-axis and Trombone position of the hardware system, respectively. Generally, a two-dimensional search method is used, sequentially adjusting different Z-axis and Trombone positions to determine the optimal focus position based on changes in image sharpness (e.g., ...). Figure 1 (As shown). However, hardware repeatability errors and resolution, minor jitter caused by environmental and random noise, and image sharpness instability caused by the optical system can all seriously affect the actual calibration results.

[0004] In semiconductor inspection equipment (especially high-end optical systems such as photolithography, mask inspection, and wafer metrology), a trombone typically refers to an optical delay line or optical path adjustment mechanism that allows for precise adjustment of the optical path length. Its core function is optical path compensation / matching. In precision optical systems such as interferometers, white light interferometers, and phase-shift measurements, it is necessary to precisely adjust the optical path difference (OPD) between the reference light and the sample light to make them equal or maintain a specific difference in order to generate interference fringes or optimize the signal. The trombone structure linearly changes the optical path length by moving a group of mirrors; its sliding mechanism is similar to the U-shaped telescoping tube of a trombone, hence its name.

[0005] The high magnification objectives and tiny defect sizes of TDI cameras place higher demands on camera calibration accuracy. Systematic errors in hardware, environmental changes, and the degradation of optical components can all severely affect subsequent inspection accuracy. Traditional calibration methods rely heavily on hardware readings, which introduces a large amount of systematic errors and environmental noise. Therefore, recalibration is required at specific intervals to maintain the accuracy of the optimal focus position. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, the present invention aims to provide a TDI camera focus position calibration method and a wafer inspection device, which can solve the above-mentioned problems.

[0007] Design Principle: The focusing curve of a TDI camera is highly sensitive to the Z-axis and Trombone positions; minute changes can be detected through variations in image sharpness. The focusing curve accurately reflects the relationship between object distance, image distance, and image quality, and is determined by the optical characteristics of the optical path. Hardware instability does not cause changes in the focusing curve. To reduce reliance on hardware readings and improve the stability of optimal focusing position calibration for TDI cameras, this invention utilizes this characteristic to propose a novel method for calibrating the optimal focusing position of a TDI camera. This method effectively avoids the significant errors and failure risks associated with solely relying on hardware readings, improving the calibration accuracy of the TDI camera and facilitating subsequent defect detection. The overall design is as follows: During initial calibration, the focusing matrix obtained from a two-dimensional search is stored as reference data, along with the Z-axis and Trombone reading positions corresponding to the optimal focus (referred to as the optimal focusing position). In subsequent calibration processes, new focusing data is compared with the reference data to calculate the offset value of the optimal focusing position, and the initial optimal focusing position is corrected to obtain the calibrated optimal focusing position. The specific scheme is as follows.

[0008] A TDI camera focus position calibration method includes the following steps: S1, generating a focus matrix. During the two-dimensional search focusing process, different Z-axis positions of the wafer inspection system and the Trombone mechanism correspond to an image sharpness value. The entire focusing process generates a complete image sharpness matrix, i.e., the focus matrix; S2, aligning the Trombone coordinates. The Trombone error is... Before registration, the coordinates of the Trombone mechanism are aligned to ensure the calibration matrix is ​​accurate. With reference matrix The Trombone coordinates are consistent, limiting the search area to... S3, Registration of the focus matrix, and calibration matrix. Select Window area of ​​size ,move With reference matrix By comparing different regions, the region with the smallest mean absolute error is found, thus obtaining the search position with the smallest mean absolute value. This is the offset value of the optimal focus position, achieving registration; S4, generate the optimal focus position based on the registration result. Calculate the focus matrix Corresponding optimal focus position , In the formula Z-axis and Trombone position coordinates corresponding to the optimal focal plane position .

[0009] Furthermore, in the process of generating the focus matrix, the two-dimensional search includes: the focus matrix The The row indicates that the Z-axis coordinate is fixed. The process of a mobile Trombone organization performing a one-dimensional search; The first of the matrix The column indicates that the Trombone mechanism coordinates are fixed. The process of moving along the Z-axis to perform a one-dimensional search.

[0010] Furthermore, the optimal focus position The solution steps include: S31, given a reference matrix and calibration matrix ,initialization The value of S32; find the window area. and S33. Accumulate the absolute error of the window region and calculate the average absolute error value; S34. Traverse the remaining... Repeat steps S32-S33 to calculate all mean absolute error values; in step S35, filter the search position corresponding to the smallest mean absolute error value among all mean absolute error values. As the optimal focus position, the offset value of the optimal focus position is obtained.

[0011] Furthermore, the optimal focus position is determined using the Grid Search algorithm. .

[0012] The present invention also provides a wafer inspection device, which includes an EFEM device and an inspection device; the EFEM device includes an EFEM controller and a robotic arm with a fork disposed in the EFEM chamber; the inspection device includes an inspection frame, a motion stage, an optomechanical module, and an inspection processor disposed in the inspection chamber; the inspection processor controls the Z-axis of the optomechanical module and the Trombone mechanism based on the aforementioned TDI camera focus position calibration method to achieve optical focusing of the TDI camera.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This application provides a method for calibrating the optimal focus position of a TDI camera for wafer bright field defect detection. The optimal focus position is calculated by registering a focus matrix generated by a two-dimensional search, which reduces the impact of system errors, improves the calibration accuracy of the TDI camera, and facilitates its application in the field of wafer optical inspection. Attached Figure Description

[0014] Figure 1 A schematic diagram illustrating the focus and image sharpness in a two-dimensional search.

[0015] Figure 2 A flowchart illustrating the focus position calibration method for a TDI camera;

[0016] Figure 3 This diagram illustrates the two-dimensional search for focus and image sharpness to obtain the optimal focus position using this method.

[0017] Figure 4 This is a schematic diagram of a wafer inspection equipment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] A method for calibrating the focus position of a TDI camera, the method is based on two-dimensional scanning controlled by the Z-axis and Trombone mechanism of the wafer inspection equipment, and calibrates the focus position of the TDI camera based on registration of two-dimensional search image data, see [link to relevant documentation]. Figure 2 The TDI camera focus position calibration method includes the following.

[0020] S1. Generate a focus matrix. During the two-dimensional search and focusing process, different Z-axis positions of the wafer inspection system and the Trombone mechanism correspond to an image sharpness value. The entire focusing process generates a complete image sharpness matrix, i.e., the focus matrix.

[0021] Specifically, during the two-dimensional search and focusing process, different Z-axis and Trombone mechanism positions correspond to a specific image sharpness value. The entire focusing process can generate a complete image sharpness matrix (referred to as the focusing matrix, as shown in Table 1 below).

[0022] Table 1. Examples of image sharpness matrices corresponding to different Z-axis coordinates and Trombone coordinates.

[0023]

[0024] Assuming the search coordinates along the Z-axis In order Search coordinates of the Trombone organization In order If m and n are positive integers, then the focusing matrix is... It can be represented as a The matrix whose coordinates The image sharpness at that location can be expressed as , .

[0025] In the formula, The first of the matrix The row indicates that the Z-axis coordinate is fixed. The process of a mobile Trombone organization performing a one-dimensional search; The first of the matrix The column indicates that the Trombone mechanism coordinates are fixed. The process involves moving along the Z-axis to perform a one-dimensional search, thus completing the two-dimensional search. During this process, an appropriate search range is selected so that the search area encompasses as much of the focus curve's peak as possible. It should be noted that the focus curve is more sensitive to the Z-axis and less sensitive to the Trombone's position, so the Trombone reading is given greater weight during the search.

[0026] S2 and Trombone coordinates are aligned, with Trombone's error being... Before registration, the coordinates of the Trombone mechanism are aligned to ensure the calibration matrix is ​​accurate. With reference matrix The Trombone coordinates are consistent, limiting the search area to... .

[0027] The Z-axis coordinate has a nanometer-scale dimension, with a repeatability error of 100nm-400nm, while the Trombone coordinate has a micrometer-scale dimension, with a repeatability error of 1µm, resulting in a significantly smaller relative error. Therefore, the search range of the Trombone coordinate can be controlled within its error range. If the Trombone error is... Before registration, the Trombone coordinates are aligned to make the calibration matrix With reference matrix The Trombone coordinates are consistent, limiting the search area to... This improves the registration efficiency of the focus matrix.

[0028] S3. Registration of the focus matrix and calibration matrix. Select Window area of ​​size ,move With reference matrix By comparing different regions, the region with the smallest mean absolute error (MAE) is found, thus obtaining the search position with the smallest mean absolute value. This is the offset value for the optimal focus position, achieving registration. Window area It should include the peak area of ​​the focus curve as much as possible.

[0029] Assumption and They represent and In coordinates The corresponding image sharpness, The offset distance between the Z-axis and the Trombone matrix is ​​represented by the following formula. The registration process of the focus matrix can be expressed as follows.

[0030] .

[0031] In the formula, , .

[0032] The solution space for the above problems is relatively small, and they can be solved directly using the Grid Search algorithm. Specifically, the optimal focus position is determined based on the Grid Search algorithm. The solution steps include the following.

[0033] S31, Given a reference matrix and calibration matrix ,initialization The value of .

[0034] S32. Find the window region and The image sharpness value at the corresponding location.

[0035] S33. Accumulate the absolute errors of the window region and calculate the average absolute error value.

[0036] S34, Traversing the Remaining Items Repeat steps S32-S33 to calculate all mean absolute error values.

[0037] S35. Filter the search position corresponding to the smallest error value among all mean absolute error values. As the optimal focus position, the offset value of the optimal focus position is obtained.

[0038] S4. Generate the optimal focus position based on the registration result. Calculate the focus matrix Corresponding optimal focus position , In the formula Z-axis and Trombone position coordinates corresponding to the optimal focal plane position .

[0039] Specifically, assuming the two-dimensional search focus matrix corresponding to the initial calibration position is: The optimal focal plane position corresponds to the Z-axis and the Trombone mechanism position as follows: , No. The focus matrix corresponding to the secondary calibration is: The offset error obtained from registration with the focus matrix It can be calculated Corresponding optimal focus position , A schematic diagram showing the Z-axis and Trombone mechanism positions corresponding to the optimal focal plane position obtained through the above method is shown below. Figure 3 , Figure 3 Corresponding to the data in Table 1, the optimal focal plane position was finally obtained. Figure 3 Dense area on the left side of the middle.

[0040] The above-mentioned TDI camera focus position calibration method is applicable to the optical inspection of wafers, including bright field and dark field inspection, as well as wafer inspection with and without patterns.

[0041] A wafer inspection device, see Figure 4 The wafer inspection equipment includes an EFEM device 100 and an inspection device 200; the EFEM device 100 includes an EFEM controller 1 and a robotic arm 2 with a fork, both located in the EFEM chamber; the inspection device 200 includes an inspection rack 3, a motion stage 4, an optical-mechanical module 5, and an inspection processor 6, all located in the inspection chamber; the inspection processor 6 controls the Z-axis and Trombone mechanism of the optical-mechanical module 5 based on the aforementioned TDI camera focus position calibration method to achieve optical focusing of the TDI camera.

[0042] The present invention also provides a computer-readable storage medium storing computer instructions thereon, which, when executed, perform the steps of the aforementioned method. The method is described in detail in the foregoing section and will not be repeated here.

[0043] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, including permanent and non-permanent, removable and non-removable media. Information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient media, such as modulated data signals and carrier waves.

[0044] The present invention also provides a terminal, including a memory and a processor. The memory stores data provider information and computer instructions executable on the processor. When the processor executes the computer instructions, it performs the steps of the aforementioned method. The method is described in detail in the foregoing section and will not be repeated here.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for calibrating the focus position of a TDI camera, characterized in that, The method includes the following steps: S1. Generate a focus matrix. During the two-dimensional search and focusing process, each of the different Z-axis and Trombone mechanism positions of the wafer inspection system corresponds to an image sharpness value. The entire focusing process generates a complete image sharpness matrix, i.e., the focus matrix. S2 and Trombone coordinates are aligned, with Trombone's error being... Before registration, the coordinates of the Trombone mechanism are aligned to ensure the calibration matrix is ​​accurate. With reference matrix The Trombone coordinates are consistent, limiting the search area to... ; S3. Registration of the focus matrix and calibration matrix. Select window area ,move With reference matrix By comparing different regions, the region with the smallest mean absolute error is found, thus obtaining the search position with the smallest mean absolute value. This is the offset value of the optimal focus position, which is used to achieve registration; S4. Generate the optimal focus position based on the registration result. Calculate the focus matrix Corresponding optimal focus position , In the formula Z-axis and Trombone position coordinates corresponding to the optimal focal plane position .

2. The TDI camera focus position calibration method according to claim 1, characterized in that, The two-dimensional search during the generation of the focus matrix includes: the focus matrix The The row indicates that the Z-axis coordinate is fixed. The process of a mobile Trombone organization performing a one-dimensional search; The first of the matrix The column indicates that the Trombone mechanism coordinates are fixed. The process of moving along the Z-axis to perform a one-dimensional search.

3. The TDI camera focus position calibration method according to claim 1, characterized in that: Optimal focus position The solution steps include: S31, Given a reference matrix and calibration matrix ,initialization The value; S32. Find the window region and The image sharpness value at the corresponding location; S33. Accumulate the absolute errors of the window region and calculate the average absolute error value; S34, Traversing the Remaining Items Repeat steps S32-S33 to calculate all mean absolute error values. S35. Filter the search position corresponding to the smallest error value among all mean absolute error values. As the optimal focus position, the offset value of the optimal focus position is obtained.

4. The TDI camera focus position calibration method according to claim 3, characterized in that: The optimal focus position is determined using the Grid Search algorithm. .

5. The TDI camera focus position calibration method according to claim 1, characterized in that: The TDI camera focus position calibration method is suitable for optical inspection of wafers.

6. A wafer inspection apparatus, comprising an EFEM device (100) and an inspection device (200); the EFEM device (100) comprising an EFEM controller (1) and a robotic arm (2) with a fork disposed in an EFEM chamber; the inspection device (200) comprising an inspection rack (3), a motion stage (4), an optomechanical module (5), and an inspection processor (6) disposed in an inspection chamber; characterized in that: The detection processor (6) controls the Z-axis and Trombone mechanism of the optical engine module (5) to achieve optical focusing of the TDI camera based on the TDI camera focus position calibration method according to any one of claims 1-4.

Images