A method for compensating exposure pattern grid accuracy

Abstract

The application discloses a compensation method for exposure pattern grid precision, which comprises the following steps: firstly, obtaining an exposure calibration plate, wherein the exposure calibration plate comprises basic points arranged in a matrix; then, selecting the basic points of the exposure calibration plate as alignment points, performing alignment exposure on the exposure calibration plate, and obtaining detection points, wherein the detection points and the basic points one-to-one correspond to a group of test points; acquiring images of the test points, and obtaining exposure compensation values according to the images of the test points. The compensation method for exposure pattern grid precision only needs to perform whole-plate exposure twice, acquires two times of coordinate data, and greatly reduces the amount of coordinate data extraction. Meanwhile, the method adjusts the difference between the data of the detection points and the data of the basic points, reduces the superposition deviation caused by the measurement error of the camera itself, and optimizes the uniformity of the whole grid and the mutual effect between double platforms.

Description

Technical Field

[0001] This invention relates to a method for compensating exposure patterns, and more particularly to a method for compensating the grid precision of exposure patterns. Background Technology

[0002] Photolithography is widely used in semiconductor and PCB manufacturing, serving as a crucial step in the production of semiconductor devices, chips, and PCB products. It prints feature patterns onto a substrate surface to ultimately obtain the desired pattern structure based on the circuit design. Direct-write exposure systems utilize digital light processing technology, employing programmable digital mirrors to edit different desired pattern structures, enabling rapid pattern switching. This not only reduces costs but also shortens process time, leading to its widespread application in photolithography. To accurately print the pattern on the workpiece, compensation data for the exposure pattern must be obtained before exposure in a direct-write exposure system. Currently, obtaining this compensation data requires multiple repeated exposures for verification. First, misaligned exposure is used to obtain global misalignment compensation data. Then, based on this global misalignment compensation data, another misaligned exposure is performed to verify the global misalignment compensation data and obtain multi-channel compensation data. Next, misaligned exposure is performed based on the obtained multi-channel compensation data to obtain and verify inter-field stitching compensation data. Finally, aligned exposure is performed based on the inter-field stitching compensation data to obtain and verify the outer layer alignment compensation data. This process requires repeating the process seven times for the entire plate, making it cumbersome, complex, and involving a large amount of data processing. Meanwhile, the exposure pattern is adjusted only based on the measurement data, and the exposure distortion includes the superposition of measurement distortion and repeatability deviation, so the error of the measurement system cannot be effectively avoided. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method for compensating the accuracy of exposure pattern grids with high efficiency while reducing data processing volume.

[0004] To achieve the above objective, the method for compensating the accuracy of the exposure pattern grid includes the following steps: First, an exposure calibration plate is obtained, which includes base points arranged in rows and columns; then, the base points of the exposure calibration plate are selected as alignment points, and the exposure calibration plate is subjected to alignment exposure to obtain detection points, wherein the detection points and the base points correspond one-to-one to form a set of test points; images of the test points are acquired, and exposure compensation values ​​are obtained based on the images of the test points.

[0005] Preferably, the calibration point of the calibration plate is the light-blocking part, and the rest is the light-transmitting part.

[0006] Preferably, the exposure calibration plate includes a substrate coated with photoresist and a calibration plate fixed above the substrate, wherein the base point is the calibration point of the calibration plate.

[0007] Preferably, the edge of the detection point is outside the range of the base point.

[0008] Preferably, when verifying the exposure compensation value, the base point is used as the alignment point, and the image is exposed again. The exposed image is an array of verification detection points that correspond one-to-one with the base point. The outline of the verification detection point is smaller than the outline of the base point, the verification detection point is located inside the base point, and there is a certain gap between the edges of the detection point and the calibration point. The exposure compensation value is added during exposure to obtain the verification test point, and the image of the test point is acquired to verify the accuracy of the exposure compensation value.

[0009] Preferably, the exposure calibration board is obtained by mask exposure. The calibration board is used as a mask and placed on a substrate coated with photoresist for mask exposure. The exposed substrate is the exposure calibration board. The base point is the point on the exposure calibration board that corresponds to the calibration point on the calibration board.

[0010] Preferably, the edge of the detection point is within the range of the base point.

[0011] Preferably, the exposure compensation value includes global misalignment compensation value, multi-path compensation value, inter-field stitching compensation value, and outer layer alignment translation compensation value.

[0012] Preferably, the global shear compensation value is obtained by calculating the degree of difference between the X coordinate of the detection point and the X coordinate of the base point from top to bottom.

[0013] Preferably, the multi-path compensation value is calculated by the difference in the Y-direction between the detection point and the corresponding base point. The difference in the Y-direction between each detection point and the corresponding base point is used as the minuend, and the difference in the Y-direction between the first detection point of the strip and the base point is used as the subtrahend. The difference obtained is the multi-path compensation amount.

[0014] Preferably, the inter-field splicing compensation value is divided into X-direction inter-field splicing compensation value and Y-direction inter-field splicing compensation value. When calculating the X-direction inter-field splicing compensation amount, the calculation starts from the second strip. The minuend is the average difference between the X-direction coordinates of each detection point in each strip after global misalignment compensation and the X-direction coordinates of the base point. The subtrahend is the average difference between the X-direction coordinates of each detection point in the X-direction after global misalignment compensation and the X-direction coordinates of the calibration plate in the first strip. When calculating the Y-direction inter-field splicing compensation value, the calculation starts from the second strip. The minuend is the difference between the Y-direction coordinates of the first detection point in each strip and the base point. The subtrahend is the difference between the Y-direction coordinates of the first detection point in the first strip and the corresponding base point.

[0015] Preferably, the outer layer alignment translation compensation value is the difference between the XY coordinate value of the first detection point of the first strip and the XY coordinate value of the base point.

[0016] Compared with existing technologies, the method for compensating the accuracy of the exposure pattern grid described in this invention only requires two full-page exposures to obtain coordinate data twice, greatly reducing the amount of coordinate data extracted. At the same time, adjustments are made based on the difference between the measured data of the detection point and the data of the base point, reducing the superposition deviation caused by the camera's own measurement error, and optimizing the uniformity of the overall grid and the interlocking effect between the two cameras. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of existing methods for compensating for the accuracy of exposure pattern grids.

[0018] Figure 2 This is a schematic diagram of the method for compensating the accuracy of the exposure pattern grid according to the present invention.

[0019] Figure 3 This is a block diagram of the method for compensating the accuracy of the exposure pattern grid according to the present invention.

[0020] Figure 4 This is an exposed image of another embodiment of the method for compensating the accuracy of the exposure pattern grid of the present invention.

[0021] Figure 5 This is a schematic diagram of the test points.

[0022] Figure 6 This is a schematic diagram of the verification test points. Implementation

[0023] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described below with reference to specific embodiments shown in the accompanying drawings.

[0024] This invention provides a method for compensating the accuracy of an exposure pattern grid. First, an exposure calibration board is obtained, which includes base points arranged in rows and columns. Then, the base points of the exposure calibration board are selected as alignment points, and the exposure calibration board is subjected to alignment exposure to obtain detection points. The detection points and base points correspond one-to-one to form a set of test points. Images of the test points are acquired, and exposure compensation values ​​are obtained based on the images of the test points.

[0025] The exposure calibration plate can be obtained by mask exposure. Using the calibration plate as a mask, it is placed above a substrate coated with photoresist for exposure, resulting in the exposure calibration plate. The calibration points on the calibration plate are light-shielding portions, while the remaining portions are light-transmitting portions. The base points on the exposure calibration plate are not exposed to light. Preferably, the photoresist can be a dry film photoresist. To protect the calibration plate, it is preferable to fix it during exposure, reduce the motor's movement speed, and slow down the movement of the calibration plate and the substrate.

[0026] The alignment points can be selected from the four base points located at the four corners of the exposure calibration plate.

[0027] When the exposure calibration plate is aligned and exposed, the exposure image is an array pattern of detection points that correspond one-to-one with the base point. The detection points are located inside the base point, and there is a certain gap between the edges of the detection points and the base point.

[0028] Before exposure, the coordinates of the reference point need to be compared with the strip cutting position to confirm that each column of points falls within the corresponding strip. If they do not fall within the corresponding strip, the position of the reference point can be adjusted to ensure that each column of points falls within the corresponding strip, so as to calculate the inter-field stitching compensation. The reference point is usually the alignment point closest to the origin among the alignment points. The detection point and the reference point are selected as concentric shapes, such as both being circles with different radii, or both being squares with different side lengths.

[0029] The exposure compensation values ​​include global shear compensation values, multi-path compensation values, inter-field stitching compensation values, and outer layer alignment translation compensation values. The global shear compensation value is used to fill in the shear elements of the image transformation matrix; the multi-path compensation values ​​are sent to the synchronization board to adjust the advance or lag of the synchronization signal at different movement positions of the scanning axis; the inter-field stitching compensation value is used to fill in the lens position relationship; and the outer layer alignment compensation value is used to fill in the translation elements of the image transformation matrix.

[0030] The global misalignment compensation value is obtained by calculating the degree of difference between the X-coordinate of the detection point and the X-coordinate of the base point from top to bottom. The specific calculation method is as follows:

[0031]

[0032] Where (x0,y0) is the first detection point of each band, (xi,yi) is all the detection points involved in the calculation of each band, (X0,Y0) is the first base point of each band, (Xi,Yi) is all the base points involved in the calculation of each band, n is the number of detection points in a column of each band, and N is the number of exposure lenses.

[0033] It is the average of the ratio of the difference in X coordinate to the difference in Y coordinate between all the detection points (xi,yi) participating in the calculation of each band and the first detection point (x0,y0) of that band.

[0034] It is the average of the ratio of the difference in X coordinates to the difference in Y coordinates between all the base points (Xi, Yi) involved in the calculation of each band and the first base point (X0, Y0) of that band.

[0035] The multi-path compensation amount is calculated by the difference in the Y-direction between the detection point and the corresponding base point. Specifically, the difference in the Y-direction between each detection point and the corresponding base point is used as the minuend, and the difference in the Y-direction between the first detection point of the strip and the base point is used as the subtrahend. The difference obtained is the multi-path compensation amount.

[0036] Inter-field stitching compensation is divided into X-direction inter-field stitching compensation and Y-direction inter-field stitching compensation. When calculating the X-direction inter-field stitching compensation, it starts from the second strip. During the calculation, the average value after removing the global misalignment is the difference between the X-direction coordinates of each detection point in each strip and the X-direction coordinates of the base point. The minuend is the average of the differences between the X-coordinate difference of each detection point in each strip and the base point, the tangent of the corresponding global misalignment compensation value, and the difference between the Y-coordinate difference of each detection point and the first detection point. The subtrahend is the average of the differences between the X-coordinate values ​​of each detection point in the X-direction of the first strip and the X-coordinate value of the base point, minus the result of subtracting the global misalignment. When calculating the inter-field stitching compensation value in the Y-direction, the calculation also starts from the second strip. During the calculation, the minuend is the difference between the Y-coordinate of the first detection point in each strip and the base point, and the subtrahend is the difference between the Y-coordinate of the first detection point in the first strip and the corresponding base point.

[0037] The outer layer alignment translation compensation value is the difference between the XY coordinate value of the first detection point of the first strip and the XY coordinate value of the base point.

[0038] To verify the exposure compensation value, repeat the above steps, adding the exposure compensation value during exposure, obtain the verification test point, acquire the image of the verification test point, and verify the accuracy of the exposure compensation value.

[0039] The method for compensating for the accuracy of the exposure pattern grid only requires two exposures of the entire plate, acquiring coordinate data twice, significantly reducing the amount of coordinate data extracted. Simultaneously, adjustments are made based on the difference between the measured data at the detection points and the data at the base points, reducing the superposition deviation caused by the camera's own measurement errors, optimizing the overall grid uniformity and the interlocking effect between the two cameras. Furthermore, it does not damage the calibration plate, which has a high reusability. The exposure calibration plate can also directly use a substrate coated with photoresist and a calibration plate fixed on top of the substrate. The calibration points on the calibration plate are light-shielding parts, and the remaining parts are light-transmitting parts.

[0040] When the exposure calibration plate is aligned and exposed, the exposure image is an array pattern of detection points that correspond one-to-one with the base point. The outline of the detection point is larger than the outline of the calibration point, the calibration point is located inside the detection point, and there is a certain gap between the edges of the detection point and the calibration point.

[0041] Remove the calibration plate to obtain test points. The test points are in a ring shape, including the base point located inside the calibration point and the exposed test point located outside.

[0042] By acquiring images of the test points, the exposure compensation value is obtained using the calculation method of the first embodiment.

[0043] To obtain better X-direction inter-field stitching compensation, during alignment exposure, it is preferable to expose multiple supplementary points simultaneously, in addition to exposing the detection points corresponding to the calibration points. These supplementary points are located between the calibration points, matching the spacing of the exposure lenses in the exposure equipment. This reduces the positional differences between detection points and improves the accuracy of the X-direction inter-field stitching compensation. As shown in Tables 1 and 2, taking an exposure lens spacing of 52.5mm as an example, the relative positions of the corresponding exposure lenses are: 0, 52.5, 105, 157.5, 210, 262.5. The minimum spacing of the calibration plate is fixed. If the minimum spacing of the calibration plate is 10mm, the relative positions of the corresponding points on the calibration plate can only be selected as: 0, 50, 110, 160, 210, 260, with differences from the exposure lenses of: 0, -2.5, 5, 2.5, 0, -2.5, with a maximum difference of 5mm and a difference between them reaching 7.5mm. To reduce these differences, in addition to... In addition to exposing the test points corresponding to the calibration points, multiple supplementary points are simultaneously exposed to optimize the test point positions based on the positional relationship of the exposure lenses. After optimization, the relative positions of the test points are 0, 55, 105, 160, 210, and 265. For positions 55, 105, and 265 that cannot be obtained from the calibration board, they are obtained through the exposure of supplementary points. The base point data in the supplementary points are obtained from the base point data corresponding to the calibration points at adjacent positions. For example, for a test point with a relative position of 105, its base point data is obtained by averaging the base point data with a relative position of 100 and a relative position of 110. After optimization, the differences from the exposure lenses are 0, 2.5, 0, 2.5, 0, and 2.5, respectively, with a maximum difference of only 2.5mm, improving the accuracy of inter-field stitching compensation in the X direction.

[0044] Table 1. Relative positional relationships of exposure lenses and corresponding relative positional relationships of calibration plate test points.

[0045] Unit: mm

[0046] Table 2. Relative positional relationships of exposure lenses and corresponding relative positional relationships of test points with supplementary points.

[0047] Unit: mm

[0048] When verifying the exposure compensation value, the base point is used as the alignment point, and the image is exposed again. The exposed image is an array of verification detection points that correspond one-to-one with the base point. The outline of the verification detection point is smaller than the outline of the base point, and the verification detection point is located inside the base point. There is a certain gap between the edges of the detection point and the calibration point. An exposure compensation value is added during exposure to obtain verification test points. The images of the test points are then acquired to verify the accuracy of the exposure compensation value.

[0049] Obtaining and verifying exposure compensation values ​​using the above method eliminates the need to repeatedly create exposure calibration boards, making the method simpler.

Claims

1. A method for compensating exposure grid precision, first obtaining an exposure calibration plate, the exposure calibration plate comprising basic points arranged in rows and columns. Then the base points of the exposure calibration board are selected as the alignment points, the exposure calibration board is exposed by alignment to obtain detection points, and the detection points and the base points correspond to each other to form a group of test points; An image of the test points is acquired, and an exposure compensation value is acquired according to the image of the test points, wherein the exposure compensation value includes a global skew compensation value, a multi-path compensation value, an inter-field splicing compensation value and an outer alignment translation compensation value; When verifying the exposure compensation value, the base points are used as the alignment points, and exposure by alignment is performed again, so that an exposure image is an array pattern of verification detection points corresponding to the base points one by one, the exposure compensation value is added during exposure, verification test points are obtained, an image of the test points is acquired, and the accuracy of the exposure compensation value is verified.

2. The exposure grid accuracy compensation method according to claim 1, wherein: When the exposure calibration board is acquired, a calibration board is used, and the calibration points of the calibration board are light shielding parts, and the remaining parts are light transmission parts.

3. The exposure grid accuracy compensation method according to claim 2, wherein: The exposure calibration board includes a substrate coated with photoresist and a calibration board fixed above the substrate, and the base points are calibration points of the calibration board.

4. The exposure grid accuracy compensation method according to claim 3, wherein: The edges of the detection points are outside the range of the base points.

5. The exposure grid accuracy compensation method according to claim 4, wherein: The outline of the verification detection point is smaller than the outline of the base point, the verification detection point is located inside the base point, and the edges of the detection point and the calibration point have a certain gap.

6. The exposure grid accuracy compensation method according to claim 2, wherein: The exposure calibration board is obtained by mask exposure, the calibration board is used as a mask plate, and is placed above the substrate coated with photoresist for mask exposure, the substrate after exposure is the exposure calibration board, and the base points are points of the exposure calibration board corresponding to the positions of the calibration points on the calibration board.

7. The exposure grid accuracy compensation method according to claim 6, wherein: The edges of the detection points are within the range of the base points.

8. The exposure grid accuracy compensation method according to claim 1, wherein: The global skew compensation value is obtained by calculating the difference from top to bottom of the deviation value of the X coordinate of the detection point and the X coordinate of the base point.

9. The exposure grid accuracy compensation method of claim 1, wherein: The multi-path compensation value is calculated by the difference value of the Y direction of the detection point and the corresponding base point, the difference value of the Y direction of each detection point and the corresponding base point is used as the minuend, the difference value of the Y direction of the first detection point and the base point of the strip is used as the subtrahend, and the obtained difference value is the multi-path compensation value.

10. The exposure grid accuracy compensation method of claim 1, wherein: The inter-field splicing compensation value includes an X direction inter-field splicing compensation value and a Y direction inter-field splicing compensation value, when the X direction inter-field splicing compensation value is calculated, the calculation starts from the second strip, the average value of the difference value of the X direction coordinate of each detection point after global skew compensation and the X direction coordinate of the base point is used as the minuend, and the average value of the difference value of the X direction coordinate of each detection point after global skew compensation and the X direction coordinate of the calibration board of the first strip is used as the subtrahend; when the Y direction inter-field splicing compensation value is calculated, the calculation starts from the second strip, the difference value of the Y direction coordinate of the first detection point of each strip and the base point is used as the minuend, and the difference value of the Y direction coordinate of the first detection point of the first strip and the corresponding base point is used as the subtrahend.

11. The exposure grid accuracy compensation method of claim 1, wherein: The outer alignment translation compensation value is the difference value of the XY coordinate of the first detection point of the first strip and the XY coordinate of the base point.

Images