A multiple overlay alignment method for sensor chip projection lithography machine

Abstract

The present invention proposes a multiple overlay alignment method for a sensor chip projection lithography machine, which belongs to the field of lithography and includes a primary and secondary overlay alignment algorithm. The primary overlay alignment algorithm identifies the outer contour of the diaphragm, calculates its position coordinates in the image and the deviations in the X and Y directions, and is used to determine the position of the diaphragm in a single exposure. The secondary overlay alignment algorithm, for a diaphragm with a layer of pattern, accurately calculates the center coordinates of the diaphragm and the deviations in the X, Y and rotation angles (R direction) by identifying the left and right positioning marks in the pattern. Through these deviations, the air-floating table workpiece stage and mask stage of the lithography machine can be compensated in real time, thereby ensuring accurate exposure of the diaphragm. The alignment method of the present invention significantly improves the alignment accuracy of each layer of pattern in the manufacturing process of the sensor chip and reduces the defects caused by alignment errors.

Description

Technical Field

[0001] The present invention belongs to the field of photolithography, comprises an image processing algorithm, and is specifically a multiple overlay alignment method for a sensor chip projection photolithography machine, which is used to realize overlay alignment exposure in the sensor chip projection photolithography machine. Background Art

[0002] In semiconductor manufacturing, photolithography is the core process for building microcircuits, and its accuracy directly affects device performance and production yield. As electronic devices develop towards smaller sizes and higher integration, the requirements for photolithography alignment technology are also increasing. Overlay alignment is a key step in the photolithography process to ensure accurate alignment of multiple layers of patterns. This technology is usually divided into two stages: primary overlay and secondary overlay, each of which performs different tasks.

[0003] The one-shot overlay alignment algorithm focuses on the initial positioning of the diaphragm. By identifying the outer contour of the diaphragm, the algorithm calculates the coordinates of the diaphragm's position in the image, as well as the deviations in the X and Y directions. This process not only determines the position of the diaphragm during the first exposure, but also lays the foundation for subsequent pattern formation. The successful implementation of one-shot overlay is critical to the entire lithography process, because if it is not accurately aligned, the subsequent pattern superposition will face a high risk, ultimately leading to product defects.

[0004] After one overlay, the first layer of pattern will be formed on the film, providing a basis for the second overlay. The second overlay alignment algorithm focuses on identifying the positioning marks (such as left and right marks) in the existing pattern to accurately calculate the center coordinates of the film and the deviations in the X, Y and rotation angles (R). This process significantly improves the alignment accuracy, ensuring that the new pattern can be accurately superimposed on the formed pattern in subsequent exposure. The accuracy of the second overlay is more complicated. If errors occur, multi-layer circuit design will be difficult to implement smoothly. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a multiple overlay alignment method for a sensor chip projection lithography machine, including a primary and secondary overlay alignment algorithm. The primary overlay alignment algorithm identifies the outer contour of the diaphragm, calculates its position coordinates in the image and the deviations in the X and Y directions, and is used to determine the position of the diaphragm in one exposure. The secondary overlay alignment algorithm, for a diaphragm with a layer of pattern, accurately calculates the center coordinates of the diaphragm and the deviations in the X, Y and rotation angles (R direction) by identifying the left and right positioning marks in the pattern. Through these deviations, the air-floating table workpiece stage and mask stage of the lithography machine can be compensated in real time to ensure accurate exposure of the diaphragm.

[0006] The overlay alignment algorithm of the present invention is systematically integrated in technical implementation, and fully considers the functional complementarity and sequential relationship of the primary and secondary overlay. The primary overlay alignment is not only responsible for the initial positioning of the diaphragm, but also the key guarantee for the subsequent process. It ensures high positioning accuracy during the first exposure by accurately identifying the external contour of the diaphragm. On this basis, the secondary overlay alignment achieves higher alignment accuracy by finely analyzing the existing pattern and using the positioning marks in the pattern. This method solves the potential errors caused by relying only on one alignment in traditional technology, especially when there are already multi-layer structures on the diaphragm, the accumulation of errors may seriously affect the performance of the device. Compared with the traditional method, the algorithm of the present invention not only improves the accuracy of alignment, but also enhances the real-time compensation capability of the lithography machine. By real-time monitoring of the deviation, the air-floating workpiece stage and the mask stage can be adjusted in real time to ensure the accurate superposition of each layer of pattern. This efficient and dynamic alignment strategy adapts to the fast pace and high-precision requirements of modern manufacturing, provides important support for the intelligent and automated process of semiconductor manufacturing, and promotes the industry to a higher technical level.

[0007] The specific technical solution proposed by the present invention is as follows: the method comprises the following steps:

[0008] Step S1: The air-floating worktable of the photolithography machine delivers the material tray with the membrane to the CCD off-axis alignment camera, and collects images of each membrane in turn;

[0009] Step S2: Identify the outer contour of the diaphragm by a primary overlay alignment algorithm, and calculate the coordinates of the diaphragm in the image and the deviations in the X and Y directions;

[0010] Step S3: using a secondary overlay alignment algorithm, identifying the left and right positioning marks of the film with a layer of pattern, and calculating the center coordinates of the film in the image and the deviations of the rotation angles in the X, Y and R directions through the left and right positioning marks;

[0011] Step S4: The air-floating workpiece stage of the lithography machine compensates for the X-direction and Y-direction deviations calculated in step S2 or S3, and the mask stage compensates for the R-direction deviation in step S3. After these compensations are completed, the film is sent to the exposure position for precise exposure.

[0012] The beneficial effects of the present invention are as follows:

[0013] 1. Improve alignment accuracy: Through the primary and secondary overlay alignment algorithms, the outer contour of the diaphragm and the positioning marks of the existing pattern can be accurately identified and calculated respectively, effectively improving the alignment accuracy in the X, Y and R directions, and ensuring the accurate superposition of each layer of pattern.

[0014] 2. Enhanced image processing effect: Use image processing techniques such as median filtering, grayscale histogram analysis, opening and closing operations to effectively remove background noise and enhance image features, thereby improving the accuracy and reliability of matching and recognition between template images and diaphragm images.

[0015] 3. Automatic compensation: By calculating the deviation between the center of the film and the center of the image and the rotation angle, and performing real-time compensation on the air-floating workpiece stage and the mask stage, the precise adjustment of the film position and angle during the exposure process is ensured, reducing manual intervention and improving production efficiency.

[0016] 4. Adapt to multiple sizes and scaling ratios: When matching and identifying film marks, a flexible scaling ratio (0.99 to 1.01) and angle range are used to make the algorithm more robust and able to adapt to slight changes in film size, thereby improving the flexibility of overlay alignment.

[0017] 5. Improve the stability of lithography production: Through precise alignment and compensation mechanisms, the exposure defects caused by alignment errors can be significantly reduced, thereby improving the production stability and product yield of the sensor chip lithography process. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of the overall overlay alignment method of the present invention;

[0019] Figure 2 This is a one-time overlay alignment flow chart of the present invention;

[0020] Figure 3 This is a comparison diagram of the membrane before and after pretreatment of the present invention;

[0021] Figure 4 This is a diagram of the recognition result of one-time overlay alignment of the membrane of the present invention;

[0022] Figure 5 It is the secondary overlay alignment diagram of the present invention;

[0023] Figure 6 This is a comparison diagram of the alignment mark of the film before and after pretreatment of the present invention;

[0024] Figure 7 This is a comparison diagram of the membrane before and after pretreatment of the present invention;

[0025] Figure 8 This is a diagram of the secondary overlay alignment and recognition results of the membrane of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in each embodiment of the present invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above-mentioned purpose, the present invention adopts the following technical scheme.

[0027] like Figure 1 As shown, it is an overall flow chart of a multiple overlay alignment method for a sensor chip projection lithography machine, comprising the following steps:

[0028] Step S1: The air-floating worktable of the photolithography machine delivers the tray with the membrane to the CCD off-axis alignment camera, and collects images of each membrane in turn.

[0029] Step S2: Identify the outer contour of the diaphragm by using an overlay alignment algorithm, and calculate the coordinates of the diaphragm in the image and the deviations in the X and Y directions.

[0030] Step S3: Use the secondary overlay alignment algorithm to identify the film positioning mark of the existing pattern, and calculate the center coordinates of the film in the image and the deviations in the X direction, Y direction and rotation angle (R direction) through the left and right positioning marks.

[0031] Step S4: The air-floating workpiece stage of the lithography machine compensates for the X-direction and Y-direction deviations calculated in step S2 or S3, and the mask stage compensates for the R-direction deviation in step S3. After these compensations are completed, the film can be sent to the exposure position for precise exposure.

[0032] The method significantly improves the alignment accuracy of each layer pattern in the sensor chip manufacturing process and reduces defects caused by alignment errors.

[0033] like Figure 2 As shown, it is a flow chart of an overlay alignment algorithm in step S2 of the present invention, which includes the following steps:

[0034] Step S21: Convert the diaphragm image into a grayscale image.

[0035] Step S22: Use The pixel rectangle performs median filtering on the grayscale image to remove background noise. Then the absolute histogram of the grayscale value of the grayscale image is calculated, and the two thresholds with the largest grayscale difference (excluding 0 and 255) are selected as the maximum and minimum thresholds to perform binary processing on the grayscale image. Subsequently, the marked area is further smoothed and denoised using the opening operation, and the image features are enhanced through the closing operation. The membrane before and after the processing is as follows Figure 3 As shown in the figure, (a) is the original image of the membrane, and (b) is the image of the membrane after preprocessing.

[0036] Step S23: Calculate the contour set of the diaphragm image. Since the diaphragm is a cylindrical steel cup and its top view is circular, the center pixel coordinates of the diaphragm are obtained by finding the minimum circumscribed circle of the contour set. and pixel radius , Including X coordinates , Y coordinate , the recognition results are as follows Figure 4 shown.

[0037] Step S24: Calculate the ratio of the pixels of the membrane to the actual space size according to formula (1) ,in, is the physical radius of the diaphragm itself:

[0038] (1)

[0039] Step S25: Calculate the X-direction offset between the center of the diaphragm and the center of the image according to equations (2) and (3): , Y offset , that is, the required compensation value, where and The air-floating workpiece stage can accurately add this offset when moving to the exposure position to ensure the accuracy of one-time overlay alignment.

[0040] (2)

[0041] (3)

[0042] Furthermore, the secondary overlay alignment algorithm is to identify the left and right positioning marks of the film with a layer of pattern, and calculate the center coordinates of the film in the image and the deviations in the X direction, Y direction and rotation angle (R direction) through the left and right positioning marks, such as Figure 5 As shown, it is a flow chart of the secondary overlay alignment algorithm of the present invention, which includes the following steps:

[0043] Step S31: Load the diaphragm marker template image and convert it into a grayscale image.

[0044] Step S32: Calculate the grayscale histogram of the grayscale image of the diaphragm marker template, and obtain each pixel by traversing the image Gray value , determine the number of pixels corresponding to each grayscale level and their proportion, and establish a grayscale distribution model. Then, the diaphragm marking template image is smoothed to filter out background noise, the grayscale threshold of the step distribution is calculated based on the grayscale histogram, and the two thresholds with the largest grayscale difference (excluding 0 and 255) are selected for binarization. Next, the marked area is smoothed using an open operation to further reduce noise, and the image features are enhanced using a closed operation. The processed image is saved as an alignment template, and finally the minimum circumscribed circle pixel radius of the diaphragm marking template is calculated using Canny edge detection. and , the diaphragm marker template image before and after processing is as follows Figure 6 As shown. Among them, (a) the original image with the template marked on the left, (b) the original image with the template marked on the right, (c) the image after preprocessing with the template marked on the left, and (d) the image after preprocessing with the template marked on the right.

[0045] Step S33: Load the diaphragm image and convert it into a grayscale image.

[0046] Step S34: When processing the grayscale image of the diaphragm, first use The pixel rectangle performs mean filtering on the image to remove background clutter, and then performs local dynamic thresholding based on equation (4) Operation, where and Calculate the size of the local area, The diaphragm image is binarized based on the local dynamic threshold and overlaid on the original diaphragm image to highlight the features of the area to be detected. The results before and after image processing are shown in Figure 2. Figure 7 As shown, (a) is the original image of the membrane, and (b) is the image of the membrane after preprocessing.

[0047] (4)

[0048] in, For coordinates Gray value of

[0049] Step S35: Match and identify the diaphragm mark, set the search angle range and the zoom ratio from 0.99 to 1.01, use the alignment template pattern to perform a match search in the detection image, and calculate the match score. Filter the matching results according to the minimum match score and the maximum number of matches, further filter according to the maximum overlap, and finally sort the matching results according to the greediness, and the one with the highest score is used as the final alignment mark matching result. Finally, the pixel center coordinates of the left and right marks of the diaphragm are obtained. and And the scale ratio of the left and right markings of the diaphragm and , the recognition results are as follows Figure 8 shown.

[0050] Step S36: Calculate the ratio of the pixels of the left and right marks of the membrane to the actual space size according to equations (5-1) and (5-2): , , and then determine the physical center coordinates of the left and right marks of the diaphragm according to formula (6-1) (6-2);

[0051] (5-1)

[0052] (5-2)

[0053] (6-1)

[0054] (6-2)

[0055] in, , It is the physical radius of the left and right marks of the diaphragm.

[0056] Step S37: The physical center coordinates of the left and right marks of the diaphragm , , calculate the deviation between the physical center coordinates of the diaphragm and the center of the diaphragm image , and the rotation angle of the diaphragm As shown in equations (7), (8), (9), and (10), the air-floating workpiece stage adds the offset when it moves to the exposure position, and the mask stage moves the corresponding rotation angle, thereby ensuring the accuracy of the secondary overlay alignment.

[0057] (7)

[0058] (8)

[0059] (9)

[0060] (10)

[0061] in, and They are The X and Y offsets, and for The X and Y coordinates of and They are The X and Y coordinates of and They are The X and Y coordinates of the

Claims

1. A multiple overlay alignment method for a sensor chip projection lithography machine, characterized in that: The method comprises the following steps: Step S1: The air-floating worktable of the photolithography machine delivers the material tray with the membrane to the CCD off-axis alignment camera, and collects images of each membrane in turn; Step S2: Identify the outer contour of the diaphragm by a primary overlay alignment algorithm, and calculate the coordinates of the diaphragm in the image and the deviations in the X and Y directions; Step S3: using a secondary overlay alignment algorithm, identifying the left and right positioning marks of the film with a layer of pattern, and calculating the center coordinates of the film in the image and the deviations of the rotation angles in the X, Y and R directions through the left and right positioning marks; Step S4: the air-floating workpiece stage of the lithography machine compensates for the X-direction and Y-direction deviations calculated in step S2 or S3, and the mask stage compensates for the R-direction deviation in step S3. After these compensations are completed, the film is sent to the exposure position for precise exposure; The step S2 comprises the following steps: Step S21: converting the diaphragm image into a grayscale image; Step S22: Use pixel rectangles to perform median filtering on the grayscale image to remove background noise, then calculate the absolute histogram of the grayscale value of the image, select the two thresholds with the largest grayscale difference excluding 0 and 255 as the maximum and minimum thresholds, perform binarization on the image, then use an open operation to further smooth and reduce noise in the marked area, and use a closed operation to enhance the image features; Step S23: Calculate the contour set of the grayscale image. The diaphragm is a cylindrical steel cup, and the top view is circular. The center pixel coordinates and pixel radius of the diaphragm are obtained by finding the minimum circumscribed circle of the contour set. Step S24: Calculate the ratio of the pixels of the membrane to the actual space size; Step S25: Calculate the offset between the center of the film and the center of the image, that is, the required compensation value, and the air-floating workpiece stage can accurately add the offset when moving to the exposure position to ensure the accuracy of the one-time overlay alignment; In step S24, the ratio of the pixels of the membrane to the actual space size is calculated according to formula (1): : (1) in, is the physical radius of the diaphragm itself, is the central pixel radius of the diaphragm; In step S25, the X-direction offset between the center of the diaphragm and the center of the image is calculated according to equations (2) and (3): , Y offset ; (2) (3) in, and are the width and height of the field of view image, respectively. , are the X and Y coordinates of the center pixel of the diaphragm; The step S3 comprises the following steps: Step S31: loading the membrane marking template image and converting it into a grayscale image; Step S32: Calculate the grayscale histogram of the grayscale image of the diaphragm marking template image, obtain the grayscale value of each pixel by traversing the image, determine the number of pixels corresponding to each grayscale level and its proportion, and establish a grayscale distribution model. Then, smooth the template image to filter out background noise, calculate the grayscale threshold of the step distribution according to the grayscale histogram, select the two thresholds with the largest grayscale difference excluding 0 and 255 for binarization operation, then use the opening operation to smooth the marked area to further reduce noise, use the closing operation to enhance the image features, and save the processed image as an alignment template. Finally, calculate the minimum circumscribed circle pixel radius of the diaphragm marking template through Canny edge detection; Step S33: loading the diaphragm image and converting it into a grayscale image; Step S34: when processing the grayscale image of the diaphragm, first use the pixel rectangle to perform mean filtering on the grayscale image to remove background clutter, then perform local dynamic threshold operation, binarize the image based on the local dynamic threshold, and overlay it on the original diaphragm image to highlight the features of the area to be detected; Step S35: matching and identifying the diaphragm mark, setting the search angle range and the scaling ratio of 0.99 to 1.01, using the alignment template pattern to perform matching search in the detection image, and calculating the matching score, filtering the matching results according to the minimum matching score and the maximum matching number, further screening according to the maximum overlap, and finally sorting the matching results according to the greediness, and the highest score is used as the final alignment mark matching result, and finally obtaining the pixel center coordinates of the diaphragm mark and its scaling ratio; Step S36: Calculate the ratio of the pixels of the diaphragm mark to the actual space size, and then determine the physical center coordinates of the diaphragm mark; Step S37: Calculate the deviation between the physical center coordinates of the left and right marks of the diaphragm and the image center, as well as the rotation angle of the diaphragm. When the air-floating workpiece stage moves to the exposure position, the deviation is added, and the mask stage moves the corresponding rotation angle, thereby ensuring the accuracy of the secondary overlay alignment.

2. The method for multiple overlay alignment of a sensor chip projection lithography machine according to claim 1, characterized in that: In step S34, a local dynamic threshold is performed based on equation (4): Operation: (4) in, and Calculate the size of the local area, To compensate for the offset, For coordinates The gray value of .

3. The method for multiple overlay alignment of a sensor chip projection lithography machine according to claim 2, characterized in that: In step S36, the ratio of the pixels of the left and right marks of the membrane to the actual space size is calculated according to equations (5-1) and (5-2): , , and then determine the physical center coordinates of the left and right marks of the diaphragm according to formula (6-1) (6-2); (5-1) (5-2) (6-1) (6-2) in, , is the physical radius of the left and right marks of the diaphragm, and the pixel center coordinates of the left and right marks of the diaphragm are and , the scale ratio of the left and right marks of the diaphragm is and , the minimum circumscribed circle pixel radius of the left and right marker templates of the diaphragm and .

4. The multiple overlay alignment method for a sensor chip projection lithography machine according to claim 3, characterized in that: In step S37, the physical center coordinates of the left and right marks of the diaphragm are , , calculate the deviation between the physical center coordinates of the diaphragm and the center of the diaphragm image , and the rotation angle of the diaphragm , as shown in formula (7), formula (8), formula (9), and formula (10): (7) (8) (9) (10) in, and They are The X and Y offsets, and for The X and Y coordinates of and They are The X and Y coordinates of and They are The X and Y coordinates of the

Images