Correlation-based overlay key centering system and method

Abstract

To provide a centering system of an overlay key based on a correlation and a method thereof that can correctly search for center coordinates of an overlay key pattern.SOLUTION: An centering method of an overlay key on the basis of a correlation includes a rough search step of adjusting an overlay target image to the predetermined set size, calculating a first correlation value in the adjusted image, and calculating the rough center coordinate, and a fine search step of calculating a second correlation value on the basis of the rough center coordinate in the original image of the overlay target image and calculating the fine center coordinate.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to semiconductor processing, and more particularly to a system and method for centering an overlay key based on correlation in semiconductor processing. [Background technology]

[0002] In the semiconductor process, a photoresist is formed on the surface of a wafer, and in an exposure process using a stepper, the photoresist is transferred onto the photoresist on the wafer surface on a reticle, and the photoresist after the exposure process is developed. After that, a circuit pattern is formed on the wafer surface by a photolithography process in which the developed photoresist is etched on the wafer surface using an etching mask, and the photolithography process is repeated to form a multilayer film having a circuit pattern on the wafer surface, thereby manufacturing a semiconductor device.

[0003] In the semiconductor manufacturing process, the exposure process for forming a fine pattern on a semiconductor substrate involves coating a photoresist on the semiconductor substrate, applying heat to the semiconductor substrate coated with the photoresist to match a pattern formed on a mask to the surface of the semiconductor substrate, partially transmitting light to expose the photoresist at the corresponding portion, spraying a developer to remove the portion through which light was transmitted or the portion through which light was not transmitted during exposure by chemical action, and measuring the alignment state after forming a pattern on the semiconductor substrate to measure the vertical alignment degree.

[0004] Meanwhile, the vertical alignment of an upper thin film layer and a lower thin film layer for manufacturing semiconductor chips on a wafer is called overlay, and an optical overlay measurement device uses an overlay key to inspect the alignment of a pattern formed on a semiconductor substrate with a pattern formed in the current process to detect minute defects and detect defects in the semiconductor process.

[0005] An overlay process, which measures and inspects the alignment of fine patterns on a semiconductor substrate, checks the alignment of the upper and lower thin film layers using an overlay mark to determine whether a lower thin film layer pattern and an upper thin film layer pattern formed on a plurality of thin film layers consisting of multiple layers on a semiconductor substrate are accurately aligned.

[0006] Optical overlay measurement equipment performs extremely high-precision measurements in semiconductor processes to determine whether lower and upper thin film layer patterns formed in multiple thin film layers consisting of multiple layers on a semiconductor substrate stacked in multiple layers are accurately aligned, and whether circuit patterns are correctly aligned using overlay marks.

[0007] In order to measure the alignment of the overlay marks, a PR (Pattern Recognition) action must be performed first.

[0008] When moving a stage to a target (Overlay Key) position entered in a recipe, there is a limit to finding the exact position due to the repeatability of the hardware (robot, stage, etc.).

[0009] PR action is a technology that corrects this by detecting the center where the model image registered in the recipe matches the live image on the FOV (Field of View), and then moving the stage further by an offset.

[0010] However, the method of further stage movement through 1:1 matching between images is possible if the images registered in the recipe are replaced with images of any target across the wafer.

[0011] If there is a large variation between images of the overlay key due to process influences, PR failure may occur.

[0012] Therefore, there is a need for a technology that can compensate for PR failure, find the center coordinates of the overlay key, and perform accurate centering. [Prior art documents] [Patent documents]

[0013] Korean Patent No. 10-1604789 (March 14, 2016) Summary of the Invention [Problem to be solved by the invention]

[0014] The problem to be solved by the present invention is to provide a system and method for centering an overlay key based on correlation in order to overcome variations between images of the overlay key that occur due to process reasons and to perform PR (Pattern Recognition).

[0015] The object of the present invention is not limited to the object mentioned above, and other objects and advantages of the present invention not mentioned can be understood from the following description and can be more clearly understood from the embodiments of the present invention. In addition, it is easy to understand that the object and advantages of the present invention can be realized by the means and combinations thereof described in the claims. [Means for solving the problem]

[0016] According to an embodiment, a system for centering an overlay key based on correlation includes an overlay measurement device including at least one processor, which receives an overlay target image, adjusts the overlay target image to a size smaller than the entire image, calculates a first correlation value in the adjusted image to calculate a rough center coordinate, and calculates a second correlation value based on the rough center coordinate in an original image of the overlay target image to calculate a fine center coordinate.

[0017] According to one embodiment, a method for centering an overlay key based on correlation includes a rough search step of adjusting an overlay target image to a preset size, calculating a first correlation value in the adjusted image, and calculating rough center coordinates, and a fine search step of calculating a second correlation value based on the rough center coordinates in an original image of the overlay target image, and calculating fine center coordinates. Effect of the Invention

[0018] According to the embodiment, the alignment state of fine patterns on a semiconductor substrate can be accurately measured and inspected by accurately centering the overlay key based on the correlation.

[0019] Also, by accurately locating the coordinates of the overlay key pattern, the precision of the overlay measurement can be improved.

[0020] In addition, it is possible to prevent PR (Pattern Recognition) failures that may occur when there is a large variation in the overlay keys.

[0021] In addition, it is possible to compensate for PR misreading that may occur in the case of a layer that is sensitive to PR parameters.

[0022] The above-mentioned effects and specific effects of the present invention will be described in conjunction with the following description of the preferred embodiment of the invention. [Brief description of the drawings]

[0023] [Figure 1] 1 is a diagram for explaining a PR (Pattern Recognition) action. [Diagram 2] 1 is a flow diagram of a method for finding a center of an overlay key based on correlation according to one embodiment. [Diagram 3] 3 is a diagram illustrating the correlation search area of ​​FIG. 2; [Figure 4] 3 is a diagram illustrating the correlation search area of ​​FIG. 2; [Diagram 5] 3 is a diagram illustrating the correlation search area of ​​FIG. 2; [Figure 6] 3 is a diagram illustrating a method for finding a center of an overlay key based on the correlation of FIG. 2; [Figure 7] 3 is a diagram showing the definition of a template for calculating the correlation in FIG. 2. [Figure 8] 1 is a flow diagram of a correlation-based overlay key centering method according to one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0024] The above-mentioned objects, means and effects will be described in detail below with reference to the accompanying drawings, so that a person having ordinary skill in the art to which the present invention pertains can easily implement the technical idea of ​​the present invention. In describing the present invention, if a detailed description of the known technology according to the present invention is deemed to make the gist of the present invention unclear, the detailed description will be omitted. Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings are used to indicate the same or similar components.

[0025] Hereinafter, a system and method for centering an overlay key based on correlation to overcome variations between images of the overlay key that occur due to process reasons and to perform PR (Pattern Recognition) will be disclosed.

[0026] Before describing the correlation-based overlay key centering method according to the embodiment, the meanings of terms used in this specification will be defined.

[0027] For purposes of this specification, overlay may be defined as the vertical alignment of upper and lower thin film layers for fabricating semiconductor chips on a wafer.

[0028] In the specification, an overlay key can be defined as a key used to detect defects in a semiconductor process by inspecting the alignment state of a pattern formed on a semiconductor substrate in an optical overlay measurement device and a pattern formed in a current process to detect minute defects.

[0029] In the specification, the PRU recognizes the overlay key to perform overlay measurements, which may include centering the overlay key within a Field of View (FOV).

[0030] In the specification, the target finder may be implemented as a PC terminal equipped with a processor, an overlay measurement device, etc. In one example, the target finder may be an application that calculates the center coordinates of an overlay key based on a correlation.

[0031] FIG. 1 is a diagram for explaining a PR (Pattern Recognition) action.

[0032] When moving the stage to the target (overlay key) position, there is a limit to finding the exact position due to the effect of the repeatability of the measurement equipment.

[0033] PR action is a technology to correct this by detecting the position where the model image registered in the recipe matches the live image on the FOV as the center, and then moving the stage further by an offset.

[0034] However, the method of further stage movement through 1:1 matching between images is possible when the images registered in the recipe are replaced with images of any target across the wafer.

[0035] If there is a large variation between images of the overlay key due to process influences, PR failure may occur.

[0036] To this end, in the present embodiment, the center position of the overlay key is detected using a template matching function that searches for a portion of an input image that matches a small template image. In one example, the center position can be detected using a template matching function that searches for a portion of a live image, i.e., an image acquired while measuring, as an input image.

[0037] The vertical alignment of upper and lower thin film layers used to create semiconductor chips on a wafer is called overlay. An optical overlay measurement system uses an overlay key to inspect the alignment of a pattern formed on a semiconductor substrate with a pattern formed in the current process to detect minute defects and defects in the semiconductor process.

[0038] The correlation-based overlay key centering system and method according to an embodiment may measure similarity using a tool that can measure how similar an input function and a target function are. In one example, when applied to an image, a correlation value may be derived using a tool that can measure how similar both images are.

[0039] Correlation analysis, used in probability theory and statistics, is a method for analyzing the sectorial relationship between two variables. In this case, the strength of the relationship between the two variables is called correlation (or correlation coefficient).

[0040] There are various techniques for analyzing statistical correlation, such as Pearson Correlation Coefficient analysis and Spearman Correlation Coefficient analysis.

[0041] An embodiment may use a template matching function that looks for portions of an input image that match a small sized template image.

[0042] The template matching principle scans the input image with a template, comparing it with every part of the image from the top left corner to the bottom right corner.

[0043] The template matching function can include various template matching functions according to the manner in which the correlation value is expressed.

[0044] FIG. 2 is a flow diagram of a method for finding the center of an overlay key based on correlation according to one embodiment.

[0045] Referring to FIG. 2, the correlation-based overlay key centering method includes a rough search step (S10) and a fine search step (S50).

[0046] First, in the rough search step (S10), the overlay target image is adjusted to a preset size, and a first correlation value in the adjusted image is calculated to calculate the rough center coordinate.

[0047] The rough search step (S10) will be described as follows.

[0048] The overlay target image is adjusted to a preset size (S20). In one example, the target finder receives the overlay target image from the measurement equipment and reduces the size of the overlay target image by half. This minimizes the processing time of the template matching function. Here, the target finder may be an application that calculates the center coordinates of the overlay key based on the correlation, or a PC terminal equipped with a processor.

[0049] Next, a first template and a second template in the adjusted image and a first correlation search region are defined, and first correlation values ​​between the first template and the second template in the first correlation search region are calculated (S30).

[0050] 3 to 5 are diagrams illustrating the correlation search area of ​​FIG.

[0051] 3-5, in one example, the first correlation search area may be defined as the entire area of ​​the adjusted image.

[0052] The first template 1 and the second template 2 may be defined to have horizontal and vertical values ​​of preset sizes and to be symmetrical to each other with respect to a diagonal line.

[0053] In one example, the embodiment may compare an image of the first template 1 and an image of the second template 2 with images rotated 180 degrees, and indicate the degree of similarity in a correlation value. In another example, the embodiment may compare an image of the first template 1 that is flipped across the diagonal between the first template 1 and the second template 2 with an image of the second template 2, and indicate the degree of similarity in a correlation value.

[0054] In one example, the embodiment may indicate that the correlation value is higher when the similarity between the first template 1 and the second template 2 is higher.

[0055] FIG. 6 is a diagram illustrating a method for finding the center of an overlay key based on the correlation of FIG.

[0056] 2 and 6, the rough center coordinates are calculated based on the first correlation values ​​between the first template 1 and the second template 2 within the entire area of ​​the adjusted image (S40).

[0057] To explain the rough center coordinate calculation in detail, the first template 1 and the second template 2 corresponding to the largest correlation value among the first correlation values ​​are found, and the rough center coordinate (Cx, Cy) is calculated using the coordinates (X1, Y1) of the first template 1 and the second template 2.

[0058] In one example, the rough center coordinate (Cx) can be calculated to the horizontal (or width) of the X1+ template, and the rough center coordinate (Cy) can be calculated to the vertical (or height) of the Y1+ template.

[0059] Next, in a fine search step (S50), a second correlation value is calculated based on the rough center coordinates in the original image of the overlay target image to calculate fine center coordinates.

[0060] The fine search step is described as follows.

[0061] The fine search step defines a second correlation search region based on the rough center coordinates (S60).

[0062] FIG. 7 is a diagram showing the definition of a template for calculating the correlation in FIG.

[0063] 7, a second correlation search area is defined based on the rough center coordinates in the original image. In one example, the second correlation search area may be defined to have twice the width and height of the first template or the second template based on the rough center coordinates.

[0064] The first and second template regions are defined based on the rough center coordinates of the original image. In one example, the first and second templates may be defined as regions symmetrical to each other with respect to a diagonal line passing through the rough center coordinates.

[0065] Then, second correlation values ​​between the first template and the second template in the second correlation search area of ​​the original image are calculated (S70).

[0066] In one example, the image of the first template 1 and the image of the second template 2 can be compared with images rotated 180 degrees to calculate the second correlation value based on the similarity, or the image of the first template 1 and the image of the second template 2 can be compared with images flipped to calculate the second correlation value based on the similarity.

[0067] Then, the fine center coordinates are calculated based on the second correlation values ​​(S80).

[0068] The step of calculating the fine center coordinates includes searching for the first template and the second template corresponding to the largest correlation value among the second correlation values, and calculating the fine center coordinates using the coordinates of the searched first template or the second template.

[0069] FIG. 8 is a flow diagram of a correlation-based overlay key centering system and method according to one embodiment.

[0070] Referring to FIG. 8, the target finder calculates the rough center coordinates through a rough search step, and calculates the fine center coordinates through a fine search step.

[0071] The rough search step reduces the size of the entire image to minimize the processing time. In one example, the size of the entire image can be resized to 1 / 2 or 1 / 4. Of course, this is not limited to this, and any modification can be made within the technical concept that can minimize the processing time for calculating the center coordinates.

[0072] The first template 1 and the second template 2 are defined as shown in Fig. 5. The entire area of ​​the resized image is defined as the first correlation search area.

[0073] A correlation value between the first template and the second template in the first correlation search area is calculated, and the template with the largest correlation value is found and used to calculate the rough center coordinates.

[0074] A second correlation region is defined based on the rough center coordinates found by the rough search of the fine center coordinates, and a correlation value between the first template and the second template in the second correlation region is calculated in the original image.

[0075] The template with the highest correlation value is found and used to calculate the fine center coordinates.

[0076] In one example, the overlay metrology equipment may include an internal target finder and register the calculated fine center coordinates in a recipe or move the stage to the target center using the fine center coordinates of the registered recipe. In this case, the overlay metrology equipment may register the PRU image in the recipe using the target finder in a first case. In a second case, the overlay metrology equipment may move the stage to the target center in an edit recipe using the target finder.

[0077] Thus, according to one embodiment, the correlation-based overlay key centering system includes an overlay measurement device including at least one processor, which receives an overlay target image, adjusts the overlay target image to a size smaller than the entire image, calculates a first correlation value in the adjusted image to calculate a rough center coordinate, and calculates a second correlation value based on the rough center coordinate in the original image of the overlay target image to calculate a fine center coordinate.

[0078] In the overlay measurement equipment, when calculating the rough center coordinate, the processor adjusts the overlay target image to a preset size, defines a first template and a second template in the adjusted image and a first correlation search area, calculates first correlation values ​​between the first template and the second template in the first correlation search area, and calculates the rough center coordinate based on the first correlation values.

[0079] A first correlation search area is defined as the entire area of ​​the adjusted image.

[0080] The first and second templates have preset horizontal and vertical values ​​and are defined to be symmetrical to each other with respect to a diagonal line.

[0081] The processor searches for the first template and the second template corresponding to the largest correlation value among the first correlation values, and calculates the rough center coordinates using the coordinates of the searched first template or the second template.

[0082] When calculating the fine center coordinates, the processor defines a second correlation search area based on the rough center coordinates, calculates second correlation values ​​between the first template and the second template in the second correlation search area in the original image, and calculates the fine center coordinates based on the second correlation values.

[0083] The processor finds the first template and the second template corresponding to the largest correlation value among the second correlation values, and calculates the fine center coordinates using the coordinates of the first template or the second template.

[0084] The overlay measurement equipment registers an image of the overlay key centered based on the fine center coordinates into the recipe.

[0085] And, according to one embodiment, the method for centering an overlay key based on correlation includes a rough search step of adjusting an overlay target image to a preset size, calculating a correlation value in the adjusted image, and calculating a rough center coordinate, and a fine search step of calculating a correlation value based on the rough center coordinate in an original image of the overlay target image, and calculating a fine center coordinate.

[0086] The rough search step includes the steps of adjusting the overlay target image to a preset size, defining a first template, a second template and a first correlation search area in the adjusted image, calculating first correlation values ​​between the first template and the second template in the first correlation search area, and calculating rough center coordinates based on the first correlation values.

[0087] A first correlation search area is defined as the entire area of ​​the adjusted image.

[0088] The first and second templates have preset horizontal and vertical values ​​and are defined to be symmetrical to each other with respect to a diagonal line.

[0089] The step of calculating the rough center coordinates includes the steps of searching for a first template and a second template corresponding to the largest correlation value among the first correlation values, and calculating the rough center coordinates using the coordinates of the first template or the second template.

[0090] The fine search step includes the steps of: defining a second correlation search area based on a rough center coordinate; calculating second correlation values ​​between the first template and the second template in the second correlation search area in the original image; and calculating the fine center coordinates based on the second correlation values.

[0091] The step of calculating the fine center coordinates includes the steps of searching for a first template and a second template corresponding to the largest correlation value among the second correlation values, and calculating the fine center coordinates using the coordinates of the first template or the second template.

[0092] The correlation based method for centering the overlay key further includes registering an image of the overlay key centered based on the fine center coordinates to a recipe.

[0093] The method for centering the overlay key based on the correlation further includes moving the stage to a target position based on the fine center coordinate to center the overlay key.

[0094] According to the embodiment, by accurately centering the overlay key based on the correlation, the alignment state of fine patterns on a semiconductor substrate can be accurately measured and inspected.

[0095] Also, by accurately finding the center coordinates of the overlay key pattern, the precision of the semiconductor process can be improved.

[0096] In addition, it is possible to prevent PR (Pattern Recognition) failures that may occur when there is a high degree of variation for each overlay key.

[0097] In addition, it is possible to compensate for PR misreading that may occur in the case of a layer that is sensitive to PR parameters.

[0098] As described above, the present invention has been described with reference to the illustrative drawings, but the present invention is not limited to the embodiments and drawings disclosed in this specification, and it is clear that various modifications can be made by those skilled in the art within the scope of the technical concept of the present invention. Even if the effects of the configuration of the present invention are not explicitly described and explained while the embodiments of the present invention are described above, it is natural that the effects that can be predicted by the configuration should also be recognized.

Claims

1. an overlay measurement device that includes at least one processor and inspects an alignment state of a lower thin film layer pattern and an upper thin film layer pattern using an overlay target image; The processor, receiving the overlay target image; Adjusting the overlay target image to a size smaller than the entire image; defining a first template and a second template that are symmetrical to each other based on a first diagonal in the adjusted image; calculating a first correlation value representing a degree of similarity between the first and second template images symmetrically with respect to the first diagonal line, and calculating rough center coordinates; defining symmetrical third and fourth templates based on a second diagonal line passing through the rough center coordinates in the original image of the overlay target image; calculating a second correlation value representing a similarity between the third template image and the fourth template image symmetrically with respect to the second diagonal line to calculate fine center coordinates of the overlay target image; centering the overlay key by moving the stage to a target position based on the fine center coordinates; Correlation-based centering system for overlay keys.

2. When calculating the rough center coordinate, the processor further defining a first correlation search area and calculating the first correlation value between the image of the first template and the image of the second template in the first correlation search area.

2. The correlation-based overlay key centering system of claim 1.

3. the first correlation search area is defined as the entire area of ​​the adjusted image; 3. The correlation-based overlay key centering system of claim 2.

4. The first template and the second template have width and height values ​​of a preset size.

4. The correlation-based overlay key centering system of claim 3.

5. The processor, searching for the first template position and the second template position corresponding to the largest correlation value, and calculating the rough center coordinates using the coordinates of the found first template position and the found second template position; 3. The correlation-based overlay key centering system of claim 2.

6. When calculating the fine center coordinates, the processor and further defining a second correlation search area based on the rough center coordinates of the original image, and calculating the second correlation value between the image of the third template and the image of the fourth template within the second correlation search area.

2. The correlation-based overlay key centering system of claim 1.

7. The processor, searching for a third template position and a fourth template position corresponding to the largest correlation value, and calculating the fine center coordinates using the coordinates of the searched third template position and the fourth template position; 7. The correlation-based overlay key centering system of claim 6.

8. the processor registers the image of the overlay key centered on the fine center coordinates to a recipe; 2. The correlation-based overlay key centering system of claim 1.

9. An overlay target image for inspecting an alignment state of the lower thin film layer pattern and the upper thin film layer pattern is adjusted to a size smaller than the entire image; defining a first template and a second template that are symmetrical to each other based on a first diagonal in the adjusted image; calculating a first correlation value representing a degree of similarity between the first template image and the second template image symmetrically with respect to the first diagonal line to calculate rough center coordinates; defining symmetrical third and fourth templates based on a second diagonal line passing through the rough center coordinates in the original image of the overlay target image; calculating a second correlation value representing a similarity between the third template image and the fourth template image symmetrically with respect to the second diagonal line to calculate fine center coordinates of the overlay target image; centering the overlay key by moving the stage to a target position based on the fine center coordinates; A correlation-based method for centering overlay keys.

10. Calculating the rough center coordinates further comprises: Defining a first correlation search region in the adjusted image; calculating the first correlation value between the first template and the second template in the first correlation search area. The correlation-based overlay key centering method according to claim 9.

11. the first correlation search area is defined as the entire area of ​​the adjusted image; The method for centering an overlay key based on correlation according to claim 10.

12. The first template and the second template have width and height values ​​of a preset size. The method for centering an overlay key based on correlation according to claim 11.

13. Calculating the rough center coordinates comprises: finding the first and second template positions corresponding to the largest correlation value; and calculating the rough center coordinate using the coordinates of the found first template position and the found second template position. The method for centering an overlay key based on correlation according to claim 10.

14. Calculating the fine center coordinates further comprises: defining a second correlation search area based on the rough center coordinates of the original image; calculating the second correlation value between the image of the third template and the image of the fourth template within the second correlation search area. The correlation-based overlay key centering method according to claim 9.

15. Calculating the fine center coordinates comprises: finding the third and fourth template locations corresponding to the largest correlation value; and calculating the fine center coordinate using the coordinates of the located third template position and the fourth template position. The correlation-based overlay key centering method of claim 14.

16. and registering the image of the overlay key centered on the fine center coordinates to a recipe. The correlation-based overlay key centering method according to claim 9.

Images