A method for correcting exposure pattern distortion caused by proximity effect

Abstract

The application belongs to the field of photolithography technology for semiconductor manufacturing, and provides a correction method for exposure pattern distortion caused by proximity effect by using conditional judgment formula rule. The application discloses a method for correcting exposure pattern distortion caused by proximity effect, which comprises the following steps: dividing an exposure pattern on a photomask into pixel units to obtain a target pattern; calculating proximity effect of the target pattern to obtain a proximity effect pattern; comparing the proximity effect pattern obtained each time with the target pattern to obtain redundant points and missing points; correcting the redundant points and the missing points to obtain a corrected pattern; when the proximity effect pattern of the corrected pattern is consistent with the target pattern, the correction is completed, otherwise, the correction is repeated.

Description

[0001] This invention relates to the field of photolithography technology in semiconductor manufacturing, and more specifically, to a method for correcting exposure pattern distortion caused by proximity effect. Background Technology

[0002] This invention relates to the field of photolithography technology in semiconductor manufacturing. Proximity effect correction is a photolithography enhancement technique primarily used in semiconductor chip production to ensure the complete transfer of the target pattern from the mask onto the chip during the manufacturing process. To create smaller target patterns, methods can be used to correct exposure distortion caused by proximity effects through pre-process computer simulation calculations, compensating for exposure imaging results by improving the mask design beforehand. However, conventional computer simulation methods for correcting proximity effects require extensive simulation and analysis of graphic data to obtain optimal results, often resulting in a massive computational burden. This invention proposes a method using a discriminant method and specific correction rules to correct proximity effects, effectively reducing the amount of computation required for simulations. Summary of the Invention

[0003] This invention provides a method for correcting exposure image distortion caused by proximity effect. It mainly uses discriminant method plus specific correction rules to reduce the amount of data calculation, while still achieving the same accurate proximity effect correction result.

[0004] The technical solution adopted in this invention is as follows:

[0005] Step 1: Convert the exposure pattern designed on the photomask into a pixel unit pattern. This pattern is defined as the target pattern. The exposed pixel units are defined as the original exposure points, and the unexposed pixel units are defined as the original non-exposure points.

[0006] Step 2: Convert the effect of the proximity effect into a pixel image corresponding to the light source intensity. This is done by simplifying the intensity data values ​​from 1 to 0 in a symmetrical manner based on the center of the light source point, and estimating the actual exposure image that the target image will produce under this light source condition. This image is defined as the proximity effect image.

[0007] Step 3: Compare the pixels of the proximity effect image with those of the target image to obtain the excess and lack of exposure points that differ from the target image. These are referred to as excess points and lack points.

[0008] Step 4: Based on the total number of redundant and missing points, if the total number of missing points is large, proceed with the process of adding missing points; if the total number of redundant points is large, proceed with the process of deleting redundant points.

[0009] Step 5: Then, based on whether the missing or extra points are the exposure points to be corrected or the non-exposure points to be corrected, determine the correction addition rules and correction deletion rules, and obtain the corresponding correction graphics;

[0010] Step 6: Estimate the corresponding proximity effect image again from the corrected image, and compare this proximity effect image with the target image. If the difference exceeds the error range, repeat steps 3-6. If the difference is within the ideal range, use the final corrected image for exposure to obtain the target image result that avoids distortion caused by the proximity effect.

[0011] Furthermore, the intensity data values ​​of the light source point, which decrease from 1 to 0 in a centrally symmetrical manner, are simplified into a nearly centrally symmetrical decreasing numerical distribution based on the actual intensity of the light source to represent the light source distribution values, where 1 represents the case of the strongest light intensity and 0 represents the case of no light illumination.

[0012] Furthermore, the proximity effect pattern is obtained by converting the exposed pixel units and the unexposed pixel units into the values ​​1 and 0 respectively, multiplying the value of each pixel point by the value of the light source distribution and the weight value caused by other exposure conditions, and then superimposing the values ​​obtained at each point. According to the definition, the value exceeding a certain threshold value is the exposure point of the proximity effect pattern.

[0013] Furthermore, the process of adding missing points is based on the rule that when the missing point is a point to correct exposure, eight points around the missing point are added as the correction rule; or when the missing point is a point to correct non-exposure, the location of the missing point is added as the correction rule.

[0014] Furthermore, the process of deleting redundant points is based on the rule of deleting the position of the redundant point when the redundant point is the exposure correction point.

[0015] Furthermore, the process of deleting redundant points is based on the fact that when the redundant point is a point to be corrected for exposure, it is necessary to further determine whether there are any newly added points around the redundant point as the basis for this correction and deletion rule.

[0016] Furthermore, when there are newly added points around the redundant point to serve as the correction deletion rule, deleting the newly added points around the redundant point has a greater impact on the redundant point and a smaller impact on the missing point.

[0017] Furthermore, when there are no newly added points around the redundant point to serve as the correction deletion rule, deleting some of the original exposure points around the redundant point has a significant impact on the redundant point but a minor impact on other original exposure points.

[0018] Furthermore, both the modified addition rule and the modified deletion rule can multiply each discrimination point by the weight value of other factors affecting exposure conditions to more accurately define the missing points to be added or the redundant points to be deleted.

[0019] In summary, the exposure pattern on the photomask is divided into pixel units to obtain the target pattern; the proximity effect is calculated on the target pattern to obtain the proximity effect pattern; the proximity effect pattern calculated each time is compared with the target pattern to obtain redundant and missing points; the redundant and missing points are corrected to obtain the corrected pattern; when the proximity effect pattern of the corrected pattern is consistent with the target pattern, the correction is completed; otherwise, the correction will be repeated. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the overall steps of the present invention;

[0021] Figure 2 This is a diagram representing the types of points in an embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram illustrating the conversion of the exposure pattern on the photomask into a target pattern of pixel units in an embodiment of the present invention;

[0023] Figure 4 The target graphic in Embodiment 1 of this invention;

[0024] Figure 5 The proximity effect graphic of the first target graphic in Embodiment 1 of the present invention;

[0025] Figure 6 The first N-1 total corrections in Embodiment 1 of this invention;

[0026] Figure 7 The (N-1)th corrected pattern in Embodiment 1 of the present invention;

[0027] Figure 8 The proximity effect graph of the (N-1)th correction graph in Embodiment 1 of the present invention;

[0028] Figure 9 The Nth modification in Embodiment 1 of this invention;

[0029] Figure 10 The Nth corrected pattern in Embodiment 1 of the present invention;

[0030] Figure 11 The proximity effect graph of the Nth correction graph in Embodiment 1 of the present invention;

[0031] Figure 12 The (N+1)th correction in Embodiment 1 of this invention;

[0032] Figure 13 The (N+1)th corrected pattern in Embodiment 1 of the present invention;

[0033] Figure 14 The proximity effect graph of the (N+1)th correction graph in Embodiment 1 of the present invention;

[0034] Figure 15 The (N+2)th correction in Embodiment 1 of this invention;

[0035] Figure 16 The (N+2)th correction pattern in Embodiment 1 of the present invention;

[0036] Figure 17 The proximity effect graph of the (N+2)th correction graph in Embodiment 1 of the present invention; Detailed Implementation

[0037] The technical solution of the present invention will be clearly and completely described below with reference to specific embodiments. Obviously, the described embodiments are merely one embodiment of the present invention, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0038] Example 1

[0039] This embodiment takes a method for correcting exposure pattern distortion caused by proximity effect on an arrow-shaped photomask as an example, including the following steps:

[0040] Step 1: Convert the exposure pattern designed on the photomask into a pixel unit pattern. This pattern is defined as the target pattern. The exposed pixel units are defined as the original exposure points, and the unexposed pixel units are defined as the original non-exposure points.

[0041] Step 2: Convert the effect of the proximity effect into a pixel image corresponding to the light source intensity. That is, simplify the intensity data value from 1 to 0 in a symmetrical manner with the center of the light source point, and estimate the exposure image that the target image will actually produce under this light source condition. This image is defined as the first proximity effect image.

[0042] Step 3: Compare the pixels of the first proximity effect image with those of the target image to obtain the redundant and missing points that differ from the target image;

[0043] Step 4: Compare with the target image. The total number of points is determined to be 5 missing points and 2 redundant points. Since the total number of missing points is large, the process of adding missing points will proceed.

[0044] Step 5: Since the missing point is the exposure correction point, a correction and addition rule is used, that is, a ring of points is added around the missing point to obtain the corresponding correction graphic;

[0045] Step 6: After N-1 corrections, a total of 19 exposure points are corrected, resulting in the N-1 corrected image. Comparing this image with the target image, 19 points have been added. Comparing the proximity effect image of the N-1 corrected image with the target image, 13 extra points and 0 missing points are identified. The Nth correction is then performed, correcting 11 exposure points (by directly deleting these points) and correcting 2 non-exposure points (deleting surrounding exposure points since no new points are added around the extra points). Comparing the Nth corrected image with the target image, a total of 8 points have been added and 19 points have been deleted. The number of points is 2. When comparing the proximity effect image of the Nth correction image with the target image, it is determined that there are 2 extra points. The N+1th correction is performed, which corrects 2 non-exposed points (no new points are added around the extra points, and some exposed points around them are deleted). When comparing the N+1th correction image with the target image, a total of 8 points are added and a total of 4 points are deleted. When comparing the proximity effect image of the N+1th correction image with the target image, it is determined that there are 2 missing points. The N+2th correction is performed, which corrects 2 non-exposed points (these points are added directly).

[0046] Step 7: Compare the proximity effect graph of the (N+1)th correction graph with the target graph. If there are no redundant or missing points, and the difference is within the ideal range, the correction ends. If the difference is not within the ideal range, repeat steps 3-6 until the difference is within the ideal range.

Claims

1. A method for correcting exposure pattern distortion caused by proximity effect, characterized in that, Includes the following steps: Step 1: Convert the exposure pattern designed on the photomask into a pixel unit pattern. This pattern is defined as the target pattern, and the exposed pixel units are defined as the original exposure points, and the unexposed pixel units are defined as the original non-exposure points. Step 2: Convert the effect of the proximity effect into a pixel image corresponding to the light source intensity. That is, simplify the actual light source intensity into intensity data values ​​decreasing from 1 to 0, and estimate the exposure image that the target image will actually produce under this light source condition. This image is defined as the proximity effect image. Step 3: Compare the pixels of the proximity effect image with those of the target image to obtain the excess and lack of exposure points that differ from the target image. These are referred to as excess points and lack points. Step 4: Based on the total number of redundant and missing points, if the total number of missing points is large, proceed with the process of adding missing points; if the total number of redundant points is large, proceed with the process of deleting redundant points. Step 5: Based on the missing point addition or redundant point deletion process determined in Step 4, select the correction addition rule or correction deletion rule, and obtain the corresponding correction graphic; specifically, when adding a missing point and the missing point is a correction exposure point, add eight points around the missing point as the correction addition rule; when adding a missing point and the missing point is a correction non-exposure point, add the missing point's location as the correction addition rule; when deleting redundant points and the redundant point is a correction exposure point, delete the redundant point's location as the correction deletion rule; when deleting redundant points and the redundant point is a correction non-exposure point, determine whether there are any newly added points around the redundant point. If there are newly added points around the redundant point, delete the points among the newly added points that have a greater impact on the redundant point and a smaller impact on the missing point; if there are no newly added points around the redundant point, delete the points among the original exposure points that have a greater impact on the redundant point and a smaller impact on other original exposure points, as the correction deletion rule. Step 6: Estimate the corresponding proximity effect image again from the corrected image, and compare this proximity effect image with the target image. If the difference exceeds the error range, repeat steps 3-6. If the difference is within the ideal range, use the final corrected image for exposure to obtain the target image result that avoids distortion caused by the proximity effect.

2. The method for correcting exposure pattern distortion caused by proximity effect according to claim 1, characterized in that: The intensity data values ​​of the light source, which are symmetrically decreasing from 1 to 0, are simplified based on the actual intensity of the light source into a nearly symmetrical decreasing numerical distribution to represent the distribution of the light source. Here, 1 represents the case of the strongest light intensity, and 0 represents the case of no light.

Images