OPC correction method for improving local protrusion phenomenon

Abstract

The invention provides an OPC correction method for improving a local protrusion phenomenon, and the method comprises the steps: determining an initial pattern, and forming a plurality of target patterns of the same type and different sizes; predicting the SRAF relative position of a part of the target pattern by adopting an inverse photoetching technology; obtaining an SRAF rule of a part of target graphs; obtaining an SRAF rule table; performing pattern matching on the target pattern to find a to-be-corrected pattern; and when a program is debugged, the SRAF rule table is imported, and OPC correction is carried out. The method comprises the following steps: predicting the relative position of a part of target graph SRAF by adopting an inverse lithography technology, measuring an SRAF rule to obtain an SRAF rule table, carrying out graph matching on the target graphs to find to-be-corrected graphs of the same type and different sizes, importing the SRAF rule table and carrying out OPC correction when a program is debugged, thereby reducing the EPE of a simulation contour and an ADI target value, and improving the simulation accuracy of the ADI target value. And the local bulge phenomenon is improved.

Description

Technical Field

[0001] The invention relates to the field of semiconductor technology, and in particular to an OPC correction method for improving local protrusion phenomenon. Background Art

[0002] The OPC correction (Optical Proximity Correction) process inevitably has hotspots in the photolithography process. Most of the local ridge phenomena (Ripple phenomenon) may cause pattern disconnection (Pinch) or bridging (Bridge) phenomena. The industry generally matches the simulation contour (Contour) with the ADI target value (target value after photolithography development) by increasing the number of OPC iterations, ADI target value or changing the mask morphology.

[0003] In the patent with application number 202211164126.3 and titled “Optical Proximity Effect Correction Method for Improving Imaging Contrast”, the image is improved by modifying the mask morphology through multiple iterations of OPC. However, if the mask is repaired to the limit, it will cause MRC (Mask Rule Check) to get stuck. In the patent with application number 202210006263.8 and titled “Optical Proximity Correction Method”, the length of the mask segment is changed to modify the mask morphology and ultimately improve the image. However, if the mask segment is too small, the mask shop will have an increased difficulty in production.

[0004] In the patent with application number 202010213032.5 and titled “A method for screening mask auxiliary graphics”, different SRAFs (Sub-Resolution Assist Features, exposure auxiliary graphics) are added to the same type of main graphics, and the SRAF rules (SRAF Rules) are obtained according to experience after collecting data on the wafer, and finally the graphics are improved. However, in this patent, a special test mask is required (different main graphics are matched with different SRAF Rules), and the data after exposure is collected on the wafer multiple times to obtain the SRAF Rules, which is time-consuming. In the patent with application number 2022113871645 and titled “Method and storage medium for adding scattering strips”, the inverse lithography idea is used to calculate the appropriate SRAF Rules for all main graphics of the same type and different sizes. However, in this patent, different main graphics of the same type need to be adapted to different SRAF Rules. This solution cannot adapt to main graphics of different sizes. In addition, the solution does not mention the specific implementation process. Summary of the invention

[0005] The object of the present invention is to provide an OPC correction method for improving the local convexity phenomenon, reducing edge placement, improving the local convexity phenomenon, and adapting to main graphics of different sizes.

[0006] In order to solve the above technical problems, the present invention provides an OPC correction method for improving the local convexity phenomenon, comprising the following steps:

[0007] Determine an initial graphic, and scale a portion of the size of the ADI layer of the initial graphic to form a plurality of target graphics of the same type but different sizes;

[0008] Predicting the relative position of the SRAF of a portion of the target pattern using an inverse lithography technique;

[0009] Acquire a SRAF rule that matches an ADI layer of at least a portion of the target graphic;

[0010] Get the SRAF rule table;

[0011] Performing pattern matching on the ADI layer of the target pattern in the ADI layer of the GDS layout to find a pattern to be corrected of the same type but different size; and

[0012] When debugging the program, the SRAF rule table is imported and OPC correction is performed.

[0013] Optionally, the method of scaling the partial size of the ADI layer of the initial graphic to form multiple target graphics of the same type but different sizes includes: sequentially enlarging the partial size of the ADI layer by multiple intervals and sequentially reducing the partial size by multiple intervals to form multiple target graphics.

[0014] Optionally, the enlarged multiple intervals are equal to the reduced multiple intervals.

[0015] Optionally, the method of scaling a portion of the size of the ADI layer of the initial graphic to form multiple target graphics of the same type but different sizes includes: enlarging and reducing a portion of the size of the ADI layer to obtain a fluctuation range, and selecting multiple sizes from the fluctuation range to form multiple target graphics.

[0016] Optionally, the initial graphic includes a first part, a second part and a third part extending along the first direction and arranged in sequence along the second direction, a contact hole area is provided in the middle position between the first part and the third part, and the contact hole area protrudes to both sides in the second direction; the length of the contact hole area in the first direction is L, the protruding width in the second direction is H, and the spacing between adjacent first parts and second parts or between adjacent second parts and third parts is P; wherein the first direction is perpendicular to the second direction.

[0017] Optionally, the length L, width H and spacing P in the ADI layer of the initial graphic are scaled to form a plurality of target graphics of the same type but different sizes.

[0018] Optionally, the SRAF rule includes: a distance O from an edge of the SRAF in the second direction to a side of the contact hole region in the second direction, a width W of the SRAF in the second direction, and a distance S between adjacent SRAFs.

[0019] Optionally, the method for obtaining the SRAF rule table includes: using an interpolation method to obtain an SRAF rule that matches the ADI layer of another part of the target graphic, thereby obtaining the SRAF rule table.

[0020] Optionally, the method of performing graphic matching on the ADI layer of the target graphic in the ADI layer of the GDS layout to find a graphic to be corrected of the same type but different size includes: setting a threshold range for a partial size of the ADI layer of the target graphic, and searching for a graphic within the threshold range in the ADI layer of the GDS layout as the graphic to be corrected.

[0021] Optionally, the method of importing the SRAF rule table and performing OPC correction when debugging a program includes: importing the SRAF rule table when debugging a program, finding the SRAF rule corresponding to each of the graphics to be corrected, adding corresponding SRAFs to the graphics to be corrected of the same type but different sizes according to the SRAF rule, and performing OPC correction.

[0022] In summary, in the OPC correction method for improving the local convex phenomenon provided by the present invention, the initial graphic is first determined, and the partial size of the ADI layer of the initial graphic is scaled to form a plurality of target graphics of the same type and different sizes, and then the relative position of the SRAF of a part of the target graphic is predicted by the inverse lithography technology, and then the SRAF rule matching the ADI layer of at least a part of the target graphic is obtained, and then the SRAF rule table is obtained, and then the ADI layer of the target graphic is pattern matched in the GDS layout to find the graphics to be corrected of the same type and different sizes, and then the SRAF rule table is imported when debugging the program and OPC correction is performed. The present invention selects a part of the target graphics of the same type and different sizes, predicts the relative position of the SRAF by the inverse lithography technology, and measures the SRAF rule, obtains the SRAF rule table, and then the ADI layer of the target graphic is pattern matched to find the graphics to be corrected of the same type and different sizes, and then the SRAF rule table is imported when debugging the program and OPC correction is performed, thereby reducing the EPE of the simulated contour and the ADI target value, improving the local convex phenomenon; and avoiding a long iterative process, reducing the time consumption of OPC.

[0023] In addition, the interpolation method is used to obtain the SRAF rule that matches the ADI layer of another part of the target graphic, which can avoid measuring all the dimensional data, improve efficiency and reduce time consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Those skilled in the art will appreciate that the accompanying drawings are provided for a better understanding of the present invention and do not constitute any limitation on the scope of the present invention.

[0025] Figure 1 It is a schematic diagram of a design layout provided by an embodiment of the present invention.

[0026] Figure 2 It is a schematic diagram of the relationship between the design layout and the ADI layer provided by an embodiment of the present invention.

[0027] Figure 3 It is a schematic diagram of the relationship between the design layout, ADI layer, mask layer pattern and simulation outline provided by an embodiment of the present invention.

[0028] Figure 4 This is a profile electron scanning microscope image of a local protrusion phenomenon in a photoresist provided by an embodiment of the present invention.

[0029] Figure 5 It is a flow chart of an OPC correction method for improving the local bulge phenomenon provided by an embodiment of the present invention.

[0030] Figures 6a to 6c It is a schematic diagram of the relative position relationship between ADI layers of the same type but different sizes and SRAF layers provided by an embodiment of the present invention.

[0031] Figure 7 This is a profile electron scanning microscope image after correction and exposure using the OPC correction method provided by an embodiment of the present invention.

[0032] Figure 8 It is a schematic diagram comparing the correction method of the present invention with the correction method of the prior art.

[0033] Fig. 9 yes Figure 8 Light intensity distribution diagram at the location of the middle dotted line. DETAILED DESCRIPTION

[0034] In order to make the purpose, advantages and features of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In addition, the structure shown in the drawings is often a part of the actual structure. In particular, the emphasis of each drawing is different, and sometimes different scales are used.

[0035] As used in the present invention, the singular forms "one", "an", and "the" include plural objects, unless the content clearly indicates otherwise. As used in the present invention, the term "or" is generally used in a sense that includes "and / or", unless the content clearly indicates otherwise. As used in the present invention, the term "several" is generally used in a sense that includes "at least one", unless the content clearly indicates otherwise. As used in the present invention, the term "at least two" is generally used in a sense that includes "two or more", unless the content clearly indicates otherwise. In addition, the terms "first", "second", and "third" are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first", "second", and "third" may explicitly or implicitly include one or at least two of the features.

[0036] The embodiment of the present invention takes a combination of “symmetrical sandwich-type patterns (Poly Landing)” of a polysilicon layer (Poly layer) as an example to illustrate an OPC correction method for improving the local protrusion phenomenon. Figure 1 This is a schematic diagram of the structure of the design layout provided by an embodiment of the present invention, please refer to Figure 1 As shown, the design layout 10 includes a first portion 11, a second portion 12 and a third portion 13 extending along a first direction x and arranged in sequence along a second direction y, wherein a contact hole region is provided in the middle of the first portion 11 and the third portion 13, and the contact hole region protrudes to both sides in the second direction y (i.e. Figure 1 The size and structure of the protrusions on the left and right sides can be symmetrical or asymmetrical, but the main morphology is similar. The contact hole area needs to wrap the contact hole (contact) and have a sufficient distance from the contact hole, so the area must protrude to the left and right. If the spacing between adjacent parts is too close, due to the photolithography diffraction and interference effects, the photoresist "attracts" phenomenon is inevitable, causing the simulated profile of the OPC simulation or the actual profile after actual exposure (Profile) to be not close to the ADI target value, which is called the local protrusion phenomenon. Figure 4 shown.

[0037] The chip design part designs the initial semiconductor design layout according to the electrical requirements, such as Figure 1 As shown, the parts with graphics in the design layout are all photoresists. After the photoresist undergoes a photolithography process and then an etching process, its line width will be reduced accordingly. Therefore, the final size after etching is obtained by adding the etching deviation and the photolithography deviation in turn to obtain the size after photolithography.

[0038] Among them, ADI layer: the layer from the design (Design) after the Global Bias (the overlay deviation caused by global lithography and etching) is called the ADI layer, which is a layer concept.

[0039] Mask layer: After OPC makes multiple OPC corrections based on the ADI layer, this patent is based on the ADI layer and adds SRAF and makes multiple OPC corrections to obtain the final mask layer, which is a layer concept.

[0040] Simulated contour layer (or simply called simulated contour): The contour of the exposed photoresist on the wafer obtained by OPC simulation based on the mask layer obtained after OPC correction. It is a layer concept.

[0041] ADI target value: The specific size of the graphics in the ADI layer, which is a series of specific size values.

[0042] OPC correction relies on the OPC model, that is, a physical model needs to be established before OPC correction to approximate the photolithography exposure process through a computer. A sufficient amount of regular test patterns are collected on a wafer generated by test mask exposure, and an OPC model is established through EDA software based on these wafer data. The simulated profile is obtained by simulating exposure through the OPC model, and the best mask layer pattern is calculated after OPC correction in combination with the simulated profile. After the mask layer pattern is exposed by the lithography machine, the ideal actual profile is obtained.

[0043] The OPC correction process is actually the process of changing the original shape of the mask so that the simulated contour is aimed at the ADI target value (the target value after photolithography development). Therefore, after obtaining the semiconductor design layout, firstly perform GlobalBias (the superposition deviation generated by global photolithography and etching) on ​​all design layouts, that is, expand a value as a whole on the basis of the original graphics to form the ADI target value, such as Figure 2 As shown, graph 20 is the ADI target value to supplement the line width reduction caused by the photolithography process and the etching process.

[0044] Figure 3 FIG. 1 is a schematic diagram showing the relationship between the design layout, the ADI layer, the pattern of the mask layer and the simulation profile provided by an embodiment of the present invention. Please refer to FIG. Figure 3 As shown, considering the optical diffraction and interference effects of the lithography process, the graph 20 formed after Global Bias is the ADI target value, the pattern of the mask layer obtained by OPC correction is the graph 30, and the simulated contour is the graph 40 in the figure.

[0045] It is known that the simulated contour can approximately reflect the actual contour of the wafer after exposure. In order to make the simulated contour close to the target, the local protrusion effect is generally improved by locally changing the ADI target value, increasing the number of OPC iterations, adding SRAF (Sub-Resolution Assist Feature, exposure auxiliary graphics), etc.

[0046] Based on the current ADI layer, the mask layer pattern obtained after OPC correction is simulated, and the EPE (Edge Placement Error, the edge placement error between the simulated contour and the ADI target value) is measured by the software process ruler to quantitatively express the severity of the local bulge phenomenon. In this embodiment, if EPE>2nm, it is considered that there is a local bulge. Please refer to Figure 3 As shown, for the design layout 10, the width of the second part is 36nm, the spacing between the second part and the third part is 90nm, and the minimum distance from the contact hole in the third part to the edge of the third part (the edge close to the second part) is 9nm. After Global Bias is performed on the design layout 10, the figure 20 is formed, the width of the second part is 56nm, the spacing between the second part and the third part is 70nm, and the distance from the contact hole in the third part to the edge of the third part (the edge close to the second part) is 16nm. The EPE of the final simulated contour in the contact hole area is 8.5nm. At this time, the number of OPC iterations has reached the upper limit of the required number, that is, the pattern shape of the mask layer can no longer be changed.

[0047] In order to shrink the pattern 40 in the contact hole area to reduce EPE, the pattern 30 also needs to be reduced, which will cause the second part of the pattern 30 to be reduced as well. However, since the mask grade is Q, the light-transmitting area spacing of the mask layer pattern is required to be 40nm, that is, the light-transmitting area spacing of the mask layer pattern needs to be greater than or equal to 40nm, that is, no matter how the number of OPC iterations is increased, the light-transmitting area spacing of the mask layer pattern F will never be reduced below 40nm.

[0048] If the ADI target value is modified directly, that is, the distance G from the contact hole to the edge is modified directly, the test shows that when G is reduced from 16nm to 7nm, the EPE of the local protrusion position is reduced from 8.5nm to 1.5nm, and the local protrusion phenomenon is improved. Because the distance between the patterns is larger, the photoresist does not have local protrusion phenomenon, that is, the visible "attraction" phenomenon, but it will lead to insufficient wrapping of the contact hole, because the minimum value of the ADI target value of G in the design rules should be 9nm.

[0049] In order not to change the ADI target value, the solution of adding SRAF is considered, that is, hollowing out a part of the middle of the graphic, which can improve the local optical scattering effect, thereby improving the local bulge phenomenon and avoiding the Bridge risk.

[0050] After further research, an embodiment of the present application provides an OPC correction method for improving the local bulge phenomenon, including the following steps: determining an initial graphic, and scaling part of the size of the ADI layer of the initial graphic to form multiple target graphics of the same type but different sizes; using inverse lithography technology to predict the relative position of the SRAF of a part of the target graphic; measuring the SRAF rule that matches the target value of the ADI layer of at least a part of the target graphic; obtaining an SRAF rule table; performing graphic matching on the ADI layer of the target graphic in the ADI layer of the GDS layout to find graphics to be corrected of the same type but different sizes; and importing the SRAF rule table and performing OPC correction when debugging the program.

[0051] Figure 5 This is a flow chart of an OPC correction method for improving local bulge phenomenon provided by an embodiment of the present invention. Figure 5 As shown, the OPC correction method for improving the local protrusion phenomenon provided by the embodiment of the present invention includes the following steps:

[0052] S1: determining an initial graphic, and scaling a portion of the size of the ADI layer of the initial graphic to form a plurality of target graphics of the same type but different sizes;

[0053] S2: predicting the relative position of the SRAF of a portion of the target pattern using an inverse lithography technique;

[0054] S3: Acquire a SRAF rule that matches the ADI layer target value of at least a portion of the target graphic;

[0055] S4: Get the SRAF rule table;

[0056] S5: performing pattern matching on the ADI layer of the target pattern in the ADI layer of the GDS layout to find a pattern to be corrected of the same type but different size; and

[0057] S6: When debugging the program, the SRAF rule table is imported and OPC correction is performed.

[0058] In step S1, an initial graphic is determined, and a portion of the size of the ADI layer of the initial graphic is scaled to form a plurality of target graphics of the same type but different sizes.

[0059] In this embodiment, the graph that produces the local convex phenomenon after the OPC model simulation is intercepted as the initial graph. Figures 1 to 3 The figure shown is used as an example for explanation.

[0060] Please refer to Figures 1 to 3 As shown, the design layout 10 of the initial graphic includes a first part 11, a second part 12 and a third part 13 extending along the first direction x and arranged in sequence along the second direction y, and a contact hole area is provided in the middle position between the first part 11 and the third part 13, and the contact hole area protrudes to both sides in the second direction y, that is, the width of the contact hole area in the second direction y is greater than the width of the remaining areas. In the ADI layer of the initial graphic, that is, in the curve 20, the length of the contact hole area in the first direction x is L, the protruding width in the second direction y is H, and the spacing between adjacent parts (that is, the adjacent first part 11 and the second part 12, or the adjacent second part 12 and the third part 13) is P, wherein the first direction x is perpendicular to the second direction y.

[0061] In this embodiment, the length L, width H and spacing P in the ADI layer of the initial graphic are scaled to form a plurality of target graphics of the same type but different sizes, but the present invention is not limited thereto.

[0062] Scaling the partial size of the ADI layer of the initial pattern may include: sequentially enlarging the partial size of the ADI layer by multiple intervals and sequentially reducing the multiple intervals, wherein the enlarged multiple intervals and the reduced multiple intervals may be equal or unequal. For example, if the length L of the ADI layer in the contact hole region is 200nm, the length L may be sequentially enlarged to 220nm and 240nm, and the length L may be sequentially reduced to 180nm and 160nm, thereby obtaining 5 lengths L in total; by analogy, 5 widths H and 5 intervals P may be obtained, thereby obtaining 125 target patterns.

[0063] Scaling the partial size of the ADI layer of the initial pattern may include: enlarging and reducing the partial size of the ADI layer to obtain a fluctuation range, and selecting multiple sizes from the fluctuation range to form multiple target patterns. For example, the length L of the ADI layer in the contact hole area is 200nm, and the range of the length L is determined to be 150nm to 250nm. Similarly, the range of the width H is determined to be 60nm to 100nm, and the range of the spacing P is determined to be 126nm to 800nm, and then an appropriate size is selected from the range of the length L, the width H and the spacing P to determine the target pattern.

[0064] In step S2, the relative position of the SRAF of a portion of the target pattern is predicted by using the inverse lithography technology.

[0065] Predicting SRAF through inverse lithography has the following advantages: no special test mask is required, it is adaptable to graphics of the same type but different sizes, and it can reversely match graphics of the same type but different sizes that may exist in the actual layout based on the ADI layer.

[0066] In this embodiment, the relative position of the SRAF of a part of the target graphics among the multiple target graphics is predicted by inverse lithography technology. Different target graphics correspond to different relative positions of SRAF. Exemplarily, the target value of the ADI layer of the target graphics is determined, and the simulated profile obtained by simulating exposure with the OPC model needs to be close to the target value of the ADI layer. The simulated profile of the contact hole area is tested for several iterations to see whether the EPE of the ADI target value converges (less than 2nm). If convergence stops iteration, the relative position of the best SRAF in the target graphics is output, and the relative position of the current SRAF in the graphics is defined as the best SRAF position corresponding to the current target graphics.

[0067] In step S3, a SRAF rule matching at least a portion of the ADI layer of the target graphic is obtained.

[0068] In step S2, a portion of the target graphic is selected, and the relative position of the SRAF of the target graphic is predicted. In this step, the SRAF rule corresponding to the portion of the target graphic is obtained. Since the ADI layer target value of the portion of the target graphic is determined, the relationship between the ADI layer and the SRAF rule can be obtained.

[0069] In this embodiment, the ruler function of the layout browser can be used to measure the SRAF rule. The SRAF rule includes the distance O from the edge of the SRAF in the second direction to the side of the contact hole region in the second direction, the width W of the SRAF in the second direction, and the distance S between adjacent SRAFs. Of course, the distance S needs to be measured only when there are more than two SRAFs. If there is only one SRAF, it is only necessary to measure the distance O from the SRAF to the side of the contact hole region and the width W of the SRAF.

[0070] Figures 6a to 6f Schematic diagram of the relative position relationship between ADI layers of the same type and different sizes and SRAF provided by an embodiment of the present invention. Figure 6a This is a schematic diagram of the ADI layer. Figure 6b is with Figure 6a Schematic diagram of the relative positions of the corresponding SRAFs. Figure 6c This is a schematic diagram of the ADI layer of another size. Figure 6d is with Figure 6c Schematic diagram of the relative positions of the corresponding SRAFs. Figure 6e This is a schematic diagram of the ADI layer of another size. Figure 6fis with Figure 6e The relative position diagram of the corresponding SRAF. Please refer to Figure 6a and Figure 6b As shown in the figure, the target length L of the ADI layer is 150nm, the width H is 60nm, and the spacing P is 126nm. In the corresponding SRAF rule, the width W is 18nm and the distance O is 67nm. Please refer to Figure 6c and 6d As shown in the figure, the target length L of the ADI layer is 200nm, the width H is 90nm, and the spacing P is 300nm. In the corresponding SRAF rule, the width W is 20nm and the distance O is 70nm. Please refer to Figure 6e and Figure 6f As shown in the figure, the target length L of the ADI layer is 250nm, the width H is 100nm, and the spacing P is 800nm. In the corresponding SRAF rule, the width W is 25nm and the distance O is 75nm. Figures 6a to 6f As shown in Table 1, the relationship between the ADI layer target value and the SRAF rule can be obtained.

[0071] Table 1

[0072] Target Graphics L H P W O 1 150 60 126 18 67 2 170 70 150 ？ ？ 3 200 90 300 20 70 4 220 95 500 ？ ？ 5 250 100 800 25 75

[0073] Exemplarily, the above Table 1 shows the values ​​of 5 target graphics, among which the 1st, 2nd and 5th target graphics are the target graphics whose SRAF positions are predicted in the previous step and whose SRAF rules are measured in this step. The 2nd target graphics and the 4th target graphics are not predicted and measured, so their SRAF rules are unknown. It should be noted that the above table only shows 5 target graphics by way of example, and in actual processing there may be a large number of target graphics, and the target graphics predicted and measured are only a few, and the present invention does not limit this.

[0074] In step S4, the SRAF rule table is obtained.

[0075] The SRAF rule table includes all the rules of the target graphics. In this embodiment, the SRAF rule matching the ADI layer of another part of the target graphics is obtained by interpolation, thereby obtaining the SRAF rule table. The SRAF rules of the majority of target graphics can be obtained from the SRAF rules of a minority of target graphics by interpolation.

[0076] The values ​​filled in the first, third and fifth rows in the following Tables 2 and 3 are obtained by measurement, while the blank ones need to be obtained by interpolation, such as W19 / O68, W19 / O67, and W19 / O72. It is worth noting that for the sake of convenience and simplicity, the present invention only shows data of three size picture structures in Figure 6, but in actual operation, more size graphic structure data will be measured, and the data will have a smaller interval, which can better prepare the SRAF rule table.

[0077] The interpolation method can avoid measuring all dimensional data, improve efficiency and reduce time consumption.

[0078] Table 2

[0079] P / L 150 170 200 220 250 126 W18 / O67 150 W19 / O68 300 W20 / O70 400 700 800 W25 / O75

[0080] Table 3

[0081] P / H 60 70 90 95 100 126 W18 / O67 W19 / O67 150 300 W20 / O70 400 W19 / O72 700 800 W25 / O75

[0082] For example, it can be seen from Table 2 that as the spacing P and length L in the ADI layer target value increase, the width W and distance O in the SRAF rule both increase. It can be seen from Table 3 that as the width H in the ADI layer target value increases, the width W in the SRAF rule increases, and when the width H increases but the spacing P remains unchanged, the width W in the SRAF rule increases but the distance O remains unchanged. When some values ​​in the SRAF rule are determined, the width W and distance O in the remaining SRAF rules can be obtained according to the rules.

[0083] In step S5, a pattern matching is performed on the ADI layer of the target pattern in the ADI layer of the GDS layout to find a pattern to be corrected of the same type but different size.

[0084] In this embodiment, a fluctuation range is set for part of the size of the ADI layer of the target graphic, and a graphic within the fluctuation range is searched in the ADI layer of the GDS layout as the graphic to be corrected. The ADI layer of the GDS layout refers to the ADI layer generated after the GDS layout is completed with Global Bias. The fluctuation range can be determined according to actual conditions, for example, according to the accuracy required for matching, but is not limited to this. A fluctuation range can also be determined in step S1, and the fluctuation range in step S1 can be the same as the fluctuation range in this step, or of course it can be different. Step S1 is to determine the fluctuation range and then select multiple sizes within the range to determine the target graphic, while in this step, matching is performed according to the fluctuation range to find as many graphics of the same type and different sizes as possible as the graphics to be corrected.

[0085] In step S6, the SRAF rule table is imported during program debugging and OPC correction is performed.

[0086] In this embodiment, the same type of graphics to be corrected with different sizes have been found in the previous step. The SRAF rule table is imported when debugging the program (Recipe), so that graphics of different sizes traverse the table to find their corresponding SRAF rules. According to the SRAF rules, the corresponding SRAF is added to the same type of graphics to be corrected with different sizes, and OPC correction is performed, so that the EPE of the simulated contour simulated by the final OPC model based on the mask exposure and the ADI target value is less than 2nm, thereby solving the local bulge phenomenon.

[0087] Figure 7 This is a profile electron scanning microscope image after correction and exposure using the OPC correction method provided by an embodiment of the present invention. Figure 4 and Figure 7 It can be seen that the OPC correction method for improving the local convexity phenomenon provided by the embodiment of the present invention can effectively improve the local convexity phenomenon.

[0088] Figure 8 is a schematic diagram comparing the correction method of the present invention with the correction method of the prior art, Fig. 9 yes Figure 8 Light intensity distribution diagram at the location of the middle dotted line. Figure 8 In the figure, curve 1 is the ADI layer, curve 2 is the mask pattern obtained by OPC correction after GlobalBias is performed in the prior art, and the simulated contour is obtained by simulating it, curve 3 is the mask pattern obtained by OPC correction after local size scaling of the ADI layer in the prior art (such as application number 202211164126.3), and the simulated contour is obtained by simulating it, and curve 4 is the mask pattern obtained by OPC correction using the method of the present invention, and the simulated contour is obtained by simulating it.

[0089] from Figure 8 It can be seen that curve 4 is closer to curve 1 (basically overlapped), that is, the OPC correction method for improving the local convexity phenomenon provided by the present invention is obviously better than the other two correction methods.

[0090] Please refer to Fig. 9 As shown, the horizontal axis is the detection point Figure 8 The relative position on the graph shown, the ordinate is the light intensity at the current detection point, the higher the ordinate value is, the higher the light intensity at the detection point position is, and the better the imaging effect is. Fig. 9 The curve shown has a maximum and a minimum. The larger the maximum value, the less likely it is that a bridge will occur. Fig. 9It can be seen that the maximum value of curve 4 is the largest, which means that compared with other correction methods, the correction method of the present invention is less likely to cause Bridge, that is, the local bulge phenomenon is improved.

[0091] In addition, the higher the normalized image logarithmic slope and the maximum value of the imaging contrast, the better, which means that the higher the imaging contrast of the pattern, the better the actual profile after photolithography development, that is, the less local protrusion phenomenon will occur, and the bridge risk (Bridge Risk) is further avoided. In addition, the lower the value of the process deviation loop, the better, which means that the actual profile after photolithography development is minimally affected by the process error disturbance. Table 4 compares three correction methods, among which GlobalBias is a method of simulating the simulated profile obtained by performing Global Bias in the prior art, and obtaining the mask pattern through OPC correction, and ADI Retarget is a method of simulating the simulated profile obtained by performing local size scaling of the ADI layer in the prior art, and obtaining the mask pattern through OPC correction. It can be seen from Table 4 that compared with the other correction methods, the correction method described in the present invention is not prone to Bridge, that is, the local protrusion phenomenon is improved, and the actual profile after photolithography development is minimally affected by the process error disturbance.

[0092] Table 4

[0093]

[0094] In summary, in the OPC correction method for improving the local convex phenomenon provided by the present invention, the initial graphic is first determined, and the partial size of the ADI layer of the initial graphic is scaled to form multiple target graphics of the same type and different sizes, and then the relative position of the SRAF of a part of the target graphic is predicted by the inverse lithography technology, and then the SRAF rule matching the ADI layer of at least a part of the target graphic is obtained, and then the SRAF rule table is obtained, and then the ADI layer of the target graphic is pattern matched in the ADI layer of the GDS layout to find the graphics to be corrected of the same type and different sizes, and then the SRAF rule table is imported when debugging the program and OPC correction is performed. The present invention selects a part of the target graphics of the same type and different sizes, predicts the relative position of SRAF and measures SRAF rules by inverse lithography technology, obtains the SRAF rule table, and then the ADI layer of the target graphic is pattern matched to find the graphics to be corrected of the same type and different sizes, and then the SRAF rule table is imported when debugging the program and OPC correction is performed, thereby reducing the EPE of the simulated contour and the ADI target value, improving the local convex phenomenon; and avoiding a long iterative process, reducing the time consumption of OPC.

[0095] In addition, the interpolation method is used to obtain the SRAF rule that matches the ADI layer of another part of the target graphic, which can avoid measuring all the dimensional data, improve efficiency and reduce time consumption.

[0096] The above description is only a description of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any changes or modifications made by a person skilled in the art in the field of the present invention based on the above disclosure shall fall within the scope of protection of the claims.

Claims

1. An OPC correction method for improving local convexity phenomenon, characterized in that: The following steps are involved: Determine an initial graphic, and scale a portion of the size of the ADI layer of the initial graphic to form a plurality of target graphics of the same type but different sizes; Predicting the relative position of the SRAF of a portion of the target pattern using an inverse lithography technique; Acquire a SRAF rule that matches an ADI layer of at least a portion of the target graphic; Get the SRAF rule table; Performing pattern matching on the ADI layer of the target pattern in the ADI layer of the GDS layout to find a pattern to be corrected of the same type but different size; as well as When debugging the program, the SRAF rule table is imported and OPC correction is performed.

2. The OPC correction method for improving the local convexity phenomenon according to claim 1, characterized in that: The method of scaling a portion of the ADI layer of the initial graphic to form a plurality of target graphics of the same type but different sizes includes: sequentially enlarging a portion of the ADI layer by a plurality of intervals and sequentially reducing a plurality of intervals to form a plurality of target graphics.

3. The OPC correction method for improving the local convexity phenomenon according to claim 2, characterized in that: The multiple spacing when enlarged is equal to the multiple spacing when reduced.

4. The OPC correction method for improving the local convexity phenomenon according to claim 1, characterized in that: The method for scaling a portion of the size of the ADI layer of the initial graphic to form a plurality of target graphics of the same type but different sizes includes: enlarging and reducing a portion of the size of the ADI layer to obtain a fluctuation range, and selecting a plurality of sizes from the fluctuation range to form a plurality of the target graphics.

5. The OPC correction method for improving the local convexity phenomenon according to claim 1, 2 or 4, characterized in that: The initial graphic includes a first part, a second part and a third part extending along the first direction and arranged in sequence along the second direction, a contact hole area is provided in the middle position between the first part and the third part, and the contact hole area protrudes to both sides in the second direction; the length of the contact hole area in the first direction is L, the protruding width in the second direction is H, and the spacing between adjacent first parts and second parts or between adjacent second parts and third parts is P; wherein the first direction is perpendicular to the second direction.

6. The OPC correction method for improving the local convexity phenomenon according to claim 5, characterized in that: The length L, width H and spacing P in the ADI layer of the initial graphic are scaled to form a plurality of target graphics of the same type but different sizes.

7. The OPC correction method for improving the local convexity phenomenon according to claim 5, characterized in that: The SRAF rule includes: a distance O from the edge of the SRAF in the second direction to the side of the contact hole region in the second direction, a width W of the SRAF in the second direction, and a distance S between adjacent SRAFs.

8. The OPC correction method for improving the local convexity phenomenon according to claim 1, 2 or 4, characterized in that: The method for obtaining the SRAF rule table includes: using an interpolation method to obtain the SRAF rule that matches the ADI layer of another part of the target graphic, thereby obtaining the SRAF rule table.

9. The OPC correction method for improving the local convexity phenomenon according to claim 1, 2 or 4, characterized in that: The method of performing graphic matching on the ADI layer of the target graphic in the ADI layer of the GDS layout to find a graphic to be corrected of the same type but different size includes: setting a threshold range for a partial size of the ADI layer of the target graphic, and searching for a graphic within the threshold range in the ADI layer of the GDS layout as the graphic to be corrected.

10. The OPC correction method for improving the local convexity phenomenon according to claim 1, 2 or 4, characterized in that: The method for importing the SRAF rule table and performing OPC correction when debugging a program includes: importing the SRAF rule table when debugging a program, finding the SRAF rule corresponding to each of the graphics to be corrected, adding corresponding SRAFs to graphics to be corrected of the same type but different sizes according to the SRAF rule, and performing OPC correction.

Images