TWI385549B - Method for amending layout patterns - Google Patents

Description

A method for correcting layout graphics

The present invention is directed to a method of modifying a layout pattern. In particular, the present invention relates to a method of modifying a layout pattern to conform to a reticle fabrication specification.

In the manufacturing process of semiconductor elements, techniques such as photolithography and etching are often used. The lithography technique involves transferring a complex integrated circuit pattern to a semiconductor wafer surface for etching, doping, and the like. These patterns need to be extremely accurate to correspond to the patterns of the front and back processes to create a precision integrated circuit. In the lithography step, when a reticle pattern is transferred to the surface of the wafer, variations often occur, affecting the performance of the semiconductor device. This deviation is related to the transferred pattern characteristics, the shape of the wafer, and various process parameters.

Among them, many methods of inspection, correction and compensation have been developed for the pattern deviation caused by optical proximity effects, process rules, optical rules, etc., in order to improve the quality after image transfer. For example, known methods include optical proximity correction (OPC), process rule check (PRC), and lithography rule check (LRC), etc., and commercially available optical proximity correction software. To detect problems such as pinches, bridges, and CD uniformity in the layout pattern. Such parties The method not only detects the problems in the layout pattern, but also corrects the layout pattern of the mask through the theoretical image. If the obtained correction pattern is correctly available, the output mask is output to obtain the correct image pattern on the wafer.

However, the above methods of inspection, correction, and compensation only consider the problem of the layout pattern itself, and do not incorporate the critical error (CD error) of the ray mask itself which is often up to several nanometers (nm). In other words, the above operations are all based on the assumption that the reticle fabrication process will perfectly transfer the corrected layout pattern. In fact, this is currently not possible. In particular, the corrected layout pattern usually only barely passes the above-mentioned operation method, and hardly leaves any process window on the fabrication of the mask. . Therefore, using these masks for lithography and etching, the resulting layout pattern is often problematic. Fig. 9 illustrates a pattern formed by the transfer of the layout pattern after the optical proximity correction. In Figure 9, a set of separate but similar primary features 910 and 920 are illustrated, respectively. However, since the main features 910 and 920 each contain errors caused by reticle fabrication, the main features 910 and 920 are similar, but the main feature 920 has bridged turns.

Not only that, there is currently no model for testing, correcting, and compensating for errors caused by making masks. The quality of the mask pattern is ultimately revealed by the layout pattern obtained after exposure and etching. However, it has not been proposed to simulate the reticle pattern obtained after the production, and to confirm whether the layout pattern of the reticle is applicable. Therefore, it is still necessary to use a manual operation method to correct the layout pattern on the reticle. This is a troublesome and correct system The way the maker is not convenient.

Therefore, there is an urgent need for a method of pre-correcting the layout pattern to conform to the reticle fabrication specifications. In addition, a method for establishing an optical proximity correction model is needed. These optical proximity correction models can be applied to optical proximity corrections, and directly obtain layout patterns that can be used to make masks, making the transfer of layout patterns more accurate.

The present invention thus provides a method of modifying the layout pattern to conform to the reticle fabrication specifications. In another aspect of the invention, a method for establishing an optical proximity correction model is also provided. These optical proximity correction models can be applied to optical proximity corrections and directly result in layout patterns that can be used to make reticle.

In one aspect of the invention, a method for modifying a layout pattern is provided. First, a layout graphic is provided that includes at least one rnain feature. Secondly, at least one optical proximity correction (OPC) is performed on the layout pattern to obtain a first correction pattern that can be used. Thereafter, the first correction pattern is subjected to a positive sizing process to obtain a forward calibration pattern. Again, confirm that the forward calibration graphic is available. Then, a negative sizing process is performed on the first correction pattern to obtain a negative calibration pattern. Then, confirm whether the negative adjustment graphic is available. Continuing, when both the forward adjustment pattern and the negative adjustment pattern are available, the output is a first correction pattern for both the forward adjustment pattern and the negative adjustment pattern to create a mask.

In another aspect of the invention, a method for establishing an optical proximity correction model is provided. First, a layout graphic is provided that includes at least one main feature. Secondly, at least one optical proximity correction is performed on the layout pattern to obtain a first correction pattern that can be used. Thereafter, the first correction pattern is subjected to a forward calibration process to obtain a forward calibration pattern. Again, confirm that the forward calibration graphic is available. Then, the negative correction process is performed on the first correction pattern to obtain a negative adjustment pattern. Then, confirm whether the negative adjustment graphic is available. When the forward calibration procedure and the negative calibration procedure are performed, the data of the forward calibration procedure and the negative calibration procedure are obtained together. Continuing, when both the forward calibration pattern and the negative calibration pattern are available, the forward calibration program data and the negative calibration program data are collected to establish the desired optical proximity correction model. This optical proximity correction model can be applied to optical proximity corrections, and directly to the layout pattern that can be used to make the reticle.

In one aspect of the invention, a method of modifying a layout pattern to increase a process window of a reticle to correct reticle fabrication errors is provided. In another aspect of the invention, a method for establishing a modified model is also provided. These modified models can be applied to optical proximity corrections and directly yield layout patterns that can be used to make masks.

Fig. 1 is a flow chart showing the main flow of the method for modifying the layout pattern of the present invention. The method for modifying the layout graphics 100 of the present invention comprises: Step 110: Providing a layout graphic.

Step 120: Perform at least one optical proximity correction on the layout graphic to obtain a first modified graphic that can be used.

Step 130: Perform a forward compensation process on the first correction pattern to obtain a forward adjustment pattern that can be used.

Step 140: Perform a negative compensation procedure on the forward calibration pattern that is available, and obtain a negative adjustment pattern that can be used.

Step 150: Output a first correction pattern that is applicable to both the optical proximity correction, the forward compensation procedure, and the negative compensation procedure to create a mask.

First, in step 110, the layout patterns may be circuit patterns that need to be transferred, such as layout patterns of any stage of doping regions, polysilicon, contact holes, etc. of the SRAM. In this layout pattern, the resulting geometry is referred to as the main feature. In other words, the layout graphic in step 110 will contain at least one main feature.

Next, in step 120, the original layout pattern is subjected to optical proximity correction to obtain a first correction pattern that is considered useful for optical proximity correction rules. Figure 2 illustrates a method of modifying a layout pattern of the present invention. The original layout pattern originally having the main features becomes the first correction main feature 210 in the first correction pattern 201 after the optical proximity correction. In addition, a first auxiliary feature 220 is also generated. Using the optical proximity correction, the portion of the original layout pattern that may be distorted by the optical proximity effect, that is, the main feature, is corrected to become the first correction main feature 210 in the first correction pattern 201, and The first modified assist feature 220 can be additionally added. These optical proximity corrections are available for use in commercially available optical proximity correction software to detect and correct problems in the narrow areas, bridges, and critical dimension uniformity of each of the main features in the layout pattern, and thus will not be repeated here.

It should be noted that in step 120, these optical proximity corrections may operate more than once for the original layout pattern, because each optical proximity correction may still have major features that still do not conform to the process/optical rules, so Step 120 can further include the following sub-steps. Figure 3 illustrates a method of modifying a layout pattern of the present invention, resulting from the sub-steps of step 120. For example, after "sub-step (120'): optical proximity correction of the layout pattern", sub-step (121): using the process rule check and the optical rule check respectively to confirm the first after the optical proximity correction Fix whether the graphics are available.

If the first correction pattern is available, the next step 130 is entered. If the first correction pattern is unusable, then proceed to: sub-step (122): Correcting the first correction pattern again with optical proximity correction.

Sub-steps (121) and sub-steps (122) may be repeated as many times as needed, until a first corrected pattern is available. It is then possible to perform multiple optical proximity corrections before finally obtaining all of the first modified main features 210 that conform to the process rules and optical rules of the layout pattern, referred to as the first corrected graphic. When the sub-step (120') and the sub-step (122) are performed, the correction amount of the first correction pattern is corrected, which may be determined depending on the first correction main feature 210 which is unacceptable. The way to correct the first correction graphic can be manual Corrections can also be automated, such as using commercially available software.

In step 130, a forward compensation procedure is initiated for the first modified graphics that are available for the purpose of obtaining a positive calibration pattern that is useful. Such forward compensation procedures may generally involve first performing a positive sizing process on the first corrected pattern to obtain a forward calibrated pattern, and then confirming whether the forward calibrated pattern is available. Therefore, the forward compensation procedure may contain many sub-steps.

The so-called forward calibration procedure may be, for example, sizing up the first correction pattern that is available to obtain a forward calibration pattern. These forward adjustment patterns can be used to verify the error in the form of the layout pattern when the mask is produced. For the method of performing the equal-magnification on the first correction pattern, for example, the first correction main feature, the first correction auxiliary feature, or both may be enlarged in an equal manner. The amount of iso-magnification is determined according to the error of the mask during the fabrication process, and even the difference of the subsequent etching process can be considered, and the process masks of different film layers have different deviation adjustments, such as the process mask of the polysilicon layer. It can be an isotropic magnification of 4~8 nanometers (nm), and the metal layer process mask can be an isometric magnification of 6~10 nanometers (nm). In addition, the first modified main feature and the first modified auxiliary feature may each have an independent amount of isomorphic amplification, and the first modified main feature and the first modified auxiliary feature may respectively have different isomorphic amplifications as needed. the amount.

Similarly, in step 130, these forward calibration procedures may operate more than once for the first modified graphics that are available, because each forward adjustment The first modified main feature, and/or the first modified auxiliary feature, which still does not conform to the process/optical rules, may remain after the school program. Thus, step 130 can further include the following sub-steps. Figure 4 illustrates a method of modifying a layout pattern of the present invention, illustrating the sub-steps derived from step 130. For example, after "sub-step (130'): performing an equal-magnification on the first corrected pattern that can be used to obtain a forward-corrected pattern", it is possible to perform: sub-step (131): using process rule inspection and optics, respectively. A rule check is made to confirm whether this positive adjustment pattern is available.

If the forward calibration graphic is available, it can proceed directly to the next step 140. If the forward calibration pattern is unusable, proceed to: sub-step (132): Correct the first correction pattern again with optical proximity correction.

Sub-steps (131) and sub-steps (132) may be repeated as many times as needed, until the forward calibration pattern is used for inspection of process rules and optical rules. Therefore, it may be necessary to perform multiple optical proximity corrections, and finally all the first modified main features and the first modified auxiliary features in the forward calibration pattern are applicable to the process rule inspection and the optical rule, which is called Positive adjustment of the graph. When the sub-step (130') and the sub-step (132) are performed, the correction amount of the forward adjustment pattern is corrected, which can be determined depending on the forward adjustment pattern which is unacceptable. Correct the way the pattern is corrected in the forward direction, either manually or automatically.

In step 140, a negative compensation procedure is continued for such available forward calibration patterns in order to obtain a negative adjustment pattern that can be used. These negative compensation procedures can generally include the first corrections that can be used first. The graph performs a negative sizing process to get a negative calibration pattern and then confirms whether the negative calibration pattern is available. Therefore, a negative compensation procedure may involve many sub-steps.

The so-called negative calibration procedure can be, for example, sizing down the available forward calibration pattern to obtain a negative calibration pattern. These negative adjustment patterns can be used to verify the error of the layout reduction pattern of the reticle during production. For the method of performing the contour reduction on the first correction pattern, for example, the first correction main feature, the first correction auxiliary feature, or both may be used in the previous step 130. The amount of contour reduction depends on the error of the mask during fabrication, and even the difference of the subsequent etching process can be considered, and the process masks of different film layers have different deviation adjustments, for example, the process mask of the polysilicon layer can be The shape is reduced by 4~8 nm, and the process mask of the metal layer can be reduced by 6~10 nm (nm). In addition, the first modified main feature and the first modified auxiliary feature may each have an independent amount of equal reduction, and the first modified main feature and the first modified auxiliary feature may respectively have different isomorphic reductions as needed. the amount.

In step 140, such negative calibration procedures may operate more than once for the positive adjustment pattern that is available, as each negative calibration procedure may still have a process/optical rule that is still not in compliance with the process/optical rules. The first modified primary feature, and/or the first modified assisted feature. Thus, step 140 can further include the following sub-steps. Figure 5 illustrates a method of modifying a layout pattern of the present invention, illustrating the sub-steps derived from step 140. For example, in the "Substeps (140'): After the equalization reduction of the first correction pattern that can be used to obtain the negative adjustment pattern, a sub-step (141) can be performed: using the rule rule test and the optical rule test to confirm whether the negative adjustment pattern is OK. use.

If the negative calibration pattern is available, then proceed directly to the next step 150. If the negative calibration pattern is not available, proceed to: sub-step (142): Correct the first correction pattern again with optical proximity correction.

Sub-steps (141) and sub-steps (142) may be repeated as many times as needed, until the negative alignment pattern is used for inspection of process rules and optical rules. Therefore, it may be necessary to perform multiple optical proximity corrections, and finally all the first modified main features and the first modified auxiliary features in the negative calibration pattern are applicable to the process rule inspection and optical rules, and are said to be useful. Negative adjustment of the graph. When the sub-step (140') and the sub-step (142) are performed, the correction amount of the negative adjustment pattern is corrected, which can be determined by the negative adjustment pattern which is unacceptable. The correction method of the negative adjustment pattern can be corrected manually or automatically.

Please note that steps 130 and 140 do not have a specific sequence. In other words, on the one hand, step 130 can be performed before step 140, and on the other hand, step 130 can also be performed after step 140.

Since the negative adjustment pattern that can be used is derived from the positive adjustment pattern that can be used, and the positive adjustment pattern that can be used is derived from the first correction pattern that can be used, the negative adjustment pattern can be used. "Step 120: Perform optical proximity correction on the layout pattern", and "Sub-step (131) of step 130 respectively use Process rule inspection and optical rule inspection to confirm whether the forward calibration pattern is available, and sub-step (141) of step 140: using process rule inspection and optical rule inspection to confirm whether the negative calibration pattern is available. To be useful. Thus, the original first correction pattern becomes a layout pattern that can be used at this time, and is a negative correction pattern that can be used, and a positive correction pattern that can be used is also the first correction pattern that can be used, that is, for optical proximity. Correction, forward compensation procedures and negative compensation procedures are all available. Subsequently, it is possible to output a first correction pattern that is applicable to both the forward adjustment pattern and the negative adjustment pattern to create a mask. The image pattern obtained by the mask can be correctly imaged on the wafer.

When the sub-steps (130'/140') and the sub-steps (132/142) of step 130 and step 140 are performed, the correction amount of the forward adjustment pattern/negative adjustment pattern is corrected separately, and the correction amount can be corrected. It depends on the adjustment pattern/negative adjustment pattern. In order to avoid correcting the positive adjustment pattern/negative alignment pattern while manually correcting it, in another aspect of the invention, a method for establishing an optical proximity correction model is provided.

The method for establishing an optical proximity correction model is to obtain a forward calibration program data and a first correction pattern when performing a forward calibration procedure on the first correction pattern to obtain a forward calibration pattern. When the negative calibration procedure is performed and the negative calibration pattern is obtained, the negative calibration program data is also obtained. When both the forward calibration pattern and the negative calibration pattern are available, the forward calibration program data and the negative calibration program data are collected to establish the desired optical proximity correction model.

Figure 6 illustrates a flow chart of the main flow of the method for establishing an optical proximity correction model of the present invention. The present invention establishes an optical proximity correction model method 500, comprising: Step 510: Providing a layout graphic.

Step 520: performing at least one optical proximity correction on the layout graphic to obtain a first modified graphic that can be used.

Step 530: Perform a forward compensation process on the first correction pattern to obtain a forward calibration pattern that can be used, and adjust the program data in the forward direction.

Step 540: Perform a negative compensation procedure on the forward calibration pattern that is available, and obtain a negative adjustment pattern that can be used, and adjust the program data in the negative direction.

Step 550: Collect forward calibration program data and negative calibration program data to establish an optical proximity correction model.

First, in step 510, the layout graphics can be a circuit graphic that needs to be transferred, such as a static random access memory layout graphic. In the layout graph, the resulting geometry is called the main feature. In other words, the layout graphic in step 510 will contain at least one primary feature.

Next, in step 520, the original layout pattern is subjected to optical proximity correction to obtain a first correction pattern that is considered useful for optical proximity correction rules. These optical proximity correction methods are similar to the step 120, and will not be described here. Additionally, step 520 can further include sub-steps similar to step 120. For example, sub-step (521) refers to sub-step (121), and sub-step (522) refers to sub-step (122).

In step 530, a forward compensation procedure is initiated for the first modified graphics that are available for the purpose of obtaining positive forward calibration graphics and forward calibration programming data. These forward compensation procedures are substantially similar to step 130. Therefore, step 530 may also include a number of sub-steps similar to step 130. Figure 7 illustrates a method of establishing an optical proximity correction model of the present invention, illustrating the sub-steps derived from step 530. For example, after the sub-step (530'): performing an equal-magnification on the first modified graphic that can be used to obtain the forward-corrected graphic and the forward-corrected program data, the sub-step (531) can be performed: Use process rule inspection and optical rule inspection to confirm whether the forward calibration pattern is available.

If the forward calibration pattern is available, then proceed directly to the next step 540. If the forward calibration pattern is unusable, proceed to: sub-step (532): correct the first correction pattern again with optical proximity correction and obtain the forward calibration program data.

Because the operation of the positive adjustment pattern can be used more than once, each time the correction is made, the corresponding correction data can be obtained, which is called the forward adjustment program data. The more the number of corrections, the more important types of corrections will be, and the more positive the content of the program will be. When the positive adjustment pattern is finally available, the forward calibration data can be quickly obtained as a guide for positive adjustment graphics.

Since it is possible to perform multiple optical proximity corrections, all the first modified main features and the first modified auxiliary features in the forward calibration pattern are finally available for the process rule inspection and optical rules, if there is no forward calibration With the assistance of the program data, it may be necessary to manually correct the first correction pattern multiple times in order to obtain the forward adjustment pattern that can be used. If the forward calibration program data is used, the forward adjustment pattern is automatically corrected, and it is expected that the time for obtaining the forward adjustment pattern can be greatly shortened.

Next, in step 540, the negative compensation procedure for the forward calibration patterns that are available is continued to obtain the negative adjustment pattern and the negative calibration program data. These negative compensation procedures are substantially similar to step 140. Therefore, step 540 may also include a number of sub-steps similar to step 140. Figure 8 illustrates a method of establishing an optical proximity correction model of the present invention, illustrating the sub-steps derived from step 540. For example, after the sub-step (540'): performing the contour reduction on the first modified graphic that is available, and obtaining the negative calibration graphic and the negative calibration data, the sub-step (541) can be performed: Use process rule inspection and optical rule inspection to confirm whether the negative calibration pattern is available.

If the negative calibration pattern is available, then proceed directly to the next step 550. If the negative calibration pattern is not available, proceed to: sub-step (542): correct the first correction pattern with the optical proximity correction again and obtain the negative calibration program data.

Because the operation of adjusting the graphics in the negative direction may be more than once, the corresponding correction data can be obtained for each correction, which is called the negative calibration program data. The more the number of corrections, the more important types of important features are corrected, and the richer the content of the negative calibration program data. When the negative adjustment pattern is finally available, the negative adjustment program data can become Get a quick guide to negative tuning graphics.

Since it is possible to perform multiple optical proximity corrections, all the first modified main features and the first modified auxiliary features in the negative calibration pattern are finally available for the process rule inspection and optical rules, if there is no negative adjustment. With the assistance of the program data, it may be necessary to manually correct the negative adjustment pattern multiple times in order to obtain the negative adjustment pattern that can be used. If negative adjustment program data is used, correcting the negative forward adjustment pattern in an automatic manner can greatly shorten the time required to obtain a negative adjustment pattern.

Please note that steps 530 and 540 do not have a specific sequence. In other words, on the one hand, step 530 can be performed before step 540, and on the other hand, step 530 can also be performed after step 540.

When both the forward calibration pattern and the negative calibration pattern are available, the previously obtained forward calibration program data and the negative calibration program data are collected to establish the desired optical proximity correction model for later use. .

The above are only the preferred embodiments of the present invention, and all changes and modifications made to the scope of the present invention should be within the scope of the present invention.

201‧‧‧First correction graphic

210‧‧‧ Main features of the first amendment

220‧‧‧First modified auxiliary features

910, 920‧‧‧ main features

Fig. 1 is a flow chart showing the main flow of the method for modifying the layout pattern of the present invention.

Figure 2 illustrates a method of modifying a layout pattern of the present invention.

Figure 3 illustrates a method of modifying a layout pattern of the present invention, resulting from the sub-steps of step 120.

Figure 4 illustrates a method of modifying a layout pattern of the present invention, illustrating the sub-steps derived from step 130.

Figure 5 illustrates a method of modifying a layout pattern of the present invention, illustrating the sub-steps derived from step 140.

Figure 6 illustrates a flow chart of the main flow of the method for establishing an optical proximity correction model of the present invention.

Figure 7 illustrates a method of establishing an optical proximity correction model of the present invention, illustrating the sub-steps derived from step 530.

Figure 8 illustrates a method of establishing an optical proximity correction model of the present invention, illustrating the sub-steps derived from step 540.

Figure 9 illustrates the pattern formed by the transfer of the layout pattern after the conventional optical proximity correction.

Claims (28)

A method of modifying a layout pattern, comprising: providing a layout pattern including at least one main feature; performing at least one optical proximity correction (OPC) on the layout pattern a first correction pattern; performing a positive sizing process on the first correction pattern to obtain a forward adjustment pattern; confirming whether the forward adjustment pattern is applicable; and the first correction pattern Performing a negative sizing procedure to obtain a negative calibration pattern; confirming whether the negative calibration pattern is available; and both the forward calibration pattern and the negative calibration pattern are applicable And outputting the first correction pattern for the positive adjustment pattern and the negative adjustment pattern to produce a mask. The method of claim 1, wherein the layout graphic comprises a static random access memory layout graphic. The method of claim 1, wherein the method for obtaining the first correction pattern further comprises: (a) using a process rule check and an optical rule check to confirm whether the first correction pattern after the optical proximity correction is applicable (b) when the first correction pattern is unusable, the optical proximity correction corrects the unusable A correction pattern; and repeating (a) and (b) until the first correction pattern is available, the correction amount is dependent on the first correction pattern. The method of claim 1, wherein the first modified graphic comprises a first modified primary feature and a first modified auxiliary feature. The method of claim 4, wherein the forward calibration process comprises: sizing up the first modified graphic to obtain the forward calibration graphic. The method of claim 5, wherein the first modified graphic is subjected to the isomorphic enlargement, comprising: at least one of the first modified primary feature and the first modified auxiliary feature is enlarged. The method of claim 6, wherein the first modified primary feature and the first modified auxiliary feature are independently enlarged. The method of claim 1, wherein the confirming whether the forward adjustment pattern is acceptable comprises: using a process rule check and an optical rule check respectively to confirm whether the forward adjustment pattern is applicable. The method of claim 8, further comprising: optically proximate to correct the first correction map when the forward calibration pattern is unusable The shape, the correction amount is determined by the forward adjustment pattern, until the forward adjustment pattern is used for the inspection of the process rule and the inspection of the optical rule. The method of claim 4, wherein the negative calibration procedure comprises: sizing down the first modified graphic to obtain the negative calibration graphic. The method of claim 10, wherein the performing the scaling of the first modified graphic comprises contouring at least one of the first modified primary feature and the first modified assisted feature. The method of claim 11, wherein the first modified primary feature and the first modified secondary feature are independently reduced in size. The method of claim 1, wherein the confirming whether the negative alignment pattern is acceptable comprises: using a process rule check and an optical rule check respectively to confirm whether the negative alignment pattern is applicable. The method of claim 13, further comprising: optically proximate to correct the first correction pattern when the negative adjustment pattern is unusable, and the correction amount is determined by the negative adjustment pattern until the negative The calibration pattern can be used for the inspection of the process rule and the inspection of the optical rule. A method of establishing an optical proximity correction model, comprising: Providing a layout graphic comprising at least one main feature; performing at least one optical proximity correction on the layout graphic to obtain a first correction graphic; performing a forward calibration process on the first correction graphic And obtaining a positive adjustment pattern, and obtaining a positive adjustment program data; confirming whether the forward adjustment pattern is applicable; performing a negative calibration procedure on the first correction pattern to obtain a negative adjustment Graphic, and obtaining a negative calibration program data; confirming whether the negative adjustment pattern is applicable; and collecting the positive adjustment when both the forward adjustment pattern and the negative adjustment pattern are available The program data and the negative calibration program data are used to establish the optical proximity correction model. The method of claim 15, wherein the layout graphic comprises a static random access memory layout graphic. The method of claim 15, wherein the method for obtaining the first correction pattern further comprises: (a) using a process rule check and an optical rule check to confirm whether the first correction pattern after the optical proximity correction is applicable (b) optically abutting the first correction pattern when the first correction pattern is unsuitable; and repeating (a) and (b) until the first correction pattern is available, the correction amount is It is unsuitable for the first correction pattern. The method of claim 15, wherein the first modified graphic comprises a first modified primary feature and a first modified auxiliary feature. The method of claim 18, wherein the forward calibration process comprises: sizing up the first modified graphic to obtain the forward calibration graphic. The method of claim 19, wherein the first modified graphic is subjected to the iso-amplification, the method comprising: at least one of the first modified primary feature and the first modified assisted feature. The method of claim 20, wherein the first modified primary feature and the first modified secondary feature are independently enlarged. The method of claim 15, wherein the confirming whether the forward calibration pattern is acceptable comprises: using a process rule check and an optical rule check respectively to confirm whether the forward calibration pattern is available. The method of claim 22, further comprising: optically approximating the first correction pattern when the forward adjustment pattern is unusable, the correction amount is determined by the forward adjustment pattern being unsuitable until the positive The calibration pattern can be used for the inspection of the process rule and the inspection of the optical rule. The method of claim 15, wherein the negative calibration procedure comprises: sizing down the first modified graphic to obtain the negative calibration graphic. The method of claim 24, wherein the performing the scaling of the first modified graphic comprises contouring at least one of the first modified primary feature and the first modified assisted feature. The method of claim 25, wherein the first modified primary feature and the first modified secondary feature are independently reduced in size. The method of claim 15, wherein the confirming whether the negative alignment pattern is acceptable comprises: using a process rule check and an optical rule check respectively to confirm whether the negative alignment pattern is applicable. The method of claim 27, further comprising: optically proximate to correct the first correction pattern when the negative adjustment pattern is unusable, the correction amount being dependent on the negative adjustment pattern until the negative The calibration pattern can be used for the inspection of the process rule and the inspection of the optical rule.