TWI644169B - Computer-implemented method, non-transitory computer-readable medium, and system for reticle inspection using near-field recovery - Google Patents

Abstract

The present invention provides systems and methods for detecting defects on a reticle. Embodiments include generating and/or using a pair of predetermined segments comprising a reticle pattern and a data structure corresponding to the near field data. The near field data of the predetermined segments may be determined by regression based on an actual image of a reticle produced by one of the reticle detection systems. Then, detecting a reticle may include: comparing two or more segments of the pattern included in one of the detection zones on the reticle to the predetermined segments; and based on at least The predetermined segment, most similar to one, assigns near field data to the at least one of the segments. The assigned near field data can then be used to simulate an image that will be formed by the detector for the reticle, and the image can be compared to an actual image produced by the detector for defect detection.

Description

Computer implemented method, non-transitory computer readable medium and system for reticle detection using near field restoration

The present invention is generally directed to systems and methods for mask detection using near field restoration.

The following description and examples are not considered to be prior art as they are included in this section.

Fabricating semiconductor devices such as logic and memory devices typically involves processing a plurality of semiconductor fabrication processes, such as a substrate of a semiconductor wafer, to form various features and levels of the semiconductor device. For example, lithography involves transferring a pattern from a mask to a semiconductor fabrication process of a resist disposed on a semiconductor wafer. Additional examples of semiconductor fabrication processes include, but are not limited to, chemical mechanical polishing (CMP), etching, deposition, and ion implantation. A plurality of semiconductor devices can be fabricated in one configuration on a single semiconductor wafer and then separated into individual semiconductor devices.

Test procedures are used at various steps during a semiconductor fabrication process to detect defects on the reticle to promote higher yields and, therefore, higher profits in the manufacturing process. Detection is always an important part of making semiconductor devices such as ICs. However, as the size of semiconductor devices decreases, detection becomes even more important for successful fabrication of semiconductor devices.

Existing reticle detection methods utilize one of several imaging modes to detect a mask. The most common detection mode known as reticle plane detection (RPI) involves capturing a high resolution transmissive and reflected image of a reticle and processing the two images together. Low numerical aperture Another detection method for (NA) detection (LNI) involves a pattern that mimics the optical conditions of the wafer scanner to capture an image of the transmitted light by one of the NAs that is lower than the RPI by the illumination conditions of the approximate scanner. Another method of contamination detection, referred to as SL, involves analyzing two images of RPI transmission and reflection to find defects that protrude from the background pattern.

Many reticle inspection methods use a die-to-database type comparison to detect defects on the reticle. This test typically involves acquiring a microscope image of a reticle. Based on a database describing a given pattern on the reticle, one of the images of the reticle that the inspection microscope is expected to observe can be calculated or simulated. The acquired optical image and the calculated or simulated image can then be compared to detect defects on the reticle.

Calculating a reticle image as described above can include calculating the light diffraction caused by the reticle. The Maxwell equation fully and accurately illustrates the diffraction of electromagnetic waves caused by a reticle. However, there is no practical way to accurately solve the Maxwell equation for an entire reticle for the required detection time (which is one to two hours). Some currently used methods use an approximation such as the Kirchhoff approximation to estimate the diffraction field. However, this limitation calculates the accuracy of the image and thus limits the smallest defect that can be detected.

The advancement of optical proximity correction (OPC) results in an ever-increasing complexity of the patterns written on a light mask or reticle. Shutter detection is typically performed using a high-end optical microscope, such as the high-end optical microscope described above, to identify the written pattern by a high-level algorithm. Due to OPC complexity, it is increasingly difficult to separate fatal (printing) defects from non-fatal (disturbing) defects. The traditional approach to solving this problem is to: (a) use manual rules based on the shape and size of the defect for manual handling, and (b) establish empirical rules and automatic defect classification (ADC) software to automatically handle a large number of defects, and (c) Using an aerial imaging tool to acquire an optical image under appropriate lithography conditions to handle defects.

However, there are several disadvantages with such currently used methods and systems. For example, currently used methods require a user to perform manual defect handling to re-detect defects one by one. The increased OPC complexity means that users often have difficulty dividing by several rules of experience. Off-print defects and non-print defects. In addition, potentially a large number of defects can overload the user and prevent defect handling from being completed within a reasonable period of time. The rule-based approach is not directly related to the printability of a defect. The increased OPC complexity of inverse lithography (ILT) can limit the usefulness of this approach. In addition, aerial imaging tools utilize a combination of hardware and algorithms to mimic scanner vector imaging effects. However, the accuracy of such tools has never been verified completely independently. More critically, these methods do not consider complex photoresist development or etching procedures after imaging is completed on a scanner. Studies have shown that there is a poor correlation between the size of the defect measured on an aerial imaging tool and the size of the defect measured on a wafer.

Accordingly, it would be advantageous to develop methods and/or systems for reticle detection that do not have one or more of the disadvantages set forth above.

The following description of various embodiments should not be construed as limiting the scope of the claims.

One embodiment relates to a computer implemented method for detecting defects on a reticle. The method comprises: separating one of the detection regions included in the reticle into two or more segments; and separately comparing the two or more segments and being included in a data structure Scheduled segmentation. The data structure includes a pair of predetermined segments of the reticle pattern and corresponding near field data. The method also includes assigning near field data to the one of the two or more segments based on one of the predetermined segments most similar to at least one of the two or more segments At least one. In addition, the method includes: generating near field data of the detection area based on the assigned near field data; and simulating the detection area based on the generated near field data to be formed by one detector of a reticle detection system An image. The method further includes: acquiring an actual image generated by the detector in the detection area on a physical version of the reticle; and detecting the reticle by comparing the simulated image with the actual image defect. Performing such separations, individual comparisons, with one or more computer systems, Assign, generate, simulate, acquire, and detect steps.

Each of the steps of the computer implemented method can be further performed as set forth herein. Additionally, the computer implemented method can include any of the other steps of any of the other methods set forth herein. Moreover, the computer implemented method can be performed by any of the systems set forth herein.

Another embodiment is directed to a non-transitory computer readable medium storing program instructions executable on a computer system for performing a computer implemented method for detecting a defect on a reticle . The computer implemented method includes the steps of the method set forth above. The computer readable medium can be further configured as set forth herein. The steps of the computer implemented method can be performed as further described herein. In addition, the computer implemented method for which the program instructions can be executed can include any other steps of any of the other methods set forth herein.

An additional embodiment relates to a system configured to detect defects on a reticle. The system includes a reticle detection subsystem including one of the actual images of one of the detection zones configured to produce one of the reticle. The system also includes one or more computer subsystems configured to perform the separate, separate comparison, assignment, generation, simulation, acquisition, and detection steps of the method set forth above. The system can be further configured as set forth herein.

Yet another embodiment is directed to a computer implemented method for setting a reticle inspection program. The method includes separating pattern information of a reticle into predetermined segments. The method also includes determining near field data for the predetermined segments based on one or more actual images of the reticle obtained by one of the reticle detection systems. The near field data of the predetermined segments is determined by regression based on the one or more actual images of the reticle. Additionally, the method includes generating a data structure including the predetermined segments in pairs and the near field data corresponding to the predetermined segments. The method further includes setting the reticle detection program by incorporating information of the generated data structure into a prescription for the reticle detecting procedure such that the reticle is Comparing two or more segments of the reticle or another reticle to the predetermined segment during the detection procedure and at least one of the two or more segments The near field material of one of the most similar predetermined segments is assigned to the at least one of the two or more segments. The separation, determination, generation, and setup steps are performed by one or more computer systems.

Each of the steps of the computer implemented method can be further performed as set forth herein. Additionally, the computer implemented method can include any of the other steps of any of the other methods set forth herein. Moreover, the computer implemented method can be performed by any of the systems set forth herein.

Still another embodiment relates to another computer implementation method for setting up a reticle inspection program. The method includes separating pattern information of a reticle into predetermined segments. The method also includes determining near field data for the predetermined segments by numerically solving an equation governing the electromagnetic fields of the predetermined segments. Additionally, the method includes generating a data structure including the predetermined segments in pairs and the near field data corresponding to the predetermined segments. The method further includes setting the reticle inspection program by incorporating information of the generated data structure into a prescription for the reticle detecting procedure such that the reticle or another is compared during the reticle detecting procedure Two or more segments of a detection zone on a reticle and the predetermined segments, and at least one of the two or more segments being most similar to the predetermined segments The near field data of one of the ones is assigned to the at least one of the two or more segments. The separation, determination, generation, and setup steps are performed by one or more computer systems.

Each of the steps of the computer implemented method can be further performed as set forth herein. Additionally, the computer implemented method can include any of the other steps of any of the other methods set forth herein. Moreover, the computer implemented method can be performed by any of the systems set forth herein.

200‧‧‧Non-transitory computer-readable media

202‧‧‧Program instructions

204‧‧‧ computer system

300‧‧‧Photomask Detection Subsystem

302‧‧‧Light source

304‧‧‧Homostat

306‧‧ ‧ aperture

308‧‧‧ Concentrating lens

310‧‧‧Photomask

312‧‧‧

314‧‧‧ objective lens

316‧‧ ‧ aperture

318‧‧‧ lens

320‧‧‧beam splitter

322a‧‧‧Detector

322b‧‧‧Detector

322c‧‧‧Detector

324‧‧‧Computer subsystem

Benefit from the following detailed description of the preferred embodiments and with reference to Further advantages of the present invention will become apparent to those skilled in the art, in which: Figure 1 illustrates an embodiment of a computer implementation for setting up a reticle inspection procedure and for detecting a reticle. One of the drawbacks is a flowchart of one embodiment of a computer implementation method; FIG. 2 is a block diagram illustrating an embodiment of a non-transitory computer readable medium for causing A computer system executes program instructions of one or more of the computer implementation methods set forth herein; and FIG. 3 illustrates one side of an embodiment of a system configured to detect defects on a reticle A schematic of one of the views.

While the invention is susceptible to various modifications and alternatives, the embodiments are shown in the drawings and are illustrated in detail herein. These drawings may not be drawn to scale. It is to be understood, however, that the invention is not intended to be All modifications, equivalent forms, and alternatives within the spirit and scope of the invention.

Turning now to the schema, it should be noted that the figures are not drawn to scale. In particular, the proportions of some of the elements of the figures are greatly exaggerated to emphasize the characteristics of the elements. It should also be noted that the figures are not drawn to the same scale. The same element symbols have been used to indicate elements that can be similarly configured in one or more of the figures. Any of the elements illustrated and shown may comprise any suitable commercially available element, unless otherwise stated herein.

Embodiments set forth herein are generally directed to detecting a reticle by utilizing a library that illustrates a reticle pattern. For example, as further described herein, the embodiments can be used to perform grain-to-database inspection of the reticle using near field restoration. Embodiments set forth herein can be used to detect reticle for pattern defects and contamination.

The terms "mask", "mask" and "light mask" are used interchangeably herein and are intended to mean any light known in the art to transfer a pattern to another substrate such as a wafer. Cover or mask.

A "substantially high resolution image" (such as the term commonly used in reticle detection technology) refers to an image of a reticle in which the features printed on the reticle are substantially rendered to be formed in the light. As in the hood (within the optical limitations of the reticle detection system used to create the image). For example, one of the reticle "substantially high resolution images" is by virtue of a substantially high resolution reticle detection system (eg, capable of a substantially high numerical aperture (eg, a value greater than 0.8) Aperture (NA) produces a smear detection system that images a solid reticle at the reticle plane to produce an image. In contrast, one of the images used to create a reticle "substantially low NA" can be less than one of the NAs of 0.5. In addition, the "substantially low NA" used to generate a mask image can be substantially the same as that on the mask side by an exposure system for projecting an image of the mask onto a wafer. The features on the cover are transferred to the NA on the wafer. Thus, in a substantially low NA image (or LNI), the reticle features can be rendered substantially differently than they are formed on the reticle. For example, in an LNI, the reticle features can appear to have a rounder corner than the corners formed on the reticle.

The terms "design" and "design data" as used herein generally refer to the physical design (layout) of an IC and the data derived from the physical design through complex simulations or simple geometry and Boolean operations. The design can be stored in a data repository such as a GDS file, any other standard machine readable file, any other suitable file known in the art, and a design database. A GDSII file is one of a class of files used to design a representation of layout data. Other examples of such files include GL1 and OASIS files. The designs used in the embodiments set forth herein can be stored in any of the files of this entire category, regardless of the data structure configuration, storage format, or storage mechanism. In addition, one of the pattern information of the reticle described in the text can be stored in one of the reticle patterns. In the library. The reticle pattern is a pattern that can be expected on a reticle when the mask writing operation is flawless. The database may contain vertex coordinates of a polygon or may be a grayscale image.

The design may include any of the other designs set forth in U.S. Patent No. 7,570,796 issued to the commonly owned Japanese Patent No. 7, 570, 796, issued toKalkarni et al. The information or design information agent, the two U.S. patents are hereby incorporated by reference in their entirety in their entirety. In addition, the terms "design" and "design data" as used herein are meant to be produced by a semiconductor device designer in a design process and thus may be well before the design is printed on any physical reticle and/or wafer. Information and materials used in the examples set forth herein.

One embodiment relates to a computer implemented method for setting up a reticle detection procedure. As shown in step 100 of Figure 1, the method includes separating the pattern information of a reticle into predetermined segments. The terms "part", "section" and "segment" of a reticle refer to a group of points in a two-dimensional (2D) plan view of the reticle. Although such terms may be defined to have the same meaning, each term is reserved in this document for a different concept.

In one embodiment, the predetermined segment has a diameter that is greater than an optical proximity of the reticle. For example, as explained above, the pattern information can be separated into relatively small segments or segments. Then, as further described herein, one of the segments of one of the masks being detected may be searched for among the predetermined segments. To increase the probability of finding a match, the predetermined segment can be selected as a relatively small segment of the database. However, the diameter of the predetermined segment should be greater than the optical proximity distance to ensure that the predetermined segment contains sufficient pattern information so that a match can be found with reasonable confidence.

In one embodiment, pattern information is obtained from the design data of the reticle. The design information for the reticle can be stored in a database such as that described above. In some such embodiments, the method can include acquiring a database of one of the patterns of the reticle. The database can be obtained from any suitable storage medium in any suitable manner.

In another embodiment, the method includes basing based on one of the patterns of the reticle The number of instances of the predetermined segment is selected based on the pattern information. For example, using a database, a portion of the reticle that represents a pattern on the rest of the reticle or another reticle can be selected.

As shown in step 102 of FIG. 1, the method also includes determining the near field data of the predetermined segment based on one or more actual images of the reticle obtained by one of the reticle detection systems. For example, the method can include electronically acquiring one or more optical microscope images of selected portions of the reticle. A detector and reticle detection system that acquires one or more actual images can be configured as further described herein. These images can be optical LNI images of the reticle. One or more actual images of the reticle will be used to determine near field data. In this way, the near field can be restored from the optical LNI image. The near field of a reticle can generally be defined as an electric field near any plane of the reticle. As such, the embodiments set forth herein can include restoring a mask near field based on images acquired by a microscope.

As further described herein, the near field data will then be used for the reticle or another reticle to simulate the reticle or one of the other reticle detector images, the detector image will be used for Defect detection of the reticle or the other reticle. Thus, the near field data may be based on one or more of the reticle that it is determined by the detector of the reticle detection system (which will be used to detect the reticle or the other reticle) or A similarly configured configuration of the reticle detection system (eg, one of the different reticle detection systems that is the same as the reticle detection system pattern and model to be used for detection) is obtained by a similarly configured detector. In other words, as further described herein, since the near field data will be used to simulate the detector image, the actual image available for determining near field data is preferably obtained under the same optical conditions as will be used for detection. In this way, the interaction between the reticle and the illumination electromagnetic waves can be measured as directly as possible using an actual reticle or another reticle. However, if the optical conditions for mask detection cannot be used to obtain the actual image used to determine the near field, the actual image can be modified in some way to simulate the actual acquisition of the mask for the mask with the detected optical conditions. image. The modified images can then be used to determine near field data.

If a reticle other than the reticle to be detected is used to obtain an image from which near field data is determined, the other reticle should have a characteristic substantially similar to the reticle to be detected. For example, the reticle and the other reticle are preferably constructed of substantially identical materials having substantially the same thickness and composition. Additionally, the two reticles may have been formed using the same procedure. The two masks may not necessarily be printed with the same pattern, as long as the pattern on the mask can be divided into substantially identical segments (eg, lines having similar widths, etc.). In addition, the reticle to be detected and the reticle for acquiring images may be a reticle and the same reticle.

The near field data of the predetermined segment can be determined by regression based on one or more actual images of the reticle. For example, the near field of a selected portion of the reticle can be restored (regressed) from the optical image it acquires or the intensity of the image recorded at a detector plane. In particular, the near field system of a reticle is restored from its intensity image to a reverse problem or a regression problem. As in most inverse problems, the near field can be reverted back and forth by minimizing a cost function (eg, energy or penalty function). The amount that is minimized may be the sum of the squared differences between one of the optical images at the detector and one of the intensity images from the near field. Additionally, any other suitable regression method and/or algorithm can be used to determine near field data for one or more actual images.

In one embodiment, one or more actual images of the reticle or another reticle are acquired by the detector or another detector with more than one focus setting. In this manner, in a preferred embodiment, images can be acquired with more than one focus setting (i.e., multiple focus settings). If more than one focus setting is used, the restoration of the near field is robust. A system of a reticle may be acquired using one or more of the focus settings of one of the detectors as further described herein.

In one embodiment, the regression comprises a Hopkins phase approximation. In another embodiment, the regression does not include a thin mask approximation. In an additional embodiment, the regression does not include the Krishhof approximation. For example, the near field of the reticle is an electromagnetic field present at the surface of the reticle when the reticle is illuminated by a substantially incident plane wave. In lithography and inspection, a reticle is illuminated by plane waves incident from many directions. According to the Hopkin approximation, when incident When the direction is changed, the direction of the diffraction order changes, and its amplitude and phase remain substantially unchanged. Embodiments set forth herein may use the Hopkin phase approximation, but do not perform a so-called thin mask or Krishhof approximation.

As shown in step 104 of FIG. 1, the method further includes generating a data structure including the paired predetermined segments and the near field data corresponding to the predetermined segments. In some embodiments, the data structure can be configured as a library. In this manner, the method can include forming a library of one of the near fields on a segment of the paired database and the corresponding segment of the reticle. In other words, the library can contain relatively small clips of the reticle pattern and associated reticle near field. In this way, the mask pattern clip and the near field can be saved as a pair of objects in the library. However, the data structure can have any other suitable file format known in the art. Additionally, although the data structure may be referred to as a "library" in the description provided herein, it will be understood that the embodiments set forth herein do not require the configuration of the data structure as a library. It is therefore possible to generate a database before detecting a reticle (or a second reticle).

As shown in step 106 of FIG. 1, the method also includes setting the reticle detection program by incorporating the information of the generated data structure into a prescription for the reticle detecting program, such that the light is Comparing two or more segments in one of the reticle or another reticle to the predetermined segment during the hood detection procedure, and at least one of the two or more segments The near field material of one of the predetermined segments most similar to one of the two or more segments is assigned. In other words, when detecting a reticle, the pattern information can be separated into segments, compared with predetermined segments in the library, and the near field data of a predetermined segment in the library can be assigned to its matching segment. segment. These steps can be performed as further described herein.

In this manner, setting the reticle inspection program as set forth herein can include generating a library to be used in a reticle inspection program and at least making the library available for use in the reticle inspection program. However, the embodiments set forth herein may or may not include other parameters (eg, optical parameters, defect detection parameters, etc.) that provide a reticle inspection procedure. For example, Other parameters may be determined in some other way or by some other system, and information about the library (eg, one of the links to the library and/or one of the libraries' IDs) may be incorporated into the program. As such, the embodiments set forth herein may include setting only one element (library) of an entire reticle inspection program. The reticle inspection program can be any set of instructions that can be used by a reticle inspection system to perform the reticle inspection procedure. These instructions can have any suitable format and can be stored in any suitable data structure.

The one or more computer systems can be configured in accordance with any of the embodiments set forth herein by one or more computer systems performing such separation, determination, generation, and setup steps.

Each of the embodiments of the methods set forth above can include any other steps of any of the other methods set forth herein. Moreover, each of the embodiments of the methods set forth above can be performed by any of the systems set forth herein.

One of the computer implementation methods for setting up a reticle detection procedure includes additional separation, generation, and setup steps as set forth above. However, this embodiment does not necessarily include the determination steps set forth above. In contrast, in this embodiment, the method includes determining near field data for the predetermined segments by numerically solving an equation governing the electromagnetic field of the predetermined segment. In other words, the embodiments can use the Maxwell equation to determine the near field of different predetermined segments. The near field determined from the Maxwell equation and its corresponding predetermined segment can then be stored in a data structure (eg, a library), as further described herein. These near field and predetermined segment pairs can then be used during a reticle detection procedure, as further described herein. The equation (or Maxwell's equation) governing the predetermined segmentation of the electromagnetic field can be solved in any suitable manner using any suitable method and/or algorithm.

The one or more computer systems can be configured in accordance with any of the embodiments set forth herein by one or more computer systems performing such separation, determination, generation, and setup steps.

Each of the embodiments of the methods set forth above can include the methods set forth herein. Any other steps of any other method. Moreover, each of the embodiments of the methods set forth above can be performed by any of the systems set forth herein.

Another embodiment relates to a computer implemented method for detecting defects on a reticle. In other words, this embodiment can include detecting a reticle or performing a reticle inspection procedure on a reticle. The reticle inspection procedure performed in the embodiments set forth further herein may be provided (at least in part) as set forth above.

As shown in step 108 of Figure 1, the method includes separating one of the patterns contained in one of the detection zones on the reticle into two or more segments. For example, a detection area of a reticle is provided, and the database of the detection area can be divided or divided into relatively small segments. The method may or may not include selecting a detection zone on the reticle. For example, the detection zone on the reticle can be determined by another method or system, and the information of the detection zone can be included in a prescription for a reticle detection procedure. The two or more segments can be configured as further described herein.

As shown in step 110 of FIG. 1, the method also includes separately comparing the two or more segments with predetermined segments included in a data structure. The data structure includes a pair of predetermined segments of the reticle pattern and corresponding near field data. In this way, each segment of the database can be queried in the library. In particular, the separate comparison step can include finding a match in one of the sections of the database of the detected mask in the database segment of the library. In other words, this step can include finding a matching pattern in the library.

In one embodiment, the diameter of the two or more segments is greater than an optical proximity of the reticle. For example, as explained above, the reticle pattern can be separated into relatively small segments or segments. Then, one of the segments of the database can be searched for in the database segment in the library. To increase the probability of finding a match, one of the relatively small sections of the database can be processed at once. However, the diameter of the section should be greater than the optical proximity distance.

In one embodiment, the method includes determining the pair of predetermined segments and corresponding near field data by separating the pattern information of the reticle or another reticle into predetermined segments; Determining the near field data of the predetermined segments based on the reticle or the one or more actual images of the reticle or the other reticle obtained by the detector or another detector. These steps can be performed as further described herein. For example, a library for the embodiments set forth herein can be created as set forth above.

In one such embodiment, the near field data of the predetermined segments are determined by regression based on one or more actual images of the reticle or the other reticle. The near field data can be determined in this manner, as explained in further detail herein.

In an additional embodiment, the method includes: determining the paired predetermined segments and corresponding near field data by separating the pattern information of the reticle or another reticle into predetermined segments; and by The equations of the electromagnetic fields governing the predetermined segments of the reticle pattern are numerically solved to determine near field data for the predetermined segments. These steps can be performed as further described herein. For example, a library for the embodiments set forth herein can be created as set forth above.

In another embodiment, comparing the two or more segments separately with the predetermined segment comprises comparing grayscale images of the two or more segments with grayscale images of the predetermined segments. For example, finding a match in the library can involve evaluating the correlation of the grayscale rasterized image. In an additional embodiment, comparing two or more segments separately with the predetermined segment includes comparing vertex coordinates of the polygons in one of the two or more segments with the predetermined segments The vertex coordinates of the polygon, and one of the two or more segments and one of the vertex coordinates in the predetermined segments are not considered to be a difference during a separate comparison. For example, if the database contains vertex coordinates of a polygon, the vertex coordinates can be matched within an arbitrary translation. Any suitable method and/or algorithm can be used to perform these separate comparison steps in any suitable manner.

As shown in step 112 of FIG. 1, the method further includes assigning near field data to the two or two based on one of a predetermined segment that is most similar to at least one of the two or more segments. At least one of the above segments. Assign near field data to at least one The segmentation can include extracting, from the library, a near field corresponding to a predetermined segment that is most similar to the at least one segment. The captured near field can then be assigned to one of the points in the detection zone. Any suitable algorithm and/or method may be used in any suitable manner to determine the degree to which a segment of a mask being detected is similar to a predetermined segment in the library. In this way, the near field of the reticle can be calculated during the detection by querying it in the library.

In one embodiment, one of the predetermined segments that are most similar when at least one of the two or more segments is not the at least one of the two or more segments An exact match: assigning the near field material to the at least one of the two or more segments comprises: determining the one of the predetermined segments into the two or more segments Mapping at least one of the mappings, applying the mapping to the near field data corresponding to the one of the predetermined segments to generate modified near field data, and assigning the modified near field data to the At least one of two or more segments. In this way, if an exact match is not found, the closest match in the library can be selected. In one such embodiment, one of the morphological operations of mapping the library database segment to the database segment of the detection region can be found. The same morphological operations can then be applied to the near field of the library. For example, suppose a near field of a 54 nm wide line is found in a bright field, and the closest match in the library is a 50 nm wide line. The near field mapping of the 50 nm wide line can be selected and extended to cover the 54 nm wide line.

In another embodiment, the predetermined segment that is most similar to the at least one of the two or more segments cannot be found for at least another of the two or more segments In one of the methods, the method includes: determining near field data of the at least one of the two or more segments based on an actual image generated by the detector; and determining the near field data to be new Increased to the data structure. For example, if no match can be found, a near field restore can be performed as described herein and the results can be included in the library.

As shown in step 114 of Figure 1, the method includes producing based on the assigned near field data. The near field data of the detection zone is generated. For example, generating near field data for the detection zone can include making a collage of the captured near field such that a single near field value is assigned to each point of the detection zone. In this way, an image can be calculated from the assigned near field.

In one embodiment, the method also includes simulating an image of the detection zone based on the generated near field data to be formed by one of the photomask detection systems. For example, the assigned near field value can be propagated to the detector array using Hopkin or Gamo-Gabor theory and an image intensity can be determined.

As shown in step 116 of FIG. 1, the method further includes acquiring an actual image produced by the detector of the detection zone on one of the physical versions of the reticle. In this way, an optical image of one of the areas to be detected can be acquired. Any of the detectors of any of the systems described herein can be used to generate an actual image of one of the detection zones on a physical version of the reticle, as further described herein.

As shown in step 118 of FIG. 1, the method also includes detecting defects on the reticle by comparing the simulated image with the actual image. For example, the acquired optical image and the calculated image intensity can be compared, and a significant difference between the computed image of the detection zone and the acquired optical image can be detected and reported. However, any other suitable method and/or algorithm may be used in any other suitable manner to detect defects on the reticle based on comparing the results of the simulated image with the actual image.

Thus, the embodiments set forth herein have several advantages over other currently used reticle detection methods. For example, the near-field ratio of recovering a mask from its optical image as explained further herein is faster than solving the Maxwell equation. In addition, creating a library of restored near-field and tiling the near-field from the library reduces the number of near-field recovery calculations.

In yet another embodiment, the method includes simulating one of the patterned features to be formed on a wafer in a program executed by the reticle by inputting the generated near field data into a model Or a plurality of characteristics; the measurement has been formed on the wafer in the program One or more characteristics of the patterned features; comparing one or more simulated characteristics to one or more measured characteristics; and modifying one or more parameters of the model based on the results of the comparing step. In this way, a lithography imaging model can be applied to the mask near field to predict a lithographic image. As such, the method can include predicting one of a critical dimension or lithography on a wafer using the predicted lithography image. In addition, a lithographic calculation model can be calibrated using a recovered near-field and wafer image (eg, a wafer scanning electron microscope (SEM) image). In other words, the embodiments set forth herein can calibrate modeling parameters in a lithography procedure that is critical for predicting wafer pattern from a set of LNI images.

In some of these embodiments, the reticle whose image is used to determine near field data (which is input into a model) can be a calibrated mask. This calibration mask can contain a set of one-dimensional (1D) and 2D patterns. The mask can be the same mask used to create an OPC model. Alternatively, the mask can be designed by a supplier of reticle inspection tools such as KLA-Tencor (Milpitas, Calif).

A set of LNI images can be acquired for the calibration mask, as explained herein. The imaging conditions used to acquire such images are preferably such that the illumination source shape and the NA on the objective side are the same or nearly the same as the illumination source shape and NA on a scanner. LNI images can be out of focus for better adjustment of subsequent calculations. An alternative approach is to acquire a set of LNI images at near-coherent illumination conditions but at different angles of incidence.

The mask near field (or MNR) can then be restored as further described herein. For example, an LNI image with a known illumination and imaging condition can be reconstructed repeatedly by minimizing the difference between the acquired LNI image and the self-mask near-field computed image. Cover plane amplitude and phase information.

The lithography program can then be modeled. For example, the mask near field can be fed to a computational engine to produce wafer level images. The calculation engine preferably includes a scanner imaging procedure for the entire photoresist material (which explicitly considers polarization and vector imaging effects). The engine also preferably contains, for example, a photoresist development process comprising acid and substrate diffusion and substrate quenching. The calculation engine may also include etch modeling as appropriate.

The simulated wafer image can then be compared to an experimentally acquired SEM image of a wafer having the same calibration mask printed thereon. The model parameters can be repetitively adjusted using any difference between the simulated wafer image and the experimentally acquired image to thereby reduce image differences (eg, such that the simulated wafer image is substantially matched and experimentally acquired) image). This repeated procedure implements the extraction of the lithography model parameters. The embodiments set forth herein thus provide a calibration process to extract one of the lithography model parameters. Additionally, the embodiments set forth herein combine a mask near field with a full lithography model and a calibration mask to extract lithography model parameters. These model parameters can then be stored and later retrieved when a defect is found.

Embodiments for calibrating a lithography model as set forth herein provide several advantages over the currently used methods and systems for calibrating lithography models. For example, the current method used to calibrate lithography models is based on inaccurate methods of solving Maxwell's equations. In addition, some currently used methods assume that a geometric database represents the actual pattern on the reticle. In addition, some currently used methods assume that a pre-selected three-dimensional (3D) contour represents the actual contour etched in the reticle. In addition, some currently used methods assume that a set of material parameters represent the materials that make up the reticle. Additional shortcomings of the methods and systems currently in use are set forth in the background section of this document.

Defects detected by reticle inspection are difficult to handle in terms of printability due to increased OPC complexity. The simple resist threshold model is not sufficient for LNI images or aerial images. A resist model calibration on top of a scanned imaging model is required to accurately predict the wafer level impact of a defect before actually printing a wafer using a physical scanner. As such, it is highly desirable to construct new methods that can be used to substantially accurately predict wafer-level behavior of a defect prior to shipping a reticle to a wafer fabrication facility.

In one such embodiment, the method includes determining, based on at least one of the defects detected on the reticle, based on an actual image of the at least one of the defects Simulating near field data of at least one of the defects; and simulating how the at least one of the defects is at least one of the defects by inputting the determined near field data of the at least one of the defects into the model The program is printed on the wafer. In this manner, the method can include substantially accurate wafer level defect printability of the image detected by the reticle. In other words, embodiments can be used to analyze reticle defect printability. For example, once a defect is detected on a reticle, a set of LNI images can be acquired at or near the defect, and the image can be used to restore the mask near field of the defect (which can be as herein) Executed as described), and lithographic modeling (by modeling parameters extracted through the embodiments set forth herein) can be performed to accurately calculate defect behavior on the wafer. In addition, once one of the models has been calibrated as described herein to obtain an image of how at least one defect will be printed on a wafer, the simulated image can be compared and how the mask region is intended to be printed. Information on the wafer (which can be obtained from one of the reticle pre-OPC databases). As such, a pre-OPC database can be used to calculate the effect of reticle defect printing on a wafer. Thus, by comparing the calculated wafer level image with a pre-OPC database, it is determined how the reticle defect will change the reticle print on the wafer from its intended print to predict defect behavior.

Embodiments set forth herein can be used to identify the reticle and accurately predict wafer level effects or behavior of defects on the reticle prior to shipping the reticle to the manufacturing facility and prior to performing extensive fabrication of the reticle. Subsequent defect handling can be performed to separate a fatal defect from a non-fatal defect. For example, once the effect of a defect on the printing of the reticle on the wafer has been determined, the defect can be disposed of or classified (eg, classified as a non-printing defect, a printing defect, a fatal printing defect). (eg, defects that will affect yield), a non-fatal print defect (eg, will print but do not affect one of the yield defects, etc.).

Thus, the defect printability embodiments set forth herein have several advantages over other methods and systems for predicting defect printability. For example, the embodiments set forth herein can be performed using a lithography model that includes scanning imaging characteristics and a photoresist development program. In addition, mask near-field restoration and lithography model calibration (for example, combined detection images) The combination with the lithography procedure step to calibrate the lithography model produces a more accurate wafer level prediction from the detected image. Thus, the embodiments set forth herein provide calibration results that can be used to substantially accurately predict wafer-level behavior of a defect for printability analysis.

The calibration steps set forth herein can be implemented on existing reticle inspection tools. In addition, defect handling by means of a calibrated model can be implemented on existing reticle inspection tools.

The one or more computer systems can be configured in accordance with any of the embodiments set forth herein by performing separate, separate comparison, assignment, generation, simulation, acquisition, and detection steps with one or more computer systems .

Each of the embodiments of the methods set forth above can include any other steps of any of the other methods set forth herein. Moreover, each of the embodiments of the methods set forth above can be performed by any of the systems set forth herein.

All of the methods set forth herein can include storing the results of one or more of the method embodiments in a computer readable storage medium. These results can include any of the results set forth herein and can be stored in any manner known in the art. The storage medium may include any of the storage media set forth herein or any other suitable storage medium known in the art. After the results have been stored, the results can be accessed in a storage medium and used by any of the methods or system embodiments set forth herein, formatted for display to a user, by another Software modules, methods or systems, etc.

Another embodiment is directed to a non-transitory computer readable medium storing program instructions executable on a computer system to perform one or more of the computer implemented methods set forth herein. One such embodiment is shown in FIG. For example, as shown in FIG. 2, non-transitory computer readable medium 200 stores program instructions 202 executable on computer system 204 to perform one or more of the computer implemented methods set forth herein. The computer implemented method can include any of the steps of any of the methods set forth herein.

Program instructions 202 that implement methods such as those described herein may be stored in a non- On a temporary computer readable medium 200. The computer readable medium can be a storage medium such as a magnetic or optical disk, a magnetic tape, or any other suitable non-transitory computer readable medium known in the art.

Program instructions may be implemented in any of a variety of ways, including processor-based techniques, component-based techniques, and/or object-oriented techniques, among other techniques. For example, the program instructions can be implemented using Matlab, Visual Pedestal, ActiveX Controls, C, C++ Objects, C#, JavaBeans, Microsoft Foundation Classes ("MFC"), or other techniques or methods as needed.

Computer system 204 can take a variety of forms, including a personal computer system, a host computer system, a workstation, a system computer, an imaging computer, a programmable image computer, a parallel processor, or any other device known in the art. In general, the term "computer system" is broadly defined to encompass any device having one or more processors that execute instructions from a memory medium.

In one embodiment, the computer system set forth above may comprise one or more computer subsystems that are part of an electronic design automation (EDA) tool (not shown), and the reticle detection described further herein. The system is not part of the EDA tool. The EDA tool and computer subsystem included in such a tool can contain any commercially available EDA tool that can be configured to perform one or more of the steps set forth above. For example, a computer subsystem configured to separate segments of a pattern contained in a detection zone on a reticle can be part of an EDA tool. In this manner, a reticle design material can be processed by the EDA tool to separate the pattern into segments that will be used by another different system or tool as set forth herein.

The computer subsystem may also be part of an EDA tool and may be included in another system or tool or simply configured as a stand-alone computer system. Additionally, the tool or computer subsystem (which separates a pattern into segments, generates a data structure, and/or sets a reticle inspection program as described herein) can be configured to take steps by The generated information is stored or transmitted to a computer readable storage medium such as a manufacturing database or by using the information Directly transfer to the tool that will use the information to provide information to another tool or computer subsystem.

Another embodiment relates to a system configured to detect defects on a reticle. An embodiment of such a system is shown in FIG. The system includes a reticle detection subsystem that includes a detector configured to generate an actual image of one of the detection zones on a physical version of the reticle. The actual image can be any of the images described further herein. As shown in FIG. 3, the system includes a reticle detection subsystem 300.

As further shown in FIG. 3, the reticle detection subsystem 300 includes one of the illumination subsystem and a collection subsystem as set forth in greater detail herein. The illumination subsystem includes a light source 302. Light source 302 can be a coherent light source, such as a laser. The light source can be configured to emit monochromatic light having a wavelength of about 248 nm, about 193 nm, about 157 nm, or another ultraviolet wavelength. Alternatively, the light source can be configured to emit light having a range of wavelengths and can be coupled to a spectral filter (not shown). An example of a broadband source includes, but is not limited to, a He-Xe arc lamp that produces light in a deep ultraviolet wavelength pattern. In this manner, the light source and the filter can emit monochromatic light having a wavelength as set forth above. The light source and the filter can be configured such that different wavelengths can be emitted from the light source and the filter depending on, for example, the type of reticle being detected or the type of detection or measurement being performed Light. The light source can also be configured to emit light other than ultraviolet light. Additionally, the light source can be configured to emit light at various time intervals, either continuously or in pulses.

The illumination subsystem can also include a number of optical components coupled to the light source. For example, light from source 302 may first pass through homogenizer 304. The homogenizer 304 can be configured to reduce the blob of light from the source. The illumination subsystem can also include an aperture 306. The aperture 306 can have an adjustable NA. For example, the aperture can be coupled to a control mechanism that can be configured to mechanically depend on receiving a control signal from a user or a program instruction received from a program prescription running on the system Change the aperture. In this way, the light can have various partial coherence factors σ. For example, aperture 306 can be modified to adjust the aperture of one of concentrating lenses 308. The aperture of the concentrating lens controls the NA of the system. Since the pupil of the concentrator is reduced, the illumination coherence is increased, thereby reducing the σ value. The σ value can be expressed as the ratio of the NA of the concentrating lens to the NA of the objective lens. The exposure system can have a sigma value in a range between about 0.3 to about 0.9. Thus, aperture 306 can be modified such that the optical subsystem has a sigma value between about 0.3 and about 0.9. The σ value can be changed depending on the characteristics on the reticle. For example, if the mask contains lines and spaces rather than if the mask contains contact holes, a higher value of σ can be used. The control mechanism can also be configured to change the aperture to provide circular or off-axis illumination. The aperture can also be configured to provide other types of illumination such as quadrupole or bipolar illumination. The aperture can be further configured to change the shape of one of the beams. For example, the aperture can be a diffractive optical element or a toe aperture.

The illumination subsystem can also include a number of additional optical components (not shown). For example, the illumination subsystem can also include a telescope configured to modify one of the beam diameters of the light. Additionally, the illumination subsystem can include one or more relay lenses, additional lenses such as a field lens, a folding mirror, an additional aperture, and a beam splitter.

The illumination subsystem can also include a concentrating lens 308. Condenser lens 308 can be configured to change the diameter of one of the lights in the plane of the object (mask) to be approximately or greater than the field of view of the system. Light exiting the concentrating lens illuminates the reticle 310 supported on the stage 312. The stage is configured to support the reticle by contacting the reticle adjacent a lateral edge of the reticle. An opening in the stage is provided to allow light illumination from the illumination subsystem to illuminate the reticle. The stage 312 can be configured to move the reticle such that one of the reticle can be aligned such that light can be scanned across the reticle. Alternatively, the illumination subsystem can include a scanning element (not shown) such as an acousto-optic deflector or a mechanical scanning assembly such that the reticle can remain substantially stable while allowing light to be scanned across the reticle. The stage 312 can also be configured to move the reticle through the focus, thereby changing one of the focus settings of the optical subsystem. The stage can also be coupled to an autofocus device (not shown) configured to change a position of the stage, thereby changing a position of the mask to One of the focus settings of the optical subsystem is maintained during a test. Alternatively, an autofocus device can be coupled to the objective to modify one of the objectives to maintain focus settings during a test.

The optical subsystem can also include a number of optical components configured to form a collection subsystem. For example, the collection subsystem includes an objective lens 314. The light transmitted by the reticle is collected by the objective lens 314. The collection subsystem also includes an aperture 316 having an adjustable NA. The NA of the aperture 316 can also be selected such that the light exiting the aperture has a selected magnification. The aperture 316 is positioned between the objective lens 314 and a lens 318 that can be configured as a barrel lens. Light from lens 318 can be directed to beam splitter 320. Beam splitter 320 can be configured to direct light to three detectors 322a, 322b, and 322c. The collection subsystem can also include a number of additional optical components (not shown), such as a magnifying lens. The magnifying lens can be positioned between the lens 318 and the beam splitter.

The detectors 322a, 322b, and 322c can be configured to form an image of light transmitted by the illuminated portion of one of the reticle. This image can be called an "air image." The detectors should also be sensitive to at least one of the wavelengths of light set forth above. However, in addition to wavelengths in other types, the detectors may be sensitive to certain wavelength ranges in the deep ultraviolet mode. For example, the detectors can include a charge coupled device (CCD) or a time delay integrated (TDI) camera. The detectors can also have a one or two dimensional array of pixels. Each of the three detectors can have a different focus setting. In this way, the three detectors can form an image of the reticle at substantially the same time with three different focus settings. For example, one detector can be substantially in focus, and the other two detectors can be out of focus in the opposite direction to the focus misalignment condition. Additionally, the reticle detection subsystem can include any number of such detectors depending on the mechanical or physical constraints of the reticle detection subsystem.

Alternatively, the reticle detection subsystem can include only one detector configured to produce an actual image of the reticle. The detector can have an approximation equal to an exposure system One focus sets one of the focus settings. The image of the reticle at different focus settings can be formed by forming a plurality of images of the reticle and changing the focus setting of the detector after forming each image. In this embodiment, the beam splitter 320 will not have to split the light into multiple detectors.

The reticle detection subsystem can include several other components not shown in FIG. For example, the system can include a load module, an alignment module, a processor such as a robotic transfer arm, and an environmental control module and can include any such components known in the art.

As explained above, the reticle detection subsystem can be configured to form an aerial image of the reticle using a set of exposure conditions. Such exposure conditions include, but are not limited to, illumination wavelength, illumination coherence, shape of the illumination beam, NA, and focus settings. The set of exposure conditions can be selected to be substantially equivalent to exposure conditions used by an exposure system to image one of the reticle onto a wafer. Thus, an aerial image formed by reticle inspection subsystem 300 can be substantially optically equivalent to an image of a reticle that will be printed on a wafer by the exposure system under the set of exposure conditions.

The system also includes one or more computer subsystems configured to perform one or more of one or more of the computer implemented methods set forth herein. In one embodiment, as shown in FIG. 3, the system includes a computer subsystem 324. In the embodiment shown in FIG. 3, computer subsystem 324 is coupled to reticle detection subsystem 300. For example, the computer subsystem can be coupled to one of the reticle detection subsystem detectors. In one such example, as shown in FIG. 3, computer subsystem 324 is coupled to detectors 322a, 322b, and 322c of reticle detection subsystem 300 (eg, by one of the dashed lines shown in FIG. 3 or A plurality of transmission media, which can include any suitable transmission medium known in the art. The computer subsystem can be coupled to the detectors in any suitable manner. The computer subsystem can be coupled to the reticle detection subsystem in any other suitable manner such that an image of the reticle produced by the reticle detection subsystem and any other information can be sent to the computer subsystem and, as appropriate, the computer subsystem Instructions can be sent to the reticle detection subsystem to perform one or more of the steps set forth herein Step. The computer subsystems included in the system can also be configured as set forth herein.

It should be noted that FIG. 3 is provided herein to generally illustrate one configuration of a reticle detection subsystem that may be included in the system embodiments set forth herein. Obviously, the configuration of the reticle detection subsystem set forth herein can be modified to optimize the performance of the system, as is typically done when designing a commercial inspection system. Additionally, the systems set forth herein can be implemented using an existing reticle detection subsystem (e.g., by adding the functionality set forth herein to an existing inspection system), such as commercially available from KLA-Tencor. Photomask inspection tool. For some such systems, the methods set forth herein may be provided for the optional functionality of the system (eg, in addition to other functionality of the system). Another option is to design the reticle inspection system described in this article "from scratch" to provide a completely new system.

Further modifications and alternative embodiments of the various aspects of the invention will be apparent to those skilled in the art. For example, systems and methods are provided for detecting defects on a reticle. Accordingly, the description is to be construed as illustrative only, and is intended to be a It is to be understood that the form of the invention as illustrated and described herein is considered to be a preferred embodiment. It will be apparent to those skilled in the art that <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; These features can be utilized independently. Variations in the elements set forth herein may be made without departing from the spirit and scope of the invention as set forth in the appended claims.

Claims (41)

A computer implementation method for detecting defects on a reticle, comprising: separating one of a detection regions included in the reticle into two or more segments; comparing the two separately Or more than two segments and a predetermined segment included in a data structure, wherein the data structure includes the pair of predetermined segments of the reticle pattern and the corresponding near field data; based on the two or more Assigning, by one of the predetermined segments, the most recent of the segments, the near field data to the at least one of the two or more segments; generating based on the assigned near field data Near field data of the detection zone; simulating the detection zone based on the generated near field data to form an image by one of the photomask detection systems; and acquiring one of the photomasks generated by the detector An actual image of the detection zone on the physical version; and detecting defects on the reticle by comparing the simulated image with the actual image, wherein the separation is performed by one or more computer systems, separately compared, Assign, generate, simulate, obtain And detection. The method of claim 1, further comprising: determining the pair of predetermined segments and the corresponding near field data by separating the pattern information of the reticle or another reticle into the predetermined segments And determining the near field data of the predetermined segments based on the reticle or the one or more actual images of the reticle or the other reticle obtained by the detector or another detector. The method of claim 2, wherein the pattern information is obtained from the design data of the reticle. The method of claim 2, further comprising one of the patterns based on the mask The number of instances of the predetermined segments in the volume selects the predetermined segments based on the pattern information. The method of claim 2, wherein the one or more actual images of the reticle or the other reticle are acquired by the detector or the other detector with one or more focus settings. The method of claim 2, wherein the one or more actual images based on the reticle or the other reticle are determined by regression to determine the near field data of the predetermined segments. The method of claim 6, wherein the regression comprises a Hopkin phase approximation. The method of claim 6, wherein the regression does not include a thin mask approximation. The method of claim 6, wherein the regression does not include a Krishhof approximation. The method of claim 1, further comprising: determining the pair of predetermined segments and the corresponding near field data by separating the pattern information of the reticle or another reticle into the predetermined segments And determining the near field data of the predetermined segments by numerically solving an equation for controlling the electromagnetic fields of the predetermined segments of the reticle pattern. The method of claim 10, wherein the pattern information is obtained from the design data of the reticle. The method of claim 10, further comprising selecting the predetermined segments based on the pattern information based on the number of instances of the predetermined segments in the entirety of the pattern of the reticle. The method of claim 1, wherein the diameter of the two or more segments is greater than an optical proximity of the reticle. The method of claim 1, wherein the individual comparison comprises comparing grayscale images of the two or more segments with grayscale images of the predetermined segments. The method of claim 1, wherein the individual comparison comprises: comparing a vertex coordinate of a polygon in one of the two or more segments with a vertex coordinate of a polygon in the predetermined segment, and wherein Do not compare the two or two during the individual comparison period One of the above segments and one of the coordinates of the vertices in the predetermined segments is considered a difference. The method of claim 1, wherein the one of the predetermined segments that are most similar to the at least one of the two or more segments is not in the two or more segments Assigning the near field material to the at least one of the two or more segments includes determining the one of the predetermined segments to the two of the at least one of the exact matches Mapping one or more of the at least one of the two or more segments; applying the mapping to the near field data corresponding to the one of the predetermined segments to generate modified near field data; The modified near field data is assigned to the at least one of the two or more segments. The method of claim 1, wherein the at least one of the two or more segments is not found to be most similar to the at least one of the two or more segments In one of the segments, the method further comprises: determining near field data of the at least one of the two or more segments based on the actual image generated by the detector; The near field data is determined to be added to the data structure. The method of claim 1, further comprising: simulating a patterned feature to be formed on a wafer in a program executed by the reticle by inputting the generated near field data into a model One or more characteristics; measuring the one or more characteristics of the patterned features that have been formed on the wafer in the program; comparing the one or more simulated characteristics to the one or more Measuring characteristics; and changing one or more parameters of the model based on the results of the comparison. The method of claim 18, further comprising: determining, for the at least one of the defects detected on the reticle, the defects based on the actual image of the at least one of the defects Near field data of the at least one; and inputting the determined near field data of the at least one of the defects into the model And simulating how at least one of the defects will be printed on the wafer in the program. A computer implemented method for setting a mask detection program, comprising: separating a pattern information of a mask into a predetermined segment; based on one of the masks obtained by a detector of a mask detection system or Determining the near field data of the predetermined segments by the plurality of actual images, wherein the one or more actual images based on the reticle determine the near field data of the predetermined segments by regression; generating includes pairwise And a predetermined one of the predetermined segments and the data structure of the near field data corresponding to the predetermined segments; and the information is obtained by incorporating the information of the generated data structure into a prescription for the reticle detecting program The reticle detecting program is configured to compare two or more segments and one of the predetermined segments in the reticle or another reticle during the reticle detecting procedure, and the two And the near field data of one of the predetermined segments that is most similar to at least one of the two or more segments is assigned to the at least one of the two or more segments, wherein One or more computer systems perform such separation, determination, and production And setup steps. A computer implementation method for setting a mask detection program, comprising: separating pattern information of a mask into predetermined segments; determining such values by numerically solving an equation for controlling an electromagnetic field of the predetermined segments Pre-scheduled near-field data; generating a data structure comprising the pair of predetermined segments and the near-field data corresponding to the predetermined segments; and incorporating information from the generated data structure The reticle detecting program is provided for use in one of the reticle detecting procedures such that two or more of the detection areas of the reticle or the other reticle are compared during the reticle detecting procedure Assigning the near field data of the segment to the predetermined segment and one of the predetermined segments most similar to at least one of the two or more segments to the two or two The at least one of the above segments, wherein the separating, determining, generating, and setting steps are performed by one or more computer systems. A non-transitory computer readable medium storing program instructions executable on a computer system to perform a method for detecting a defect in a reticle, wherein the computer implementation method comprises: One of the patterns in one of the masks is separated into two or more segments; the two or more segments are individually compared with a predetermined segment included in a data structure, wherein the data structure includes a pair of predetermined segments of the reticle pattern and corresponding near field data; one of the predetermined segments that are most similar based on at least one of the two or more segments will have near field data Assigning to the at least one of the two or more segments; generating near field data of the detection region based on the assigned near field data; simulating the detection region based on the generated near field data One of the reticle detection systems forms an image; acquires an actual image of the detection zone on a physical version of the reticle generated by the detector; and by comparing the simulated image with the actual Image detection Cover the defect. A system configured to detect defects on a reticle, comprising: a reticle detection subsystem configured to generate an actual image of one of the detection zones on a physical version of the reticle a detector; and one or more computer subsystems configured to: Separating one of the patterns included in the detection zone into two or more segments; separately comparing the two or more segments with a predetermined segment included in a data structure, wherein the data structure includes a a pair of predetermined segments of the reticle pattern and corresponding near field data; assigning near field data based on one of the predetermined segments most similar to at least one of the two or more segments And at least one of the two or more segments; generating near field data of the detection region based on the assigned near field data; simulating the detection region based on the generated near field data by the detection The detector forms an image; the actual image is acquired from the detector; and the defect on the mask is detected by comparing the simulated image with the actual image. The system of claim 23, wherein the one or more computer subsystems are further configured to: determine the pairwise by separating the pattern information of the reticle or another reticle into the predetermined segments Determining the predetermined segments and the corresponding near field data; and determining, by the detector or another detector, the predetermined segments based on the one or more actual images of the mask or the other mask The near field information. The system of claim 24, wherein the pattern information is obtained from design information of the reticle. The system of claim 24, wherein the one or more computer subsystems are further configured to select the number based on the number of instances of the predetermined segments in the entirety of the pattern of the reticle Wait for the scheduled segment. The system of claim 24, wherein the one or more actual shadows of the reticle or the other reticle are acquired by the detector or the other detector with one or more focus settings image. The system of claim 24, wherein the one or more actual images based on the reticle or the other reticle are determined by regression to determine the near field data of the predetermined segments. The system of claim 28, wherein the regression comprises a Hopkin phase approximation. The system of claim 28, wherein the regression does not include a thin mask approximation. The system of claim 28, wherein the regression does not include a Krishhof approximation. The system of claim 23, wherein the one or more computer subsystems are further configured to: determine the pairwise by separating the pattern information of the reticle or another reticle into the predetermined segments The predetermined segments and the corresponding near field data; and determining the near field data of the predetermined segments by numerically solving an equation for controlling the electromagnetic fields of the predetermined segments of the reticle pattern. The system of claim 32, wherein the pattern information is obtained from design information of the reticle. The system of claim 32, wherein the one or more computer subsystems are further configured to select the number based on the number of instances of the predetermined segments in the entirety of the pattern of the reticle Wait for the scheduled segment. The system of claim 23, wherein the diameter of the two or more segments is greater than an optical proximity of the reticle. The system of claim 23, wherein the individual comparison comprises comparing grayscale images of the two or more segments with grayscale images of the predetermined segments. The system of claim 23, wherein the individual comparison comprises: comparing one of the vertex coordinates of the polygons in the two or more segments to one of the vertex coordinates of the polygons in the predetermined segments. The system of claim 23, wherein the one of the predetermined segments that are most similar to the at least one of the two or more segments is not in the two or more segments Assigning the near field data when one of the at least one matches exactly The at least one of the two or more segments includes: determining a mapping of the one of the predetermined segments to the one of the at least one of the two or more segments, The mapping is applied to the near field data corresponding to the one of the predetermined segments to generate modified near field data; and assigning the modified near field data to the two or more segments The at least one. A system of claim 23, wherein the at least one of the two or more segments is not found to be most similar to the at least one of the two or more segments In one of the segments, the one or more computer subsystems are further configured to: determine the at least one of the two or more segments based on the actual image generated by the detector The near field data of the person; and adding the determined near field data to the data structure. The system of claim 23, wherein the one or more computer subsystems are further configured to: by inputting the generated near field data into a model, the simulation will be formed in a program executed by the reticle Measuring one or more characteristics of the patterned features on a wafer; measuring the one or more characteristics of the patterned features that have been formed on the wafer in the program; comparing the one or more The simulated characteristic and the one or more measured characteristics; and modifying one or more parameters of the model based on the result of the comparison. The system of claim 40, wherein the one or more computer subsystems are further configured to target at least one of the defects detected on the reticle: based on the at least one of the defects Determining the near field data of the at least one of the defects by the actual image; and simulating the defects by inputting the determined near field data of the at least one of the defects into the model How at least one of the at least one will be printed on the wafer in the program.