Print inspection repeater defect detection

Abstract

Systems and methods for detecting defects on a magnification reticle are provided. A system is configured for generating different stacked difference images for multiple instances of a first patterned region in different rows on a wafer based on images generated for the first patterned region in the different rows. The system is also configured for performing dual detection based on the different stacked difference images. Next, the system identifies defects on the magnification reticle based on defects detected by the dual detection. As further described herein, the systems and methods detect defects from multiple rows of reticles printed on a wafer, which may reduce noise and enable detection of substantially small repeater defects. Embodiments are particularly useful for high sensitivity repeater defect detection of extreme ultraviolet (EUV) reticles and multi-die reticles (MDR).

Description

Technical Field

[0001] The present invention generally relates to systems and methods for repeater defect detection on a print inspection wafer to detect defects on reticles such as extreme ultraviolet (EUV) single die and multi-die reticles. Background Art

[0002] The following descriptions and examples are not admitted to be prior art by virtue of their inclusion in this section.

[0003] Inspection processes are used at various steps during the semiconductor manufacturing process to detect defects on the wafer to promote higher yields and therefore higher profits in manufacturing. Some current inspection methods detect repeater defects on the wafer to thereby detect defects on the reticle. For example, if a defect (a "repeater defect") is repeatedly detected at multiple locations on the wafer that correspond to the same location on the reticle, then the defect may be caused by the reticle itself. Therefore, a repeater defect may be analyzed to determine whether it is caused by a reticle defect rather than some other cause.

[0004] Generally, Repeater Defect Detection (RDD) is performed as a wafer post-processing (PP) operation. For example, an inspection tool may perform normal die-to-die defect detection (DD) in a multi-die shrink reticle case, and after all wafer defects are reported, RDD may be performed in a post-processing step. A repeater defect is defined in this document as a defect that is located at the same location (within a specific tolerance) in several shrink reticles.

[0005] However, there are several disadvantages to the currently used RDD methods and systems. For example, many currently used RDD methods and systems have been produced in an attempt to reduce noise so that repeater defects can be detected more accurately, with fewer detected disturbing points, and with higher sensitivity. However, many attempts at wafer noise reduction are unable to detect the smaller-sized EUV repeater defects required by next-generation design rules. For example, some currently used RDD methods use a set wafer for steps such as image alignment. The set wafer and the test wafer may have relatively small differences caused by, for example, variations in process conditions. Those differences in the wafer may be sufficient to cause errors in image alignment, which may thereby cause noise in defect detection results. Then, the noise may actually cause errors in defect detection results, such as missed defects, disturbing points, and noise being detected as defects and the like.

[0006]

[0006] It would therefore be advantageous to develop systems and methods for detecting repeater defects on a wafer, and thereby defects on a shrink reticle, that do not have one or more of the above-mentioned disadvantages. Summary of the invention

[0007] The following description of various embodiments should in no way be construed as limiting the subject matter of the appended claims.

[0008] One embodiment relates to a system configured to detect defects on a reticle. The system includes an inspection subsystem configured to generate an image of a wafer. The reticle is used to print patterned areas on the wafer during a lithography process. The patterned areas include first patterned areas that correspond to the same area on the reticle and are printed in different rows on the wafer. The system also includes a computer subsystem configured to generate different stacked difference images of multiple instances of the first patterned areas in the different rows based on the images generated by the inspection system for the first patterned areas in the different rows. The computer subsystem is also configured to detect defect candidates in the multiple instances of the first patterned areas on the wafer by applying a defect detection method to the different stacked difference images. The defect detection method determines that the defect candidate exists at the location in each of the multiple instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images. The computer subsystem is further configured to identify defects on the reticle based on the detected defect candidates. The system may be further configured as described herein.

[0009] Another embodiment is directed to a computer-implemented method for detecting defects on a reticle. The method includes acquiring an image generated by an inspection subsystem for a wafer. The reticle is used to print patterned areas on the wafer during a photolithography process. The patterned areas include a first patterned area corresponding to the same area on the reticle and printed in a different row on the wafer. The method also includes the generating, detecting, and identifying steps described above, the steps being performed by a computer subsystem coupled to the inspection subsystem.

[0010] Each step of the method may be further performed as further described herein. The method may include any other steps of any other method described herein. The method may be performed by any of the systems described herein.

[0011] Another embodiment relates to a non-transitory computer-readable medium storing program instructions executable on a computer system for performing a computer-implemented method for detecting defects on a retracted reticle. The computer-implemented method includes the steps of the method described above. The computer-readable medium may be further configured as described herein. The steps of the computer-implemented method may be executed as further described herein. Additionally, the computer-implemented method for which the program instructions may be executed may include any other steps of any other method described herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Further advantages of the present invention will become apparent to those skilled in the art having the benefit of the following detailed description of the preferred embodiments and after reference to the accompanying drawings, in which:

[0013] Figure 1 and 2 is a schematic diagram illustrating a side view of an embodiment of a system configured as described herein;

[0014] Figure 3 is a schematic diagram illustrating a plan view of one example of a wafer having patterned areas printed thereon using a reticle and one embodiment of an interleaved scan of different rows on the wafer performed by an inspection subsystem;

[0015] Figures 4 to 7 is a flow chart illustrating an embodiment of steps of generating different stacked difference images and detecting defect candidates that may be performed for detecting defects on a reticle;

[0016] Figure 8 is a flow chart illustrating an embodiment of identifying defects on a reticle based on defect candidates detected on a wafer;

[0017] Fig. 9 is a schematic diagram illustrating a plan view of a test image generated for three rows of patterned areas printed on a wafer that may be processed together to detect defects on a shrunken reticle;

[0018] Fig.10 This illustrates that three rows of patterned areas can be processed together. Figure 4 A flowchart of an embodiment of the steps of generating different stacked difference images and detecting defect candidates for detecting defects on a reticle shown in FIG.

[0019] Fig.11 is a flow chart illustrating an embodiment of identifying defects on a reticle based on defect candidates detected on a wafer that may be performed when three rows of patterned areas are processed together; and

[0020] Fig.12is a block diagram illustrating one embodiment of a non-transitory computer-readable medium storing program instructions that execute on a computer system for performing one or more of the computer-implemented methods described herein.

[0021] Although the present invention is susceptible to various modifications and alternative forms, specific embodiments of the invention are shown by way of example in the drawings and described in detail herein. The drawings may not be drawn to scale. However, it should be understood that the drawings and detailed description thereof are not intended to limit the invention to the specific forms disclosed, but on the contrary, are intended to cover all modifications, equivalents and alternatives as falling within the spirit and scope of the invention as defined by the claims. DETAILED DESCRIPTION

[0022] Referring now to the drawings, it should be noted that the drawings are not drawn to scale. In particular, the proportions of some elements of the drawings are greatly exaggerated to emphasize the characteristics of the elements. It should also be noted that the drawings are not drawn to the same scale. The same element symbols have been used to indicate elements shown in more than one figure that may be similarly configured. Unless otherwise mentioned herein, any of the elements described and shown may include any suitable commercially available elements.

[0023] One embodiment relates to a system configured to detect defects on a retracted reticle. Embodiments described herein provide new and improved methods and systems for high sensitivity repeater defect detection (RDD) for detecting mask defects by inspecting wafers printed with the mask. The terms "retracted reticle" and "mask" are used interchangeably herein. The term "repeater" is used interchangeably herein with the term "repeater defect".

[0024] exist Figure 1 One embodiment of such a system is shown in . The system includes an inspection subsystem configured to generate an image of a wafer. The inspection subsystem described herein may include at least an energy source and a detector. The energy source is configured to generate energy directed to a wafer. The detector is configured to detect energy from the wafer and generate an output (e.g., an image) in response to the detected energy.

[0025] In one embodiment, the energy directed to the wafer comprises light, and the energy detected from the wafer comprises light. Figure 1 In the embodiment of the system shown in FIG. 1 , the inspection subsystem 10 includes an illumination subsystem configured to direct light to the sample 14. The illumination subsystem includes at least one light source. For example, as shown in FIG. Figure 1 , the illumination subsystem includes a light source 16. In one embodiment, the illumination subsystem is configured to direct light to the wafer at one or more incident angles, which may include one or more oblique angles and / or one or more normal angles. Figure 1, light from light source 16 is directed through optical element 18 and then through lens 20 to beam splitter 21, which directs the light at a normal angle of incidence to wafer 14. The angle of incidence may include any suitable angle of incidence, which may vary depending on, for example, the characteristics of the wafer and the defects to be detected on the wafer.

[0026] The illumination subsystem may be configured to direct light to the wafer at different incident angles at different times. For example, the inspection subsystem may be configured to modify one or more characteristics of one or more elements of the illumination subsystem so that light is directed at different incident angles. Figure 1 In one such example, the inspection subsystem may be configured to move light source 16, optical element 18, and lens 20 so that light is directed to the wafer at different angles of incidence.

[0027] In some examples, the inspection system may be configured to direct light to the wafer at more than one incident angle at the same time. For example, the illumination subsystem may include more than one illumination channel, one of which may include a plurality of illumination channels, such as in FIG. Figure 1 1 , and the other of the illumination channels (not shown) may include similar elements (which may be differently or identically configured), or may include at least one light source and possibly one or more other elements (such as those described further herein). If this light is directed to the wafer simultaneously with the other light, one or more characteristics (e.g., wavelength, polarization, etc.) of the light directed to the wafer at different angles of incidence may be different, so that the light originating from illuminating the wafer at different angles of incidence may be distinguished from each other at a detector.

[0028] In another example, the illumination subsystem may include only one light source (e.g., Figure 1 ) and the light from the light source may be separated into different optical paths (e.g., based on wavelength, polarization, etc.) by one or more optical elements (not shown) of the illumination subsystem. The light in each of the different optical paths may then be directed to the wafer. Multiple illumination channels may be configured to direct light to the wafer at the same time or at different times (e.g., when different illumination channels are used to sequentially illuminate the wafer). In another example, the same illumination channel may be configured to direct light with different characteristics to the wafer at different times. For example, in some examples, the optical element 18 may be configured as a spectral filter and the properties of the spectral filter may be changed in a variety of different ways (e.g., by swapping spectral filters) so that light of different wavelengths may be directed to the wafer at different times. The illumination subsystem may have any other suitable configuration known in the art for directing light with different or the same characteristics to the wafer sequentially or simultaneously at different or the same incident angles.

[0029] In one embodiment, the light source 16 may include a broadband plasma (BBP) light source. In this way, the light generated by the light source and directed to the wafer may include broadband light. However, the light source may include any other suitable light source, such as a laser. The laser may include any suitable laser known in the art and may be configured to generate light of any suitable wavelength or wavelengths known in the art. In addition, the laser may be configured to generate monochromatic or near-monochromatic light. In this way, the laser may be a narrow-band laser. The light source may also include a polychromatic light source that generates light of multiple discrete wavelengths or bands.

[0030] Light from the optical element 18 can be focused by the lens 20 to the beam splitter 21. Figure 1 2 is shown as a single refractive optical element, but in reality, lens 20 may include several refractive and / or reflective optical elements that combine to focus light from the optical element onto the wafer. Figure 1 The illumination subsystem shown in and described herein may include any other suitable optical elements (not shown). Examples of such optical elements include, but are not limited to, polarizing elements, spectral filters, spatial filters, reflective optical elements, apodizers, beam splitters, apertures, and the like, which may include any such suitable optical elements known in the art. Additionally, the system may be configured to alter one or more elements of the illumination subsystem based on the type of illumination to be used for the inspection.

[0031] The inspection subsystem also includes a scanning subsystem that is configured to scan light across the wafer. For example, the inspection subsystem may include a stage 22 on which the wafer 14 is disposed during inspection. The scanning subsystem may include any suitable mechanical and / or robotic assembly (including the stage 22) that may be configured to move the wafer so that light may be scanned across the wafer. Additionally, or alternatively, the inspection subsystem may be configured so that one or more optical elements of the inspection subsystem perform some kind of scanning of light across the wafer. The light may be scanned across the wafer in any suitable manner.

[0032] The inspection subsystem further includes one or more detection channels. At least one of the one or more detection channels includes a detector configured to detect light from the wafer as a result of illuminating the sample through the inspection subsystem and to generate an output in response to the detected light. For example, in Figure 1 The inspection subsystem shown in FIG. 1 includes two detection channels: one channel is formed by a light collector 24, an element 26, and a detector 28, and the other channel is formed by a light collector 30, an element 32, and a detector 34. Figure 1, two detection channels are configured to collect and detect light at different collection angles. In some examples, one detection channel is configured to detect specularly reflected light and another detection channel is configured to detect light that is not specularly reflected from the wafer (e.g., scattered, diffracted, etc.). However, two or more detection channels may be configured to detect the same type of light from the wafer (e.g., specularly reflected light). Although Figure 1 An embodiment of an inspection subsystem including two detection channels is shown, but the inspection subsystem may include a different number of detection channels (e.g., only one detection channel or two or more detection channels). Figure 1 , but each of the light collectors may include one or more refractive optical elements and / or one or more reflective optical elements.

[0033] One or more detection channels may include any suitable detector known in the art, such as a photomultiplier tube (PMT), a charge coupled device (CCD), and a time delay integration (TDI) camera. The detector may also include a non-imaging detector or an imaging detector. If the detector is a non-imaging detector, each of the detectors may be configured to detect a specific characteristic of the scattered light, such as intensity, but may not be configured to detect such characteristics that vary as a function of position within the imaging plane. Thus, the output generated by each detector in each detection channel included in the inspection subsystem may be a signal or data, rather than an image signal or image data. In such examples, the computer subsystem further described herein may be configured to generate an image of the wafer from the non-imaging output of the detector. In other examples, the detector may be configured as an imaging detector, which is configured to generate an image signal or image data. Thus, the system may be configured to generate the images described herein in a number of ways.

[0034] It should be noted that in this article Figure 1 The configuration of the inspection subsystem that may be included in the system embodiments described herein is generally illustrated. Obviously, the inspection subsystem configuration described herein may be modified to optimize the performance of the system as is typically performed when designing a commercial inspection system. In addition, the system described herein may be implemented using an existing inspection system (e.g., by adding the functionality described herein to the existing inspection system) (e.g., the 29xx / 39xx series of tools available from KLA, Inc., Milpitas, California). For some such systems, the methods described herein may be provided as optional functionality of the system (e.g., in addition to other functionality of the system). Alternatively, the system described herein may be designed "from scratch" to provide an entirely new system.

[0035] The computer subsystem 100 of the system may be coupled to the detector of the inspection subsystem in any suitable manner (e.g., via one or more transmission media, which may include "wired" and / or "wireless" transmission media) so that the computer subsystem may receive the output generated by the detector during wafer scanning. The computer subsystem 100 may be configured to perform a number of functions using the output of the detector as described herein and any other functions further described herein. The computer subsystem 100 may be further configured as described herein.

[0036] Computer subsystem 100 (and other computer subsystems described herein) may also be referred to herein as a computer system. Each of the computer subsystems or systems described herein may take various forms, including personal computer systems, image computers, mainframe computer systems, workstations, network devices, Internet devices, or other devices adapted according to the various embodiments described herein. In general, the term "computer system" may be broadly defined to encompass any device having one or more processors that execute instructions from a memory medium. A computer subsystem or system may also include any suitable processor known in the art, such as a parallel processor. In addition, a computer subsystem or system may include a computer platform (as a stand-alone tool or a network tool) with high-speed processing and SW.

[0037] If the system includes more than one computer subsystem, the computer subsystems may be coupled to each other so that images, data, information, instructions, etc. can be sent between the various computer subsystems. For example, the different computer subsystems may be coupled to each other via any suitable transmission media, which may include any suitable wired and / or wireless transmission media known in the art. Two or more computer subsystems may also be effectively coupled via a shared computer-readable and writable storage media (not shown).

[0038] Although the inspection subsystem is described above as an optical or light-based inspection subsystem, the inspection subsystem may be an electron beam-based inspection subsystem. For example, in one embodiment, the energy directed to the wafer includes electrons, and the energy detected from the wafer includes electrons. In this manner, the energy source may be an electron beam source. Figure 2 In one such embodiment shown in , the inspection subsystem includes an electronic column 122 coupled to a computer subsystem 200 .

[0039] Likewise Figure 2, the electron column includes an electron beam source 126 configured to generate electrons that are focused to a sample 128 by one or more elements 130. The electron beam source may include, for example, a cathode source or emitter tip, and the one or more elements 130 may include, for example, a gun lens, an anode, a beam limiting aperture, a gate valve, a beam current selection aperture, an objective lens, and a scanning subsystem, all of which may include any such suitable elements known in the art.

[0040] Electrons returning from the sample (eg, secondary electrons) may be focused to a detector 134 by one or more elements 132. One or more elements 132 may include, for example, a scanning subsystem, which may be the same scanning subsystem included in element 130.

[0041] The electron column may include any other suitable elements known in the art. In addition, the electron column may be further configured as described in U.S. Pat. No. 8,664,594 issued to Jiang et al. on April 4, 2014, U.S. Pat. No. 8,692,204 issued to Kojima et al. on April 8, 2014, U.S. Pat. No. 8,698,093 issued to Gubbens et al. on April 15, 2014, and U.S. Pat. No. 8,716,662 issued to MacDonald et al. on May 6, 2014, which are incorporated herein by reference as if fully set forth.

[0042] Although the electron column Figure 2 1 is shown as being configured so that electrons are directed to the wafer at an oblique angle of incidence and scattered from the wafer at another oblique angle, but the electron beam may be directed to the wafer and scattered from the wafer at any suitable angle. In addition, the electron beam-based subsystem may be configured to generate an image of the wafer using multiple modes (e.g., employing different illumination angles, collection angles, etc.). The multiple modes of the electron beam-based subsystem may differ in any image generation parameter of the subsystem.

[0043] Computer subsystem 200 may be coupled to detector 134 as described above. The detector may detect electrons returning from the surface of the wafer, thereby forming an electron beam image of the wafer. The electron beam image may include any suitable electron beam image. Computer subsystem 300 may be configured to use the output of the detector and / or the electron beam image to perform any of the functions described herein. Computer subsystem 200 may be configured to perform any additional steps described herein. Figure 2 The system of inspection subsystems shown in may be further configured as described herein.

[0044] It should be noted that this article provides Figure 2to generally illustrate configurations of electron-beam-based inspection subsystems that may be included in the embodiments described herein. As with the optical inspection subsystems described above, the configurations of the electron-beam-based inspection subsystems described herein may be modified to optimize the performance of the inspection subsystems as is typically performed when designing commercial inspection systems. Additionally, the systems described herein may be implemented using existing inspection systems (e.g., by adding the functionality described herein to the existing inspection systems). For some such systems, the methods described herein may be provided as optional functionality of the system (e.g., in addition to other functionality of the system). Alternatively, the systems described herein may be designed "from the ground up" to provide an entirely new system.

[0045] Although the inspection subsystem is described above as a light-based or electron-beam-based inspection subsystem, the inspection subsystem may be an ion-beam-based inspection subsystem. Figure 2 This inspection subsystem is configured as shown in FIG. 1 , except that any suitable ion beam source known in the art may be used in place of the electron beam source. Additionally, the inspection subsystem may be any other suitable ion beam-based subsystem, such as inspection subsystems included in commercially available focused ion beam (FIB) systems, helium ion microscope (HIM) systems, and secondary ion mass spectrometer (SIMS) systems.

[0046] As mentioned above, the inspection subsystem may be configured to direct energy (e.g., light, electrons) to and / or scan energy (e.g., light, electrons) across a physical version of the wafer, thereby producing an actual image of the physical version of the wafer. In this manner, the inspection subsystem may be configured as a "real" imaging system rather than a "virtual" system. Figure 1 The storage medium (not shown) and computer subsystem 100 shown in the figure may be configured as a "virtual" system. Systems and methods configured as a "virtual" inspection system are described in commonly assigned U.S. Patent Nos. 8,126,255, issued on February 28, 2012 to Bhaskar et al. and 9,222,895, issued on December 29, 2015 to Duffy et al., both of which are incorporated herein by reference as if fully set forth. The embodiments described herein may be further configured as described in these patents.

[0047] As further mentioned above, the inspection subsystem may be configured to have multiple modes. In general, a "mode" is defined by the value of a parameter of the inspection subsystem used to produce an output of the sample. Thus, different modes may differ in the value of at least one of the optical or electron beam parameters of the inspection subsystem (in addition to the location of the output on the sample that produces the image). For example, for a light-based inspection subsystem, different modes may use light of different wavelengths. The modes may differ in the wavelength of light directed to the sample as further described herein for different modes (e.g., by using different light sources, different spectral filters, etc.). In another embodiment, different modes may use different illumination channels. For example, as mentioned above, the inspection subsystem may include more than one illumination channel. Thus, different illumination channels may be used for different modes.

[0048] The multiple modes may also differ in illumination and / or collection / detection. For example, as further described above, the inspection subsystem may include multiple detectors. Thus, one of the detectors may be used for one mode and another of the detectors may be used for another mode. Furthermore, the modes may differ from each other in more than one of the ways described herein (e.g., different modes may have one or more different illumination parameters and one or more different detection parameters). For example, depending on the ability to scan a sample using multiple modes simultaneously, the inspection subsystem may be configured to scan a sample with different modes in the same scan or in different scans.

[0049] A reticle is used to print patterned areas on a wafer during a lithography process. The lithography process may include any suitable lithography process. In general, a lithography process uses some form of energy (e.g., light, electrons, etc.) to transfer a pattern formed on the reticle to a material on the wafer. The material on the wafer may be some form of photoresist, one or more properties of which change after exposure to the energy used in the lithography process. In this way, a pattern of energy (light, electrons, etc.) may be directed from or through the reticle to the photoresist on the wafer to thereby transfer the pattern from the reticle to the photoresist. One or more additional steps (e.g., post-exposure baking, development, etc.) may be performed on the photoresist to complete the pattern transfer process. Therefore, as is clear from this general description of the lithography process, during the lithography process, defects present on the reticle may be transferred to the wafer, which is clearly disadvantageous. Sometimes, such as during reticle inspection, it is impossible or impractical to inspect the reticle directly. In such instances, the reticle is printed on a wafer, the wafer is inspected for defects, and the wafer defects are used to determine whether the defects were caused by the reticle. Such a determination may be performed as further described herein.

[0050] In one embodiment, the reticle is a single die reticle (SDR). In another embodiment, the reticle is a multi-die reticle (MDR). For example, the embodiments described herein can be used for both SDR inspection where the mask contains only one die and MDR inspection where a mask contains several identical dies.

[0051] In a further embodiment, the retractor is configured for use at an extreme ultraviolet (EUV) wavelength. Such a mask may then be referred to as an EUV mask. In other words, the mask may be configured for use in a lithography process that uses light at one or more EUV wavelengths to transfer a pattern from the mask to a wafer (e.g., by reflecting EUV light from the mask to the wafer). In this way, the actinic wavelength of the mask (i.e., the wavelength of light used to transfer the pattern from the mask to the wafer to cause a photochemical reaction in one or more materials (e.g., photoresist) on the wafer) may be an EUV wavelength. Although the embodiments described herein are particularly useful for defect detection on EUV masks, the embodiments may also be used for defect detection on other types of SDRs and MDRs, such as masks designed to be used at one or more deep ultraviolet (DUV) wavelengths of light (i.e., single die or multi-die, optical or DUV masks).

[0052] Embodiments described herein provide substantially high sensitivity repeater defect detection (RDD) for EUV print inspection. "Print inspection" is defined herein as a type of mask inspection that involves printing a mask on a wafer, inspecting the wafer, and identifying defects on the mask based on the wafer inspection results. Print inspection can be used for a variety of applications that can be performed to routinely or periodically inspect EUV masks going into production for possible contamination, including identifying reticle repeaters for EUV mask qualification and / or monitoring. Print inspection can be used for both SDR and MDR. In the embodiments described herein, reference images from multiple reticle rows are used to generate difference images, and the difference images from or across reticle rows are stacked for improved noise reduction, which enables detection of smaller size EUV MDR repeater defects on print inspection wafers.

[0053] The patterned area includes a first patterned area that corresponds to the same area on the reticle and is printed in different rows on the wafer. In this way, the "rows" described herein are "reticle rows" printed on the wafer, and each "reticle row" is defined as a row of multiple instances of the entire reticle printed on the wafer. For a single-die reticle, each row of dies printed on the wafer may correspond to a row of reticle instances printed on the wafer. However, more than one row of dies printed on the wafer may correspond to a single row of reticle instances printed on the wafer with an MDR. For example, when an MDR including multiple rows of dies is printed on the wafer at one time, the one reticle instance will include multiple rows of dies. In this way, when this reticle is printed multiple times in a single row across the wafer, the row will include multiple rows of dies. For the embodiments described herein, different rows refer to different reticle rows that may or may not be different rows of dies.

[0054] The image of the first patterned area produced by the inspection subsystem is mapped to different instances of the same area on the reticle printed on the wafer. Depending on whether the reticle is an SDR or an MDR, what constitutes different instances of the same area of ​​the reticle printed on the wafer may change. For example, for an SDR, different instances of the same area on the reticle printed on the wafer may be corresponding areas in different dies printed separately on the wafer using the reticle. In the case of an MDR, different instances of the same area on the reticle printed on the wafer may be corresponding areas in different instances of the reticle printed on the wafer. In such an example, the corresponding areas of the MDR may be multiple instances of the same area of ​​only one die on the MDR.

[0055] In one embodiment, the inspection subsystem is configured to generate an image by interleaving a scan of a first patterned area in a different row on the wafer with a scan of a second patterned area in a different row on the wafer. Figure 3 One embodiment of a staggered reticle-based scanning (i.e., scanning) strategy that can be used with the embodiments described herein is described. The staggered reticle-based scanning strategy scans (i.e., scans) the same Y position of a printed reticle instance on a wafer with scanning strips across multiple reticle rows, such as Figure 3 For example, Figure 3, a wafer 300 may have a reticle instance 302 formed thereon. In the case of MDR, each reticle instance contains a plurality of die 304 that may be configured as further described herein. In this embodiment, the inspection subsystem may scan S1, S2, S3, S4, S5, and S6 in sequence. S1, S3, and S5 all correspond to the same Y position of the reticle instance in each of the different rows. Similarly, S2, S4, and S6 all correspond to the same Y position of the reticle instance in each of the different rows, which is different from the Y position of S1, S3, and S5. (Of course, there may be some overlap in Y between S1 and S2, S3 and S4, and S5 and S6, but S1 and S2, S3 and S4, and S5 and S6 still have different Y positions). This reticle-based interleaved scan-band scanning strategy is different from the die-based interleaved scan-band scanning strategy currently used. The staggered swath scanning strategy based on the reticle ensures that reference images across reticle rows are available for computational reference (CR) image generation and difference image stacking as further described herein.

[0056] As described above, Figure 3 , the system scans the swaths in the following order: S1, S2, S3, S4, S5, and S6 (of course more than 6 swaths are possible). The system can then inspect S1 and S3 simultaneously as further described herein and inspect S2 and S4 simultaneously as further described herein. In other words, the swaths generated at different Y positions relative to the reticle area are processed separately for defect detection. In this way, defect detection is performed for S1, S3, and S5 independently of the defect detection performed for S2, S4, and S6. Once the first set of two reticle rows are completed, the system can then proceed to inspection involving the third row (e.g., inspect S1 and S3, then inspect S3 and S5). However, as further described herein, more than two (e.g., 3) reticle rows can also be inspected together (e.g., inspect S1, S3, and S5 simultaneously). For the embodiments described herein, at least two reticle rows are scanned and then processed simultaneously. In this manner, scanning of a first set of scanning swaths that are processed together is interleaved with scanning of other scanning swaths that are processed together and independently of the first set of scanning swaths.

[0057] Although this article describes and Figure 3 The interleaved swath scanning strategy shown in is particularly advantageous for the embodiments described herein, but the systems and methods described herein may be used with any other suitable type of swath scanning strategy known in the art. An appropriate swath scanning strategy may be selected based on a number of system layout factors, such as available memory for temporarily storing images of swaths to be processed together for defect detection.

[0058] When any two or more images generated at different locations on the wafer are used together for some purpose, some type of image alignment should be performed to ensure that the images are properly aligned with each other (e.g., for image subtraction, for CR generation, etc.). One particularly advantageous way in which the embodiments described herein may perform this image alignment is to use an alignment die positioned in the first reticle row that is scanned in a scan band to align other reticle rows. For example, the image generated in the first die scanned in scan band S1 by the inspection subsystem may be used as an alignment die for aligning the images generated in the scan band with each other and with the images generated in the other scan bands. Alignment using this image as an alignment die may be performed in other ways in any suitable manner known in the art. While this is a particularly useful method for aligning images between different rows or scan bands, the embodiments may also be used with any other suitable alignment methods known in the art.

[0059] In one such embodiment, the first patterned area and the second of the patterned areas correspond to different portions of the die on the retraction reticle. For example, when the dimension of the scan swath across the wafer in the Y direction is smaller than the dimension of the die printed on the wafer in the Y direction, more than one scan swath may be scanned across the wafer to cover the entire die. In this manner, different patterned areas in different rows on the wafer (e.g., different patterned areas in S1 and S2) corresponding to different portions of the same die on the retraction reticle may have different patterned features formed therein. Thus, in the embodiments described herein, such patterned areas will be processed separately for defect detection.

[0060] In another such embodiment, the first patterned area and the second of the patterned areas correspond to different instances of a die on a reticle printed on a wafer. For example, when the dimension of the scan band across the wafer in the Y direction is equal to or greater than the dimension of the die printed on the wafer in the Y direction, one scan band may be scanned across the wafer to cover the entire die. In this way, different patterned areas in different rows on the wafer (e.g., different patterned areas in S1 and S2) may correspond to different dies on the wafer. In some such examples, different dies on a wafer covered by different scan bands may be printed in the same reticle row (e.g., as when a single printing of a reticle on a wafer results in multiple rows of dies, and each scan band covers the rows of those dies). In other such examples, different dies on a wafer covered by different scan bands may be printed in different reticle rows (e.g., as when a single printing of a reticle on a wafer results in a single row of dies, and each scan band covers the rows of those dies).

[0061] In some embodiments, the shrink mask is an MDR, and the first patterned area includes at least a portion of only the first die of the MDR printed on the wafer. For example, the first patterned area may correspond to only one die of the MDR, and each first patterned area is a different instance of the MDR printed on the wafer. In this case, each of those first patterned areas may correspond to the entire die printed on the wafer or less than all of the die printed on the wafer. In this way, in a row of repeated patterned areas printed on the wafer with an MDR, the first patterned area may correspond to the first die of the MDR printed in the row, the second patterned area may correspond to the second die of the MDR printed in the row, and so on.

[0062] The system also includes a computer subsystem configured to generate different stacked difference images of multiple instances of the first patterned areas in different rows based on images generated by the inspection subsystem for the first patterned areas in different rows. The embodiments described herein provide new methods and systems for CR image generation and difference image stacking. In the embodiments described herein, reference images from multiple reticle rows are used to generate difference images and / or stack difference images from and / or across reticle rows. Several ways of performing this step are described herein, but all of the different ways have in common that the images used to generate inputs to the defect candidate detection step include images generated from patterned areas in different rows formed on the wafer. These new ways of generating inputs for the defect candidate detection step provide several important improvements to the embodiments described herein over currently used systems and methods for reticle defect detection. For example, the four new methods further described herein provide advantages such as maximum noise reduction and maximum upset point suppression.

[0063] In contrast to the embodiments described herein, some currently used methods and systems for print inspection use reference images from the same reticle row and stack difference images within the reticle row to achieve noise reduction. However, this wafer noise reduction is not sufficient to detect the smaller size EUV repeater defects required by next generation design rules. Some currently used methods for print inspection type reticle inspection require training that stores full die images from a setup wafer and uses them to align with production wafers. Since the setup wafer and production wafers may be different for a variety of reasons (e.g., processed with different process conditions due to process drift), the aligned die images from the setup wafer may have color variations compared to the production wafer images, and any image misalignment caused by this may introduce false defects at run time.

[0064] In one embodiment, generating a first of the different stacked difference images ("Stacked Difference 1") includes generating first and second difference images and combining at least a portion of the first and second difference images to generate Stacked Difference 1, and generating a second of the different stacked difference images ("Stacked Difference 2") includes generating third and fourth difference images and combining at least a portion of the third and fourth difference images to generate Stacked Difference 2. This step is common to the first three methods described herein.

[0065] Figure 4 The first method ("Type 1") shown in FIG. 1 uses reference images from 2 rows to generate a CR image for each test image and stacks the resulting difference images across 2 rows for maximum noise reduction. In this embodiment and other embodiments described herein, A1 and A2 refer to the reference images printed on the wafer ( Figure 4 , not shown, that correspond to the same area on the reticle and are printed in different rows on the wafer. Similarly, B1 and B2 (and C1 and C2) refer to different examples of the patterned area printed on the wafer with the reticle during the lithography process, that correspond to the same area on the reticle and are printed in different rows on the wafer. In this way, A1 and A2 correspond to different examples of the same portion of a die printed on the wafer with an MDR, B1 and B2 correspond to different examples of the same portion of different dies on the wafer printed with the MDR, and C1 and C2 correspond to different examples of the same portion of another different die on the wafer printed with the MDR. In this way, a test image 400 of A1, A2, B1, B2, C1, and C2 can be generated by scanning two different rows on the wafer. For example, in Figure 3 In S1 shown in FIG. 1 , test images A1, B1, and C1 can be generated, and a single MDR example printed on a wafer can be generated. Figure 4 Each group A1, B1 and C1 in the test image 400 shown in FIG. Figure 3 In S3 of FIG. 1 , test images of A2, B2, and C2 can be generated, and a single MDR example printed on a wafer can be generated. Figure 4 Each group A2, B2 and C2 in the test image 400 shown in FIG. Figure 3 In S5 shown in FIG. 1 , test images A3, B3 and C3 can be generated (eg Fig. 9 ), and each set of A3, B3 and C3 can be generated for a single MDR instance printed on a wafer.

[0066] Figure 4Each of A1, A2, B1, B2, C1, and C2 shown in other figures described herein may include only a portion of the die in the Y direction and possibly the entire die in the X direction. In other words, a scan strip may be generated by scanning Figure 4 and each of A1, A2, B1, B2, C1, and C2 shown in other figures described herein, the scanning strip has a height in Y that is less than the height of the die and covers the entire width of the die in the X direction for the height of the scanning strip. Alternatively, Figure 4 Each of A1 , A2 , B1 , B2 , C1 , and C2 shown in other figures described herein may include all dies in both the X direction and the Y direction.

[0067] In one such embodiment, generating the first difference image includes separately subtracting the first CR from images generated for multiple instances of the first patterned area in a first one of different rows on the wafer, and generating the second difference image includes separately subtracting the first CR from images generated for multiple instances of the first patterned area in a second one of different rows on the wafer. Figure 4 , generating a first difference image (e.g., top row of set 402 for difference 1) includes subtracting the first CR alone from the image generated for A1 and generating a second difference image (e.g., bottom row of set 402) includes subtracting the first CR alone from the image generated for A2, as shown in step 404. All image subtractions described herein may be performed in any manner known in the art.

[0068] In some such embodiments, the computer subsystem is configured to generate the first CR from images of multiple instances of the second one of the patterned regions in different rows on the wafer. Figure 4 , the first CR subtracted from A1 in step 404 may be generated as CR(B1, B2). This same CR may be subtracted from A2 alone, as shown in step 404. This CR and all other CRs described herein may be generated in any suitable manner known in the art.

[0069] In one embodiment, CR(B1, B2) indicates that the CR is calculated from an image of each instance of B1 and B2 scanned in two rows. However, the B1 and B2 images used to generate this CR may include images of less than all of the B1 and B2 patterned areas scanned in two rows. For example, in some examples, this CR may be generated from a single B1 image and a single B2 image, but the quality of the CR will generally increase as the number of reference images used to generate the CR increases. Therefore, it may be desirable, but not necessary, to use all available reference images to generate the CR. The same may apply to all other CRs described herein.

[0070] In additional embodiments, generating the third difference image includes separately subtracting the second CR from images generated for multiple instances of the first patterned area in a first one of different rows on the wafer and generating the fourth difference image includes separately subtracting the second CR from images generated for multiple instances of the first patterned area in a second one of different rows on the wafer. Figure 4 , generating a third difference image (e.g., the top row of set 406 for Difference 2) includes subtracting the second CR alone from the image generated for A1 in step 408 and generating a fourth difference image (e.g., the bottom row of set 406) includes subtracting the second CR alone from the image generated for A2, as shown in step 408.

[0071] In another embodiment, the computer subsystem is configured to generate a second CR from images of multiple instances of a third party in a patterned area in a different row on the wafer. Figure 4 As shown in , the second CR subtracted from A1 in step 408 may be generated as CR(C1, C2). This same CR may be subtracted from A2 alone, as shown in step 408.

[0072] like Figure 4 , the computer subsystem may generate stacked difference 1 410 by combining at least a portion of the first and second difference images (i.e., at least some difference 1 in two rows in set 402) to generate stacked difference 1 and generate stacked difference 2 412 by combining at least a portion of the third and fourth difference images (i.e., at least some difference 2 in two rows in set 406) to generate stacked difference 2. Generating stacked difference 1 may be performed using all difference 1 generated by A1-CR(B1,B2), A2-CR(B1,B2), and generating stacked difference 2 may be performed using all difference 2 generated by A1-CR(C1,C2), A2-CR(C1,C2). In other words, stacked difference 1 410 may be generated from all difference 1 in set 402, and stacked difference 2 412 may be generated from all difference 2 in set 406. Thus, in this method, each of the stacked difference images is generated from a different image generated with a CR generated across rows. In other words, the stacked difference image is produced from images from more than one row on the wafer.This multiple row source of images to produce a single stacked difference image has a number of advantages further described herein.

[0073] Although it may be advantageous to use as many difference images as possible (i.e., all of the difference images) to generate the stacked difference image, less than all of the difference images may be used to generate the stacked difference image when appropriate (e.g., when one or more of the difference images clearly contain some outlier characteristics). Combining at least a portion of the plurality of difference images to generate these and any other stacked difference images described herein may be performed in other ways in any suitable manner known in the art. For example, any suitable mathematical operation (e.g., a linear combination) may be used to combine the plurality of difference images into the stacked difference image.

[0074] The computer subsystem is also configured to detect defect candidates in multiple instances of the first patterned area on the wafer by applying the defect detection method to different stacked difference images. The defect detection method determines that the defect candidate exists at the location in each of the multiple instances of the first patterned area only when the defect detection method detects the defect candidate at the location in the different stacked difference images. For example, the computer subsystem may be configured to perform dual detection 414 using stacked difference 1 410 and stacked difference 2 412. Dual detection means comparing a test image with two different references to generate two different difference images, performing defect detection on the two difference images, and then performing arbitration so that only defects detected at the same location by two defect detections are reported in the defect detection results.

[0075] In the embodiments described herein, two different difference images have been generated prior to the dual detection step 414. Thus, in step 414, each of the stacked difference images may be individually compared to a certain defect detection threshold. Any signal or pixel in each stacked difference image having a value above the threshold may be determined as a defect candidate (conversely, a signal or pixel having a value less than the threshold may not be identified as a potential defect). The dual detection may then compare the defect candidates detected in the two stacked difference images and report a defect candidate only if a defect is detected at the same location in the two stacked difference images. Obviously, this may be the simplest method that can be used for the dual detection step, and more complex or obscure methods may be used. The stacked difference image generated in one of the novel approaches described herein may be input and used for defect detection in the same manner as any other difference image known in the art. In other words, the novel stacked difference images described herein are not specific to any kind of dual detection.

[0076] The output of the dual detection step may be a defect detection result 416, where the location of any defect candidate detected by dual detection is identified in each of the patterned areas with some kind of mark (in this case, an asterisk). In this way, the output of the dual detection step is a full correlation of the defect candidates reported, without individual frame information. In other words, because dual detection is performed on the stacked difference images, the computer subsystem cannot determine whether each individual test image contains a defect candidate based solely on the results of the detection step. Therefore, when the computer subsystem detects a defect candidate at a location via dual detection, the computer subsystem will assign a defect candidate (or mark the location or something equivalent) at the corresponding location in each test image. The additional steps described herein may then be performed to determine whether a defect is detected at the location of the defect candidate in each test image location.

[0077] although Figure 4 (and other figures described herein) show that only one defect candidate is detected by the dual detection step, but it is apparent that more than one defect candidate may be detected by the dual detection step. In this case, the result 416 may contain some label for each detected defect candidate in each test image. Each detected defect candidate may then be evaluated individually or separately to determine which test images actually contain defects at the location of the detected defect candidate. Additionally, although the results of the dual detection step are Figure 4 (and other figures described herein), the results of the dual detection steps may be reported in any suitable file and format known in the art.

[0078] Thus, as further described above, the Type 1 method uses images generated from different reticle rows on a wafer in new and important ways. For example, the embodiments described above perform reference image generation using images from different reticle rows on a wafer (e.g., CR(B1, B2)). Additionally, the embodiments described herein perform difference image stacking by stacking difference images across 2 reticle rows (e.g., generating a stacked difference 1 from a difference 1 image generated from an A1 image and a difference 1 image generated from an A2 image). Additionally, dual detection is performed based on a comparison between a test patterned area and two other patterned areas in a reticle row (e.g., AB and AC).

[0079] The type 1 method described above has significant advantages for interference point reduction. For example, the type 1 described above will advantageously start with fewer repeater types than the currently used MDR RDD method. In particular, interference point repeater types that do not have signals in more than one row that can be detected by the currently used MDR RDD method will be suppressed by the type 1 method described above. When fewer repeater types are detected, there are fewer interference point repeaters to be filtered. In addition, for some use cases, type 1 may have the highest noise reduction capability and the best original sensitivity of all different types described herein. Therefore, this embodiment may be most useful for scenarios where maximum sensitivity is the highest priority.

[0080] The same Type 1 method described above for detecting defect candidates in the A1 and A2 images may also or alternatively be used to detect defect candidates in the B1 and B2 images (and / or the C1 and C2 images). For example, to detect defect candidates in the B1 and B2 images, the difference image generation step 404 may be performed as described above except for B1-CR(A1, A2), B2-CR(A1, A2), and the difference image generation step 408 may be performed as described above except for B1-CR(C1, C2), B2-CR(C1, C2). In a similar manner, to detect defect candidates in the C1 and C2 images, the difference image generation step 404 may be performed as described above, except with respect to C1-CR(A1, A2), C2-CR(A1, A2), and the difference image generation step 408 may be performed as described above, except with respect to C1-CR(B1, B2), C2-CR(B1, B2). The CR for each of these defect detections may be generated using appropriate images as described above. Each of the other methods described herein may also be used in a similar manner to detect defects in the B1 and B2 images (and / or the C1 and C2 images), instead of or in addition to being used to detect defects in the A1 and A2 images.

[0081] If the reticle includes more than three dies in a single row, the Type 1 method described above may also be modified. For example, for detection of defects in any one of the dies, any two of the other dies are used (for detection of defects in the A1 and A2 images, images from B1, B2, C1, and C2 are used or images from C1, C2, D1, and D2 are used (not shown)). In another example, for detection of defects in any one of the dies, multiple inspections may be performed using different combinations of dies (for detection of defects in the A1 and A2 images, a first inspection may be performed using images from B1, B2, C1, and C2, and a second inspection may be performed using images from C1, C2, D1, and D2 (not shown)). In such an example, both the first and second inspections may be performed using dual inspection as described above, wherein a defect candidate is identified at a location in the A1 and A2 images only if both dual inspections detect the defect candidate at that location. Each of the other methods described herein may be modified in a similar manner.

[0082] Figure 5 The second method ("Type 2") shown in FIG. 1 uses a reference image from the same row as the test image to generate a CR image for each test image and stacks the resulting difference images across 2 rows. In this embodiment, by scanning the wafer ( Figure 5 500 for generating A1, A2, B1, B2, C1, and C2 for two different rows on the wafer (not shown). In one embodiment, generating the stacked difference 1 includes generating first and second difference images. In one such embodiment, generating the first difference image includes separately subtracting the first CR from an image generated for multiple instances of the first patterned area in the first of the different rows on the wafer, and generating the second difference image includes separately subtracting the second CR from an image generated for multiple instances of the first patterned area in the second of the different rows on the wafer. For example, in Figure 5 , generating a first difference image (e.g., the top row of set 502 for difference 1) includes subtracting the first CR alone from the image generated for A1 and generating a second difference image (e.g., the bottom row of set 502) includes subtracting the second CR alone from the image generated for A2, as shown in step 504.

[0083] In some embodiments, the computer subsystem is configured to generate a first CR from an image of multiple instances of a second patterned area in a first one in different rows on a wafer and generate a second CR from an image of multiple instances of a second patterned area in a second one in different rows on a wafer. Figure 5, the first CR subtracted from A1 in step 504 may be generated as CR(B1), and the second CR subtracted from A2 in step 504 may be generated as CR(B2). In one embodiment, CR(B1) indicates that the CR is calculated from an image generated by scanning each instance of B1 in row 1. However, CR(B1) may be generated from less than all images of the B1 patterned area scanned in row 1. For example, in some instances, a suitable CR may be generated from two B1 images, but the quality of the CR will generally increase as the number of reference images used to generate the CR increases. Therefore, it may be desirable, but not necessary, to use all available reference images to generate the CR. The same applies to all other CRs described herein.

[0084] In one embodiment, generating stacked difference 2 includes generating third and fourth difference images. In additional embodiments, generating the third difference image includes separately subtracting the third CR from multiple instances of the image generated for the first patterned area in the first of the different rows on the wafer, and generating the fourth difference image includes separately subtracting the fourth CR from multiple instances of the image generated for the first patterned area in the second of the different rows on the wafer. For example, in Figure 5 , generating a third difference image (e.g., the top row of set 506 for difference 2) in step 508 includes subtracting the third CR alone from the image generated for A1 and generating a fourth difference image (e.g., the bottom row of set 506) includes subtracting the fourth CR alone from the image generated for A2, as shown in step 508.

[0085] In another embodiment, the computer subsystem is configured to generate a third CR from an image of multiple instances of a third one of the patterned areas in a first one of the different rows on the wafer and generate a fourth CR from an image of multiple instances of a third one of the patterned areas in a second one of the different rows on the wafer. Figure 5 As shown in , the third CR subtracted from A1 in step 508 may be generated as CR(C1), and the fourth CR subtracted from A2 in step 508 may be generated as CR(C2).

[0086] like Figure 5 As shown in , the computer subsystem can generate stacked difference 1 510 by combining at least a portion of the first and second difference images (i.e., at least some difference 1 in the two rows in set 502) to generate stacked difference 1 and generate stacked difference 2 512 by combining at least a portion of the third and fourth difference images (i.e., at least some difference 2 in the two rows in set 506) to generate stacked difference 2.

[0087] The computer subsystem is also configured to detect defect candidates in multiple instances of the first patterned area on the wafer by applying a defect detection method to different stacked difference images. The defect detection method determines that the defect candidate is present at the location in each of the multiple instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images. For example, the computer subsystem may be configured to perform dual detection 514 using stacked difference 1 510 and stacked difference 2 512. Dual detection 514 may be performed as further described herein. The output of the dual detection step may be a defect detection result 516, where the location of any defect candidate detected by dual detection is identified in each of the patterned areas with some mark (in this case, an asterisk). In this way, the output of the dual detection step is a full correlation of reported defects without individual frame information. The result 516 may be further configured as described herein.

[0088] Thus, the Type 2 method also uses images generated from different reticle rows on the wafer in a new and important way. For example, while the embodiments described above perform reference image generation using images from the same reticle row on the wafer as test images (e.g., CR(B1) for A1 and CR(B2) for A2), the embodiments perform difference image stacking by stacking difference images across 2 reticle rows (e.g., generating a stacked difference 1 from a difference 1 image generated from the A1 image and a difference 1 image generated from the A2 image). In addition, dual detection is performed based on a comparison between the test patterned area and two other patterned areas in the reticle row (e.g., AB and AC).

[0089] The Type 2 method has significant advantages for noise reduction. For example, the Type 2 method will have more noise reduction capabilities than the currently used MDR RDD systems and methods.

[0090] Figure 6 The third method ("Type 3") shown in FIG. 1 uses a reference image from the same row as the test image to generate a CR image for each test image and stack the resulting difference images in the same row. In this embodiment, the CR image can be generated by scanning the wafer ( Figure 6600 of generating A1, A2, B1, B2, C1, and C2 for two different rows on the wafer (not shown). In one embodiment, generating the stacked difference 1 includes generating the first and second difference images. In some such embodiments, generating the first difference image includes separately subtracting the first CR from the image generated for multiple instances of the first patterned area in the first of the different rows on the wafer, and generating the second difference image includes separately subtracting the second CR from the image generated for multiple instances of the first patterned area in the first of the different rows on the wafer. For example, in Figure 6 , generating a first difference image (e.g., the top row of set 602 for difference 1) includes subtracting the first CR alone from the image generated for A1 and generating a second difference image (e.g., the bottom row of set 602) includes subtracting the second CR alone from the image generated for A1, as shown in step 604.

[0091] In another such embodiment, the computer subsystem is configured to generate a first CR from an image of multiple instances of a second one of the patterned areas in a first one of the different rows on the wafer and generate a second CR from an image of multiple instances of a third one of the patterned areas in a first one of the different rows on the wafer. Figure 6 As shown in , the first CR subtracted from A1 in step 604 may be generated as CR(B1), and the second CR subtracted from A1 in step 604 may be generated as CR(C1).

[0092] In one embodiment, generating stacked difference 2 includes generating third and fourth difference images. In additional embodiments, generating the third difference image includes separately subtracting the third CR from images generated for multiple instances of the first patterned area in the second of the different rows on the wafer, and generating the fourth difference image includes separately subtracting the fourth CR from images generated for multiple instances of the first patterned area in the second of the different rows on the wafer. For example, in Figure 6 , generating a third difference image (e.g., the top row of set 606 for difference 2) includes subtracting the third CR alone from the image generated for A2 and generating a fourth difference image (e.g., the bottom row of set 606) includes subtracting the fourth CR alone from the image generated for A2, as shown in step 608.

[0093] In a further embodiment, the computer subsystem is configured to generate a third CR from an image of multiple instances of the second one of the patterned areas in a different row on the wafer and generate a fourth CR from an image of multiple instances of the third one of the patterned areas in a different row on the wafer. Figure 6As shown in , the third CR subtracted from A2 in step 608 may be generated as CR(B2), and the fourth CR subtracted from A2 in step 608 may be generated as CR(C2).

[0094] like Figure 6 As shown in , the computer subsystem can generate stacked difference 1 610 by combining at least a portion of the first and second difference images (i.e., at least some difference 1 in the two rows in set 602) to generate stacked difference 1 and generate stacked difference 2 612 by combining at least a portion of the third and fourth difference images (i.e., at least some difference 2 in the two rows in set 606) to generate stacked difference 2.

[0095] The computer subsystem is also configured to detect defect candidates in multiple instances of the first patterned area on the wafer by applying a defect detection method to different stacked difference images. The defect detection method determines that the defect candidate is present at the location in each of the multiple instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images. For example, the computer subsystem may be configured to use stacked difference 1 610 and stacked difference 2 612 to perform dual detection 614, which may be performed as further described herein. The output of the dual detection step may be a defect detection result 616, where the location of any defect candidate detected by the dual detection is identified in each of the patterned areas with some marking (in this case, an asterisk). In this way, the output of the dual detection step is a full correlation of reported defects without individual frame information. The result 616 may be further configured as described herein.

[0096] Thus, the Type 3 method also uses images generated from different reticle rows on the wafer in new and important ways. For example, although the embodiments described above perform reference image generation using images from the same reticle row on the wafer as test images (e.g., CR(B1) for A1 and CR(B2) for A2), and the embodiments described above perform difference image stacking by stacking difference images in 1 reticle row (e.g., generating stacked difference 1 from difference 1 images generated only from row 1 images and not row 2 images), dual inspection is performed across rows (e.g., one inspection is performed for 1 row and another inspection is performed for another row).

[0097] Figure 7 The fourth method ("Type 4") shown in FIG. 1 uses reference images from two rows to generate a CR image for each test image and stacks the resulting difference images in the same row. In this embodiment, the CR image can be generated by scanning the wafer ( Figure 7In the embodiment, generating stacked difference 1 includes generating a first difference image (difference 1) and combining at least a portion of difference 1 to generate stacked difference, and generating stacked difference 2 includes generating a second difference image (difference 2) and combining at least a portion of difference 2 to generate stacked difference 2.

[0098] In one such embodiment, generating difference 1 includes individually subtracting the first CR from images generated for multiple instances of the first patterned region in a first one of the different rows on the wafer. Figure 7 , generating difference 1 (e.g., set 702 of difference 1) includes subtracting the first CR alone from the image generated for A1, as shown in step 704. In another embodiment, the computer subsystem is configured to generate the first CR from images of multiple instances of the second in patterned areas of the first and the second in different rows on the wafer. For example, Figure 7 As shown in , the first CR subtracted from A1 in step 704 may be generated as CR(B1, B2).

[0099] In a further embodiment, generating difference 2 includes separately subtracting the second CR from the images generated for multiple instances of the first patterned area in a second one of the different rows on the wafer. Figure 7 , generating difference 2 (e.g., set 706 of difference 2) includes subtracting the second CR from the image generated for A2 alone, as shown in step 708. In some such embodiments, the computer subsystem is configured to generate the second CR from images of multiple instances of the third one of the patterned areas in the first and second ones in different rows on the wafer. Figure 7 As shown in , the second CR subtracted from A2 in step 708 may be generated as CR(C1, C2).

[0100] like Figure 7 As shown in , the computer subsystem may generate stacked difference 1 710 by combining at least a portion of difference 1 (i.e., at least some of difference 1 in set 702) to generate stacked difference 1 and generate stacked difference 2 712 by combining at least a portion of difference 2 (i.e., at least some of difference 2 in set 706) to generate stacked difference 2.

[0101] The computer subsystem is also configured to detect defect candidates in multiple instances of the first patterned area on the wafer by applying a defect detection method to different stacked difference images. The defect detection method determines that the defect candidate is present at the location in each of the multiple instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images. For example, the computer subsystem may be configured to perform a dual detection 714 using stacked difference 1 710 and stacked difference 2 712. The dual detection 714 may be performed as further described herein. The output of the dual detection step may be a defect detection result 716, where the location of any defect candidate detected by the dual detection is identified in each of the patterned areas with some mark (in this case, an asterisk). In this way, the output of the dual detection step is a full correlation of reported defects without individual frame information. The result 716 may be further configured as described herein.

[0102] Thus, the Type 4 method also uses images generated from different reticle rows on the wafer in new and important ways. For example, the Type 4 method performs reference image generation using images (e.g., CR(B1, B2)) from different reticle rows on the wafer. In addition, while the Type 4 method performs difference image stacking by stacking difference images in 1 reticle row (e.g., generating a stacked difference 1 from a difference 1 image generated only from row 1 images and not row 2 images), dual inspection is performed across rows (e.g., one inspection is performed for the A1 image in row 1 and another inspection is performed for the A2 image in row 2).

[0103] The Type 4 method has significant advantages over other methods and systems currently used and described herein. For example, the Type 4 method will advantageously start with fewer total repeater types than the other types described herein. In particular, disturber point repeater types that do not have a signal in more than one row that can be detected by the currently used MDR RDD method will be suppressed. Therefore, the Type 4 method may have the highest disturber point suppression capability and best sensitivity of the embodiments described herein compared to the currently used RDD methods and systems. Therefore, the Type 4 method may be particularly useful for scenarios where the defect inspection budget is relatively limited.

[0104] The computer subsystem is further configured to identify defects on the retracted reticle based on the detected defect candidates. As further described herein, when a defect candidate is detected by dual detection, the defect candidate is initially identified in each of the test images for which the dual detection is performed. However, the defect may not actually be located at the defect candidate location in each test image. The computer subsystem may perform one or more steps to first determine whether the defect is located at the defect candidate location in each test image and then determine whether the defect is sufficiently repeated in multiple instances of the patterned area to be considered a retracted reticle defect. These steps may be performed at least in part based on some local characteristics of the test images at the location of the detected defect candidate as further described herein.

[0105] In some embodiments, identifying defects on the reticle includes determining a local image characteristic at a location in each of the images of the plurality of instances of the first patterned area, comparing the determined local image characteristic at the location in each of the images of the plurality of instances of the first patterned area to a threshold, determining the number of images of the plurality of instances of the first patterned area having the determined local image characteristic above the threshold, and comparing the determined number to a repeater threshold. For example, the defect detection result generated by any of Types 1 to 4 described above may be used as Figure 8 The result 800 shown in is input to the recognition step.

[0106] The computer subsystem can then calculate local image properties, such as Figure 8 800. For example, the computer subsystem may calculate a local signal-to-noise ratio (SNR) for each defect candidate location in the result 800. More specifically, the computer subsystem may calculate a local SNR for each defect candidate location in each A1 and A2 test image. Although some embodiments may be described herein with respect to local SNR, the embodiments may use any other suitable local image characteristics or metrics for this step, including but not limited to magnitude, point similarity, local saliency, energy parameters, repeater signals, etc. Any of these local image characteristics may be determined in any suitable manner known in the art.

[0107] The computer subsystem may compare the determined local image characteristics at the locations in each of the images of the multiple instances of A1 and A2 to a threshold value (essentially a defect detection threshold value). The result of this step may be result 804, where asterisks indicate defect candidate locations in the test image have local image characteristics above the threshold value, and degree signs indicate defect candidate locations in the test image have local image characteristics below the threshold value.

[0108] The computer subsystem may perform disturber point filtering by local image characteristics as shown in step 806, thereby producing defect detection results 808. In one such example, disturber point filtering may include eliminating any defect candidates having a local SNR below a threshold (i.e., all defect candidates whose locations are shown with a degree symbol in result 804) and keeping any defect candidates having a local SNR above a threshold (i.e., all defect candidates whose locations are shown with an asterisk in result 804). In this manner, result 808 (the output of the disturber point filtering step) is a partial correlation of reported defects. In this context, partial correlation of reported defects means that the local SNR test need not pass every test image for a defect to be reported. At least some of the test images must pass the local SNR test to be considered a repeater defect, and how a test image may have to pass to be considered a repeater defect is controlled by the repeater threshold. The repeater threshold may be pre-set by a user or in another suitable manner and may vary depending on a number of variables, such as the level of process variation expected across the wafer, which may cause the defect to repeat in less than all test images and / or may cause noise in the test images that may be erroneously detected as a defect. Thus, even if a defect does not repeat at every instance of a test image, it may still be appropriately considered a repeater defect under appropriate circumstances.

[0109] Final repeater defect candidates (i.e., defects that repeat in a number greater than a repeater threshold) may all be identified as defects on the reticle. However, identifying defects on the reticle based on the final repeater defect candidates may include some additional analysis as desired. For example, the repeaters may be input to any suitable post-processing known in the art, such as defect classification (e.g., via any suitable defect classification method or algorithm). Repeater analysis may also include sampling defects for review, which may be performed on a subset of all final repeater defect candidates to verify and / or classify the sampled defects. This review sampling may produce a review sample that includes information for a selected subset of the final repeater defect candidates, which may then be sent to a defect review tool (not shown) that performs defect review.

[0110] In general, the embodiments described herein are configured for detecting reticle repeater defects, but some reticle repeater defects may actually be die repeater defects. For the purposes of the applications described herein, the user may wish to keep only the reticle repeaters. If this is the case, the die repeaters of the MDR should be removed from the final report. In particular, the embodiments described herein may be used for applications where the user is looking for repeater defects that are caused by problems with the reticle (mask). In such cases, defects at the same location in different dies on the same reticle are most likely caused by wafer processing rather than by the reticle itself.

[0111] In other words, a defect that occurs at substantially the same location in multiple dies on a single reticle on a wafer is likely not due to an actual reticle defect (e.g., a damaged patterned feature, a particle or foreign matter, and the like). Instead, if a defect is repeated in multiple dies on a reticle on a wafer, then the defect that is repeated in dies on the wafer is likely due to an interaction problem between the reticle, the wafer, and the process (e.g., an edge design that failed due to a design / process interaction problem). Such die repeater defects may be of interest in some use cases, and the embodiments described herein may be used in such cases. However, the embodiments described herein may also be specifically used to detect defects that repeat from printed reticle instance to printed reticle instance and therefore may be considered to be caused by a reticle defect.

[0112] In some examples, the computer subsystem will not report the defect in reticle coordinates. In other words, when the computer subsystem detects a defect, the coordinates of the location of the defect may not have been determined by the computer subsystem relative to the reticle. In this way, the coordinates of the defect determined by the computer subsystem may be translated into reticle coordinates by the computer subsystem or another system or method. The coordinates may be translated in a number of different ways. For example, one relatively simple way is to translate the coordinates reported for the defect to a reference common to the wafer and the reticle (e.g., the design of the wafer and the reticle). In one such example, the image generated by the inspection subsystem for the defect may be aligned to the design of the wafer. Once the design coordinates of the defect are determined, those coordinates may be translated into reticle coordinates based on the known relationship between the design and the reticle.

[0113] The computer subsystem may preferably store information of all detected defects so that the information of all detected defects is available to the RDD and any other functions. Once the RDD is executed by the computer subsystem, the information stored for defects determined not to be retract reticle and / or die repeater may be eliminated from the stored information before being sent to another computer subsystem, another method, back to memory, etc.

[0114] The computer subsystem may be configured to store the results of any one or more of the steps described herein in a computer-readable storage medium. The results may include any of the results described herein and may be stored in any manner known in the art. The storage medium may include any storage medium described herein or any other suitable storage medium known in the art. After the results have been stored, the results may be accessed in the storage medium and used by any of the method or system embodiments described herein, formatted for display to a user, used by another software module, method or system, etc. For example, defects detected on a reticle may be used to control and / or alter a reticle repair or cleaning process, where the goal is to remove or repair one or more reticle defects to thereby prevent such reticle defects from being printed on other wafers. In one such example, information of defects detected on a reticle may be used by a computer subsystem described herein, another system (e.g., a reticle repair system (not shown), which may or may not be part of or coupled to system embodiments described herein), or another method (e.g., a reticle repair method) to determine where on the reticle a repair or cleaning process should be performed. Information of defects detected on the reticle may also be used to determine one or more parameters of such a process, such as the area on which the process will be performed, the length of time the process will be performed, which chemicals or gases should be used in the process, etc. In this way, the altered reticle may be used to print other wafers in a lithography process. The altered reticle may also be re-inspected at another time after it is reintroduced into the lithography process using the embodiments described herein.

[0115] Although using a 2-row sample plan as described above in Types 1-4 is particularly useful for the embodiments described herein (e.g., minimizing system memory requirements and reducing upset point correlations), more than two rows may be sampled and processed together for detecting defects on a reticle. Fig. 9 Displayed on chip ( Fig. 9, and the like. The test images 900 generated in three rows on a wafer (not shown) can be processed together for reticle defect detection. In this embodiment and other embodiments described herein, A1, A2, and A3 refer to patterned areas printed on a wafer with a reticle during a photolithography process, the patterned areas corresponding to the same area on the reticle and printed in different rows on the wafer. Similarly, B1, B2, and B3 (and C1, C2, and C3) refer to patterned areas printed on a wafer with a reticle during a photolithography process, the patterned areas corresponding to the same area on the reticle and printed in different rows on the wafer. In this way, A1, A2, and A3 can be the same portion of a die printed on a wafer with an MDR, B1, B2, and B3 can be the same portion of a different die printed on the wafer with the MDR, and C1, C2, and C3 can be the same portion of another different die printed on a wafer with an MDR. Test images 900 of A1, A2, A3, B1, B2, B3, C1, C2, and C3 may be generated by scanning three different rows on the wafer. Figure 3 In S1 shown in FIG. 1 , test images of A1 , B1 , and C1 can be generated, and each set of A1 , B1 , and C1 can be generated for a single MDR example printed on a wafer. Figure 3 In S3 of FIG. 1 , test images of A2, B2, and C2 may be generated, and each set of A2, B2, and C2 may be generated for a single MDR example printed on a wafer. Figure 3 In S5 shown in FIG. 1 , test images of A3, B3, and C3 may be generated, and each set of A3, B3, and C3 may be generated for a single MDR example printed on a wafer.

[0116] If the only goal is to achieve maximum noise reduction, then the approach would be to stack difference images across all three reticle rows in the inspection sample plan, but such an approach has substantially limited disturbing point suppression capabilities. For example, only one set of stacked difference images may be produced with a 3-row stacking approach, making disturbing point suppression more challenging. Alternatively, the embodiments described herein may be configured to inspect every two reticle rows by stacking difference images. This may be performed using any of the methods described herein, but in some embodiments, the difference images may be stacked to minimize the noise reduction. Figure 4 and 10 Type 1 is used as an example to illustrate.

[0117] In one embodiment, the computer subsystem is configured to perform generation and detection for the first and second in different rows on the wafer, perform generation and detection separately for the second and third in different rows on the wafer, and input the results of the two detection steps into the recognition step. In this way, two sets of stacked difference images are independently generated across 3 reticle rows using the 2-row stacking method. Figure 4The method shown in is executed for rows 1 and 2, but Fig.10 The method shown in is performed for rows 2 and 3. In this way, row 2 is Figure 4 and 10 In other words, for the triple-retract row sample plan, two rows are tested at a time (row 1, row 2 and row 2, row 3), where row 2 is redundant in both tests.

[0118] like Figure 4 The first method shown in Fig.10 The separately performed Type 1 method shown in also uses reference images from 2 rows to generate a CR image for each test image and stacks the resulting difference images across the 2 rows for maximum noise reduction. In one such embodiment, generating a first difference image includes separately subtracting the first CR from an image generated for multiple instances of the first patterned area in a first one of different rows on the wafer, and generating a second difference image includes separately subtracting the first CR from an image generated for multiple instances of the first patterned area in a second one of different rows on the wafer. For example, in Fig.10 , generating a first difference image (e.g., the top row of set 1002 for difference 1) includes subtracting the first CR alone from the image generated for A2 and generating a second difference image (e.g., the bottom row of set 1002) includes subtracting the first CR alone from the image generated for A3, as shown in step 1004.

[0119] In some such embodiments, the computer subsystem is configured to generate the first CR from images of multiple instances of the second one of the patterned regions in different rows on the wafer. Fig.10 , the first CR subtracted from A2 in step 1004 may be generated as CR(B2, B3). This same CR may be subtracted from A3 alone, as shown in step 1004.

[0120] In additional embodiments, generating the third difference image includes separately subtracting the second CR from images generated for multiple instances of the first patterned area in a first one of different rows on the wafer and generating the fourth difference image includes separately subtracting the second CR from images generated for multiple instances of the first patterned area in a second one of different rows on the wafer. Fig.10 , generating a third difference image (e.g., the top row of set 1006 for difference 2) includes subtracting the second CR alone from the image generated for A2 and generating a fourth difference image (e.g., the bottom row of set 1006) includes subtracting the second CR alone from the image generated for A3, as shown in step 1008.

[0121] In another embodiment, the computer subsystem is configured to generate a second CR from images of multiple instances of a third party in a patterned area in a different row on the wafer. Fig.10 As shown in , the second CR subtracted from A2 in step 1008 may be generated as CR(C2, C3). This same CR may be subtracted from A3 alone, as shown in step 1008.

[0122] like Fig.10 As shown in , the computer subsystem can generate stacked difference 1 1010 by combining at least a portion of the first and second difference images (i.e., at least some of the difference 1s in the two rows in set 1002) to generate stacked difference 1 and generate stacked difference 2 1012 by combining at least a portion of the third and fourth difference images (i.e., at least some of the difference 2s in the two rows in set 1006) to generate stacked difference 2.

[0123] The computer subsystem is also configured to detect defect candidates in multiple instances of the first patterned area on the wafer by applying a defect detection method to different stacked difference images. The defect detection method determines that the defect candidate is present at the location in each of the multiple instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images. For example, the computer subsystem may be configured to perform dual detection 1014 using stacked difference 1 1010 and stacked difference 2 1012. Dual detection 1014 may be performed as further described herein. The output of the dual detection step may be a defect detection result 1016, where the location of any defect candidate detected by dual detection is identified in each of the patterned areas with some mark (in this case, an asterisk). In this way, the output of the dual detection step is a full correlation of the reported defect candidates without individual frame information. The result 1016 may be configured as further described herein.

[0124] The Type 1, 3-row approach described above has significant advantages for disruptor reduction. For example, the Type 1 described above will advantageously start with a lower number of repeater types than the currently used MDR RDD approach, regardless of the rows on which the Type 1 is performed. In particular, disruptor repeater types that do not have a signal in more than one row that can be detected by the currently used MDR RDD approach will be suppressed, regardless of the rows that are tested.

[0125] Fig.11 An embodiment of inputting the results of the two detection steps into the identification step is shown in FIG. In this embodiment, the defect detection result 416 generated by executing the type 1 method for row 1 and row 2 is as follows: Figure 4 is produced as shown in Fig.10 1016 produced by performing the Type 1 method for rows 2 and 3 as shown in , are input to the identification step together. Disturbing point filtering may be performed separately for each of the defect detection results. For example, disturbing point filtering by local image characteristics may be performed for defect detection result 416 in step 1100, and disturbing point filtering by local image characteristics may also be performed for defect detection result 1016 in step 1102. Defects may be disturbing point filtered by limiting local image characteristics as described above, thereby producing defect detection result 1104. In one such example, disturbing point filtering may include eliminating any defect candidates having a local SNR below a threshold and retaining any defect candidates having a local SNR above a threshold. In this way, result 1104 (the output of the disturbing point filtering step) is a partial correlation of the reported defect.

[0126] Therefore, the embodiments described above have many advantages over currently used methods and systems for detecting defects on shrunken reticles. For example, if a mask repeater defect is not detected, then the mask repeater defect causes a yield loss on each shrunken reticle printed. As the design rules are reduced, the size of the repeater defects that should not be missed becomes smaller and smaller. The signal for the smaller repeater size also becomes weaker, which requires a way to minimize the defect inspection load in a way that achieves preferred noise reduction and preferred upset point suppression. The embodiments described herein provide solutions that meet different user use cases. Some users seek maximum raw sensitivity for the smallest possible repeater, while other users may prefer to limit the defect inspection budget by requiring a minimum total number of repeaters.

[0127] These and other advantages described herein are achieved by a number of important new features of the embodiments described herein. One such new feature is the use of staggered swath scanning based on a swath scan to acquire a swath image. Another new feature includes the use of an alignment die in a first swath row positioned in a swath scan to align other swath rows. An additional new feature is the use of a reference image across multiple swath rows for CR image generation. A further new feature is the stacking of difference images across multiple swath rows for maximum noise reduction. Yet another new feature is the stacking of difference images within the same row as the test image and dual detection across swath rows for maximum disturbing point suppression.

[0128] Each embodiment of the system described herein may be combined with any other embodiment of the system described herein.

[0129] Another embodiment is directed to a computer-implemented method for detecting defects on a reticle. The method includes acquiring an image generated by an inspection subsystem for a wafer. The reticle is used to print patterned areas on the wafer during a photolithography process. The patterned areas include a first patterned area corresponding to the same area on the reticle and printed in a different row on the wafer. The method also includes the above-described generating, detecting, and identifying steps performed by a computer subsystem coupled to the inspection subsystem.

[0130] Each step of the method may be performed as further described herein. The method may also include any other steps that may be performed by the inspection subsystem and / or computer subsystem described herein that may be configured according to any embodiment described herein. In addition, the method described above may be performed by any of the system embodiments described herein.

[0131] Additional embodiments relate to a non-transitory computer-readable medium storing program instructions that are executed on a computer system for performing a computer-implemented method for detecting defects on a retracted reticle. Fig.12 In particular, as shown in Fig.12 , a non-transitory computer-readable medium 1200 includes program instructions 1202 executable on a computer system 1204. The computer-implemented method may include any of the steps of any method described herein.

[0132] Program instructions 1202 implementing methods such as those described herein may be stored on computer readable medium 1200. The computer readable medium may be a storage medium such as a magnetic or optical disk, a tape, or any other suitable non-transitory computer readable medium known in the art.

[0133] The program instructions may be implemented in any of a variety of ways, including process-based techniques, component-based techniques, and / or object-oriented techniques, etc. For example, the program instructions may be implemented using ActiveX controls, C++ objects, JavaBeans, Microsoft Foundation Classes ("MFC"), SSE (Streaming SIMD Extensions), or other techniques or methods as desired.

[0134] Computer system 1204 may be configured according to any of the embodiments described herein.

[0135] In view of this description, further modifications and alternative embodiments of various aspects of the present invention will be apparent to those skilled in the art. For example, systems and methods for detecting defects on a retracted reticle are provided. Therefore, this description should be construed as being illustrative only and for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It will be understood that the forms of the invention shown and described herein are to be considered as currently preferred embodiments. As will be apparent to those skilled in the art after having the benefit of this description of the invention, elements and materials may be substituted for those illustrated and described herein, parts and processes may be reversed, and certain features of the present invention may be utilized alone. Changes may be made to the elements herein without departing from the spirit and scope of the invention as described in the appended claims.

Claims

1. A system configured to detect defects on a reticle, comprising: an inspection subsystem configured to generate an image of a wafer, wherein a reticle is used to print patterned regions on the wafer during a photolithography process, and wherein the patterned regions include first patterned regions corresponding to the same region on the reticle and printed in different rows on the wafer; and A computer subsystem configured to: generating different stacked difference images of a plurality of instances of the first patterned areas in the different rows based on the images generated by the inspection subsystem for the first patterned areas in the different rows; detecting defect candidates in the plurality of instances of the first patterned area on the wafer by applying a defect detection method to the different stacked difference images, wherein the defect detection method determines that the defect candidate is present at the location in each of the plurality of instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images; and Defects on the reticle are identified based on the detected defect candidates.

2. The system of claim 1 , wherein the inspection subsystem is further configured to generate the image by interleaving a scan of the first patterned area in the different row on the wafer with a scan of a second one of the patterned areas in the different row on the wafer. 3 . The system of claim 2 , wherein the first patterned area and the second one of the patterned areas correspond to different portions of a die on the reticle.

4. The system of claim 2, wherein the first patterned area and the second of the patterned areas correspond to different instances of a die printed on the reticle on the wafer.

5. The system of claim 1, wherein the reticle is a multi-die reticle, and wherein the first patterned area includes at least a portion of only a first die of the multi-die reticle printed on the wafer.

6. The system of claim 1 , wherein generating a first of the different stacked difference images comprises generating first and second difference images and combining at least a portion of the first and second difference images to generate the first of the different stacked difference images, and wherein generating a second of the different stacked difference images comprises generating third and fourth difference images and combining at least a portion of the third and fourth difference images to generate the second of the different stacked difference images.

7. The system of claim 6 , wherein generating the first difference image comprises separately subtracting a first calculated reference from the image generated for the plurality of instances of the first patterned area in a first one of the different rows on the wafer, and generating the second difference image comprises separately subtracting the first calculated reference from the image generated for the plurality of instances of the first patterned area in a second one of the different rows on the wafer.

8. The system of claim 7 wherein the computer subsystem is further configured to generate the first computational reference from the images of a plurality of instances of a second one of the patterned areas in the different rows on the wafer.

9. The system of claim 7 , wherein generating the third difference image comprises separately subtracting a second calculated reference from the image generated for the plurality of instances of the first patterned area in the first one of the different rows on the wafer and generating the fourth difference image comprises separately subtracting the second calculated reference from the image generated for the plurality of instances of the first patterned area in the second one of the different rows on the wafer.

10. The system of claim 9, wherein the computer subsystem is further configured to generate the second computational reference from the images of multiple instances of a third one of the patterned areas in the different rows on the wafer.

11. The system of claim 6, wherein generating the first difference image comprises separately subtracting a first calculated reference from the image generated for the plurality of instances of the first patterned area in a first one of the different rows on the wafer and generating the second difference image comprises separately subtracting a second calculated reference from the image generated for the plurality of instances of the first patterned area in a second one of the different rows on the wafer.

12. The system of claim 11 wherein the computer subsystem is further configured to generate the first computational reference from the images of multiple instances of the second one of the patterned areas in the first one of the different rows on the wafer and to generate the second computational reference from the images of multiple instances of the second one of the patterned areas in the second one of the different rows on the wafer.

13. The system of claim 11 , wherein generating the third difference image comprises individually subtracting a third calculated reference from the plurality of instances of the image generated for the first patterned area in the first one of the different rows on the wafer and generating the fourth difference image comprises individually subtracting a fourth calculated reference from the image generated for the plurality of instances of the first patterned area in the second one of the different rows on the wafer.

14. The system of claim 13 wherein the computer subsystem is further configured to generate the third computational reference from the images of multiple instances of a third one of the patterned areas in the first one of the different rows on the wafer and to generate the fourth computational reference from the images of multiple instances of the third one of the patterned areas in the second one of the different rows on the wafer.

15. The system of claim 6 , wherein generating the first difference image comprises separately subtracting a first calculated reference from the image generated for the plurality of instances of the first patterned area in a first one of the different rows on the wafer and generating the second difference image comprises separately subtracting a second calculated reference from the image generated for the plurality of instances of the first patterned area in the first one of the different rows on the wafer.

16. The system of claim 15 wherein the computer subsystem is further configured to generate the first computational reference from the images of multiple instances of a second one of the patterned areas in the first one of the different rows on the wafer and to generate the second computational reference from the images of multiple instances of a third one of the patterned areas in the first one of the different rows on the wafer.

17. The system of claim 15 wherein generating the third difference image comprises separately subtracting a third calculated reference from the image generated for the plurality of instances of the first patterned area in a second one of the different rows on the wafer and generating the fourth difference image comprises separately subtracting a fourth calculated reference from the image generated for the plurality of instances of the first patterned area in the second one of the different rows on the wafer.

18. The system of claim 17 wherein the computer subsystem is further configured to generate the third computational reference from the images of multiple instances of the second one of the patterned areas in the second one of the different rows on the wafer and to generate the fourth computational reference from the images of multiple instances of the third one of the patterned areas in the second one of the different rows on the wafer.

19. The system of claim 1 , wherein generating a first one of the different stacked difference images comprises generating a first difference image and combining at least a portion of the first difference image to generate the first one of the different stacked difference images, and wherein generating a second one of the different stacked difference images comprises generating a second difference image and combining at least a portion of the second difference image to generate the second one of the different stacked difference images.

20. The system of claim 19, wherein generating the first difference image comprises individually subtracting a first calculated reference from the images generated for the plurality of instances of the first patterned area in a first one of the different rows on the wafer.

21. The system of claim 20, wherein the computer subsystem is further configured to generate the first computational reference from the images of a plurality of instances of the second one of the patterned areas in the first and second ones in the different rows on the wafer.

22. The system of claim 20, wherein generating the second difference image comprises individually subtracting a second calculated reference from the images generated for the plurality of instances of the first patterned area in a second one of the different rows on the wafer.

23. The system of claim 22, wherein the computer subsystem is further configured to generate the second computational reference from the images of a plurality of instances of a third one of the patterned areas in the first and second ones in the different rows on the wafer.

24. The system of claim 1 , wherein the computer subsystem is further configured to perform the generating and detecting for a first and a second of the different rows on the wafer, to perform the generating and detecting separately for the second and a third of the different rows on the wafer, and to input the results of the two detecting steps into the identifying step.

25. The system of claim 1 , wherein identifying the defect on the reticle comprises determining a local image characteristic at the location in each of the images of the plurality of instances of the first patterned area, comparing the determined local image characteristic at the location in each of the images of the plurality of instances of the first patterned area to a threshold, determining a number of the images of the plurality of instances of the first patterned area having the determined local image characteristic above the threshold, and The determined number is compared to a repeater threshold.

26. The system of claim 1, wherein the shrink reticle is a single die shrink reticle.

27. The system of claim 1, wherein the shrink reticle is a multi-die shrink reticle.

28. The system of claim 1, wherein the retractor is configured for use at extreme ultraviolet wavelengths.

29. A non-transitory computer readable medium storing program instructions that are executed on a computer system for performing a computer implemented method for detecting defects on a retraction reticle, wherein the computer implemented method comprises: Acquiring an image generated by an inspection subsystem for a wafer, wherein a reticle is used to print patterned regions on the wafer during a photolithography process, and wherein the patterned regions include first patterned regions corresponding to the same region on the reticle and printed in different rows on the wafer; generating different stacked difference images of a plurality of instances of the first patterned areas in the different rows based on the images generated by the inspection subsystem for the first patterned areas in the different rows; detecting defect candidates in the plurality of instances of the first patterned area on the wafer by applying a defect detection method to the different stacked difference images, wherein the defect detection method determines that the defect candidate is present at the location in each of the plurality of instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images; and Defects on the reticle are identified based on the detected defect candidates, wherein the acquiring, generating, detecting, and identifying are performed by the computer system coupled to the inspection subsystem.

30. A computer-implemented method for detecting defects on a reticle, comprising: Acquiring an image generated by an inspection subsystem for a wafer, wherein a reticle is used to print patterned regions on the wafer during a photolithography process, and wherein the patterned regions include first patterned regions corresponding to the same region on the reticle and printed in different rows on the wafer; generating different stacked difference images of a plurality of instances of the first patterned areas in the different rows based on the images generated by the inspection subsystem for the first patterned areas in the different rows; detecting defect candidates in the plurality of instances of the first patterned area on the wafer by applying a defect detection method to the different stacked difference images, wherein the defect detection method determines that the defect candidate is present at the location in each of the plurality of instances of the first patterned area only if the defect detection method detects the defect candidate at the location in the different stacked difference images; and Defects on the reticle are identified based on the detected defect candidates, wherein the acquiring, generating, detecting, and identifying are performed by a computer system coupled to the inspection subsystem.

Images