Measurement-guided inspection using local maximum values ​​of wafer signatures obtained through clustering.

Abstract

A method and system are provided for generating information to be used in setting up a process to be performed on a sample. One method includes clustering dies on a sample based on the color assigned to the dies according to the predicted defect density in the dies determined from measurements performed on the sample, thereby generating initial die clusters. The method also includes analyzing the initial die clusters in position space to determine whether any of the initial die clusters contain two or more die clusters. In addition, the method includes designating initial die clusters that do not contain two or more die clusters, and any of the initial die clusters that contain two or more die clusters, as final die clusters. The method further includes storing information on the final die clusters for use in setting up a process to be performed on a sample.

Description

Technical Field

[0001] The present invention generally relates to methods and systems for generating information for use in setting up a set of processes to be performed on a specimen. Certain embodiments relate to measurement-guided inspection using local maxima of wafer signatures by clustering.

Background Art

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

[0003] The manufacture of semiconductor devices such as logic devices and memory devices typically involves processing a substrate such as a semiconductor wafer using a number of semiconductor manufacturing processes to form various functions and multiple layers of the semiconductor device. For example, lithography is a semiconductor manufacturing process involving the transfer of a pattern from a reticle to a resist disposed on a semiconductor wafer. Further examples of semiconductor manufacturing processes include, but are not limited to, chemical mechanical polishing (CMP), etching, deposition, and ion implantation. Multiple semiconductor devices can be fabricated and disposed on a single semiconductor wafer and then separated into individual semiconductor devices.

[0004] To increase the yield of the manufacturing process and thereby increase profits, inspection processes are used at various steps during semiconductor manufacturing to detect defects on the wafer. Inspection has always played an important role in the manufacture of semiconductor devices. However, as the dimensions of semiconductor devices shrink, inspection becomes even more critical for successful manufacture of qualified semiconductor devices.

[0005] The inspection tool has various parameters that can be changed based on the specimen being examined. These changeable parameters often include parameters related to the imaging hardware and those related to image processing. While changing the tool's parameters for each specimen can be advantageous, determining the correct parameters for any single specimen can be particularly challenging.

[0006] Many methods have been developed not only to determine which test parameters are suitable for a given sample, but also which test parameters are suitable for different regions on the same sample within the same test. For example, the variability across samples described herein may mean that a test parameter that is best suited to one sample region may be suboptimal or completely useless to another sample region.

[0007] Methods for selecting inspection parameters that vary from sample area to sample have been developed based on numerous characteristics of the sample and tool. Some of the sample characteristics considered are primarily related to the design formed on the sample, including the characteristics of the patterned mechanisms in the design and where they are formed on the sample. Design characteristics that can be used include those that are independent of the inspection tool and inspection process, such as which of the patterned mechanisms in the design is most important to the device function and therefore should be inspected with the greatest sensitivity. Design characteristics may also include, or alternatively, design characteristics that may have some effect on the output of the inspection tool and therefore affect both the signal and noise in the output of the inspection tool, such as dimensions, orientation, roughness, material properties, and patterned mechanisms in the underlying layers.

[0008] Several other sample characteristics considered relate to noise in the inspection tool output across different regions of the sample. This noise may be related to the design formed on the sample, as mentioned above. It may also be related to other characteristics of the sample, such as variability in the sample resulting from processes performed on it. One obvious reason why inspection parameters might be noise-based is that noise can mask defect signals. For example, if the noise is too similar to, or even more pronounced than, the defect signal may be missed, leading to a failed inspection.

[0009] While it may appear relatively simple, the characteristics of these and other samples, as well as the configuration of the inspection tool and its sample-dependent performance, significantly complicate the setup of the inspection process. For example, even when using only design information for the inspection setup, the setup can be complex due to the complexity of the design and its potential impact on the output of the inspection tool. As a further example, predicting the noise present on a sample and how that noise may vary across samples is becoming increasingly difficult, making noise-related inspection setups challenging. When combinations of factors are considered in the setup of an inspection process, the difficulty of the setup likely increases exponentially with the number of variables.

[0010] Other factors that can complicate the setup of a test recipe include performance considerations and requirements. For example, the fastest test processes generally have the fewest different parameters per sample. In particular, if a test process involves testing with different sets of parameters, it usually takes longer than if the same test could be performed with only one set of parameters. Similarly, a test process usually takes longer if there are more parameter switches, for example, per region of the sample. Therefore, some test recipe setup processes have focused on how to group regions on the sample that are similar in the characteristics described above. Test recipe setup methods may further or alternatively focus on how to achieve test objectives (e.g., overall defect capture rate, capture rate of specific defects of interest (DOI), Nusance detection rate, etc.) in the shortest possible time. These methods may involve considering the potential trade-offs between throughput and performance.

[0011] While significant progress has been made in setting up test recipes over the past few decades, many challenges remain. For example, the increasing complexity of sample designs and the ever-smaller pattern sizes make it increasingly difficult to predict how the sample itself will affect the output of the test tool. In addition, defects of interest are becoming smaller and harder to detect, especially since most test tools are being pushed to or used at the limits of their performance tolerances. These factors, along with the degree to which samples can vary across samples, make it particularly challenging to create test recipes that work well across the entire sample. [Prior art documents] [Patent Documents]

[0012] [Patent Document 1] U.S. Patent Application Publication No. 2020 / 0372631 [Overview of the project] [Problems that the invention aims to solve]

[0013] Therefore, it is advantageous to develop systems and methods for generating information used in setting up processes performed on a sample that do not have one or more of the disadvantages described above. [Means for solving the problem]

[0014] The following descriptions of various embodiments shall not be construed in any way as limiting the subject matter of the appended claims.

[0015] One embodiment relates to a system configured to generate information for use in setting up a process to be performed on a sample. The system includes one or more computer systems configured to cluster dies on a sample based on the color assigned to the dies according to the predicted defect density in the dies determined from measurements performed on the sample, thereby generating initial die clusters. One or more computer systems are also configured to analyze the initial die clusters in position space to determine whether any of the initial die clusters contain two or more die clusters. In addition, one or more computer systems are configured to designate initial die clusters that do not contain two or more die clusters, and any of the initial die clusters that contain two or more die clusters, as final die clusters. One or more computer systems are further configured to store information on the final die clusters for use in setting up a process to be performed on a sample. The system can be further configured as described herein.

[0016] Another embodiment relates to a computer implementation method for generating information used to set up a process to be performed on a sample. The method includes the clustering step, analysis step, designation step, and storage step described above. The steps of the method are performed by one or more computer systems. The steps of the method can be further performed as described herein. The method may include any other steps of any other method described herein. The method can be performed by any of the systems described herein.

[0017] Another embodiment relates to a non-temporary computer-readable medium storing program instructions executable on a computer system for performing a computer implementation method for generating information used to set up a process to be performed on a specimen. The computer implementation method includes steps of the method described above. The computer-readable medium can be further configured as described herein. The steps of the computer implementation method can be performed as described herein. In addition, a computer implementation method on which program instructions can be executed may include any other steps of any other method described herein.

[0018] Further advantages of the present invention will become apparent to those skilled in the art by referring to the detailed description of the following preferred embodiments and the accompanying drawings. [Brief explanation of the drawing]

[0019] [Figure 1] This is a schematic diagram showing a side view of an embodiment of the system configured as described herein. [Figure 2] This is a schematic diagram showing a side view of an embodiment of the system configured as described herein. [Figure 3] This is a schematic diagram showing one example of a die-level wafer map of defect density probabilities with four local clusters. [Figure 4] It is a schematic diagram showing an example of a die-level wafer map of a clustering result generated by KMeans. [Figure 5] It is a schematic diagram showing an example of a die-level wafer map of the final die clusters that can be generated by the embodiments described herein. [Figure 6] It is a flowchart showing an embodiment of steps that can be executed to generate the final die clusters as described herein. [Figure 7] It is a block diagram showing an embodiment of a non-transitory computer-readable medium storing program instructions for causing a computer system to execute the computer-implemented method described herein. **Embodiments for Carrying Out the Invention**

[0020] Although the present invention is capable of accepting various changes and alternative forms, specific embodiments thereof are shown by way of example in the figures and will be described in detail herein. The drawings may not be to scale. However, it should be understood that the drawings and their detailed description are not intended to limit the present invention to the specific forms disclosed, but rather, are intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present invention as defined by the appended claims.

[0021] Attention should be paid to the fact that the figures are not drawn to a fixed scale. In particular, the scale of some of the elements in the figures is greatly exaggerated to emphasize the characteristics of the elements. Also, note that the figures are not drawn to the same scale. Elements shown in multiple figures that may be similarly configured are shown using the same reference numerals. Unless otherwise stated herein, any of the elements described and illustrated can include any suitable commercially available elements.

[0022] One embodiment relates to a system configured for generating information to be used in setting up a process to be performed on a specimen. In some embodiments, the specimen is a wafer. The wafer can include any wafer known in the semiconductor technology. Some embodiments may be described herein in relation to one or more wafers, but embodiments are not limited to specimens on which they may be used. For example, embodiments described herein can be used for specimens such as reticles, flat panels, personal computer (PC) boards, and other semiconductor specimens.

[0023] One embodiment of such a system is shown in Figure 1. The system may include an imaging system comprising at least an energy source and a detector. The energy source is configured to generate energy directed to a specimen. The detector is configured to detect energy from the specimen and generate an output in response to the detected energy.

[0024] In one embodiment, the energy directed to the specimen includes light, and the energy detected from the specimen includes light. For example, as shown in Figure 1, the imaging system 10 includes an illumination subsystem configured to direct light to the specimen 14. The illumination subsystem includes at least one light source, e.g., light source 16. The illumination subsystem is configured to direct light to the specimen at one or more angles of incidence, which may include one or more tilt angles and / or one or more perpendicular angles. For example, as shown in Figure 1, light from light source 16 is directed through an optical element 18, then through a lens 20, to a beam splitter 21, which directs the light to the specimen 14 at a perpendicular angle of incidence. The angle of incidence may include any appropriate angle, which may vary depending on, for example, the characteristics of the specimen, defects detected on the specimen, measurements performed on the specimen, etc.

[0025] The illumination subsystem can be configured to direct light onto the specimen at various angles of incidence at different times. For example, the imaging system can be configured to modify one or more properties of one or more elements of the illumination subsystem so that light can be directed onto the specimen at angles of incidence different from those shown in Figure 1. In one such example, the imaging system can be configured to move the light source 16, optical element 18, and lens 20 so that light can be directed onto the specimen at different angles of incidence.

[0026] The imaging system can be configured to direct light onto the specimen at two or more angles of incidence simultaneously. For example, the imaging subsystem may include two or more illumination channels, one of which may include a light source 16, an optical element 18, and a lens 20, as shown in Figure 1, and another illumination channel (not shown) may include similar elements that may be configured differently or in the same way, or may include at least a light source and, optionally, one or more other components such as those further described herein. When such light is directed onto the specimen simultaneously with the other light, one or more properties (e.g., wavelength, polarization, etc.) of the light directed onto the specimen at different angles of incidence may be different, allowing the detector to distinguish between the light produced by the illumination of the specimen at different angles of incidence.

[0027] However, the illumination subsystem may include only one light source (for example, light source 16 shown in Figure 1), and the light from the light source may be separated into various optical paths (for example, 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 can then be directed onto the specimen. Multiple illumination channels may be configured to direct light onto the specimen simultaneously or at different time points (for example, when sequentially illuminating the specimen using different illumination channels). In another example, the same illumination channel may be configured to direct light having different characteristics onto the specimen at different time points. For example, optical element 18 may be configured as a spectral filter, and the properties of the spectral filter can be changed in a wide variety of ways (for example, by changing the spectral filter) so that light of different wavelengths can be directed onto the specimen at different time points. The illumination subsystem may have any other suitable configuration known in the art for sequentially or simultaneously directing light having different or the same characteristics onto the specimen at different or the same angle of incidence.

[0028] 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 specimen may include broadband light. However, the light source may include any other suitable light source, such as any suitable laser known in the art, configured to generate light at any suitable wavelength known in the art. In addition, the laser may be configured to generate monochromatic or nearly monochromatic light. Thus, the laser may be a narrowband laser. The light source may also include a polychromatic light source that generates light at multiple discrete wavelengths or wavelength bands.

[0029] Light from the optical element 18 can be focused to the beam splitter 21 by the lens 20. In Figure 1, the lens 20 is shown as a single refractive optical element, but in practice, the lens 20 can include several refractive optical elements and / or reflective optical elements that work together to focus light from the optical elements onto the wafer. The illumination subsystem shown in Figure 1 and described herein may include any other suitable optical elements (not shown). Examples of such optical elements include, but are not limited to, polarization components, spectral filters, spatial filters, reflective optical elements, apodizers, beam splitters, apertures, etc., and these may include any suitable optical elements known in the art. In addition, the system can be configured to modify one or more elements of the illumination subsystem based on the type of illumination used for inspection, measurement, etc.

[0030] The imaging system may also include a scanning subsystem configured to scan light across the specimen. For example, the imaging system may include a stage 22 on which the specimen 14 is placed during inspection, measurement, etc. The scanning subsystem may include any suitable mechanical and / or robotic assembly (including the stage 22) that can be configured to move the specimen so that light can be scanned across it. In addition, or alternatively, the imaging system may be configured so that one or more optical elements of the imaging system scan some light across the specimen. The light can be scanned across the specimen in any suitable form.

[0031] The imaging system 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 a specimen resulting from the imaging system's illumination of the specimen and to generate an output in response to the detected light. For example, the imaging system shown in Figure 1 includes two detection channels, one formed by a condenser 24, an element 26, and a detector 28, and the other formed by a condenser 30, an element 32, and a detector 34. As shown in Figure 1, the two detection channels are configured to collect and detect light at different collection angles. In some cases, one detection channel is configured to detect specularly reflected light, and the other detection channel is configured to detect light that is not specularly reflected (e.g., scattered, diffracted, etc.) from the specimen. However, two or more of the detection channels may be configured to detect the same type of light (e.g., specularly reflected light) from the specimen. Although Figure 1 shows an embodiment of an imaging system including two detection channels, the imaging system may include a different number of detection channels (e.g., one detection channel or two or more detection channels). In Figure 1, each light condenser is shown as a single refractive optical element, but each light condenser may include one or more refractive optical elements and / or one or more reflective optical elements.

[0032] One or more detection channels may include any suitable detector known in the art, such as photomultiplier tubes (PMTs), charge-coupled devices (CCDs), and time-delay integral (TDI) cameras. Detectors may also include non-imaging detectors or imaging detectors. If the detectors are non-imaging detectors, each detector may be configured to detect specific properties of scattered light, such as intensity, but not necessarily as a function of position in the imaging plane. Thus, the output produced by each detector included in each detection channel may be a signal or data, not an image signal or image data. In such an example, a computer system, such as the system's computer system 36, may be configured to generate an image of the sample from the non-imaging outputs of the detectors. However, in other examples, the detectors may be configured as imaging detectors configured to generate imaging signals or image data. Thus, the system may be configured to generate images in several ways.

[0033] Figure 1 is provided herein to schematically illustrate the configuration of an imaging system that may be included in the system embodiments described herein. Naturally, the configuration of the imaging system described herein may be modified to optimize the system's performance, as is commonly done when designing commercial inspection systems, measurement systems, etc. In addition, the systems described herein may be implemented using existing inspection or weighing systems, such as the 29xx and 39xx series tools, the SpectraShape family of tools, and the Archer series of tools, which are commercially available from KLA Corp. in Milpitas, California (for example, by adding the functions described herein to an existing inspection or weighing system). For some such systems, the embodiments described herein may be provided as optional functions of an inspection or weighing system (for example, in addition to other functions of an inspection or weighing system). Alternatively, the imaging systems described herein may be designed "from scratch" to provide a completely new inspection or weighing system.

[0034] The system's computer system 36 may be coupled to the detector of the imaging system in any suitable manner (for example, via one or more transmission media, which may include "wired" and / or "wireless" transmission media) so that the computer system can receive the output generated by the detector during scanning of the specimen. The computer system 36 may be configured to perform several functions using the output of the detector as described herein, and any other functions further described herein. The computer system may be further configured as described herein.

[0035] This computer system (and other computer systems described herein) may also be referred to herein as a computer subsystem. Each of the computer subsystems or systems described herein can take various forms, including personal computer systems, image computers, mainframe computer systems, workstations, network equipment, internet equipment, or other devices. Generally, the term “computer system” can be broadly defined to include 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 with high-speed processing and software, whether standalone or network-connected.

[0036] If the system includes two or more computer systems, different computer subsystems can be coupled to one another so that images, data, information, instructions, etc., can be transmitted between them, as will be further described herein. For example, computer system 36 can be coupled to computer system 102 by any suitable transmission medium (as shown by the dashed line in Figure 1), which may include any suitable wired and / or wireless transmission medium known in the art. Two or more such computer subsystems can also be coupled in effect by a shared computer-readable storage medium (not shown).

[0037] Although the imaging system has been described above as an optical or light-based system, the imaging system may also be an electron beam-based system. For example, in one embodiment, the energy directed to the specimen includes electrons, and the energy detected from the specimen includes electrons. Thus, the energy source may be an electron beam source. In one such embodiment shown in Figure 2, the imaging system includes an electron column 122 coupled to a computer system 124.

[0038] As shown in Figure 2, the electron column includes an electron beam source 126 configured to generate electrons focused onto a sample 128 by one or more elements 130. The electron beam source may include, for example, a cathode source or an 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 selecting aperture, an objective lens, and a scanning subsystem, all of which may include any suitable elements known in the art.

[0039] Electrons returning from the sample (e.g., secondary electrons) can be focused to the 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 as the scanning subsystem included in element 130.

[0040] The electronic column may include any other suitable elements known in the art. In addition, the electronic column may be further configured as described in U.S. Patent No. 8,664,594 issued April 4, 2014 by Jiang et al., No. 8,692,204 issued April 8, 2014 by Kojima et al., No. 8,698,093 issued April 15, 2014 by Gubbens et al., and No. 8,716,662 issued May 6, 2014, which are incorporated herein by reference as being fully described.

[0041] In Figure 2, the electron column is shown configured so that electrons are directed towards the specimen at a tilted incidence angle and scattered from the specimen at another tilted angle. However, it should be understood that the electron beam may be directed towards the specimen at any appropriate angle and scattered from the specimen. In addition, the electron beam system may be configured to generate images of the specimen using multiple modes (e.g., using various illumination angles, acquisition angles, etc.). Multiple modes of the electron beam system may have different image generation parameters for the system.

[0042] As described above, the computer system 124 can be coupled to the detector 134. The detector can form an electron beam image of the specimen by detecting electrons returning from the surface of the specimen. The electron beam image may include any suitable electron beam image. The computer system 124 can be configured to perform any of the functions described herein using the output of the detector and / or the electron beam image. The computer system 124 can be configured to perform any additional steps described herein. The system including the imaging system shown in Figure 2 can be further configured as described herein.

[0043] Figure 2 is provided herein to schematically illustrate the configuration of an electron beam-based imaging system that may be included in the embodiments described herein. Similar to the optical systems described above, the electron beam system configurations described herein may be modified to optimize the system's performance, as is typically done when designing commercial inspection or metering systems. In addition, the systems described herein may be implemented using existing inspection, metering, or high-resolution defect review systems, such as tools commercially available from KLA (for example, by adding the functions described herein to an existing inspection, metering, or defect review system). For some such systems, the embodiments described herein may be provided as optional functions of the system (for example, in addition to other functions of the system). Alternatively, the systems described herein may be designed "from scratch" to provide a completely new system.

[0044] Although the imaging system has been described above as either an optical or electron-based system, the imaging system may also be an ion beam system. Such an imaging system can be configured as shown in Figure 2, except that the electron beam source can be replaced with any suitable ion beam source known in the art. In addition, the imaging system may be any other suitable ion beam-based imaging system, such as a commercially available focused ion beam (FIB) system, a helium ion microscope (HIM) system, or an ion beam-based imaging system included in a secondary ion mass spectrometry (SIMS) system.

[0045] The imaging systems described herein can be configured to generate a sample output, such as an image, in multiple modes. Generally, a “mode” is defined by the value of a parameter of the imaging system used to generate an image of the sample (or an output used to generate an image of the sample). Thus, different modes may differ in the value of at least one of the parameters of the imaging system (other than the position on the sample where the output is generated). In this way, an image can be generated by the imaging system using two or more different values ​​of the parameters of the imaging system. For example, in an optical imaging system, different modes may use light of different wavelengths for illumination. Different modes may have different illumination wavelengths (e.g., by using different light sources, different spectral filters, etc.), as will be further described herein. In addition, as mentioned above, the imaging system may include two or more illumination channels. Thus, different illumination channels may be used for different modes. Modes may differ in any one or more modifiable parameters of the imaging system (e.g., polarization, angle, wavelength of illumination, polarization, angle, wavelength of detection, etc.).

[0046] Similarly, images generated by an electron beam imaging system may include images generated by the electron beam imaging system with two or more different values ​​of the electron beam imaging system's parameters. Multiple modes of the electron beam imaging system can be defined by the values ​​of the electron beam imaging system's parameters used to generate images on a sample. Therefore, different modes may differ in at least one value of the electron beam parameters of the electron beam imaging system. For example, different modes may use different illumination incidence angles.

[0047] The embodiments of imaging systems described herein may be configured for inspection, measurement, defect review, or other quality control-related processes performed on specimens. For example, the embodiments of imaging systems described herein and shown in Figures 1 and 2 may have one or more parameters modified to provide different imaging capabilities depending on the application in which they are used. In one such example, the imaging system shown in Figure 1 may be configured to have higher resolution when used for defect review or measurement rather than inspection. In other words, the embodiments of imaging systems shown in Figures 1 and 2 describe several general and diverse configurations of imaging systems, which are adjustable in several ways that will be obvious to those skilled in the art, and can produce imaging systems with different imaging capabilities that are better or worse suited to different applications.

[0048] As described above, the imaging system can be configured to generate an actual image of the physical version of a specimen by directing energy (e.g., light, electrons) to and / or scanning the energy over the physical version of the specimen. In this way, the imaging system can be configured as a “real” imaging system rather than a “virtual” system. However, the storage medium (not shown) and computer system 102 shown in Figure 1 may be configured as a “virtual” system. In particular, the storage medium and computer system are not part of the imaging system 100 and do not have the ability to handle the physical version of a specimen, but can be configured using the stored detector output as a virtual inspector that performs functions such as inspection, a virtual weighing system that performs functions such as weighing, a virtual defect review tool that performs functions such as defect review, and so on. Systems and methods configured as “virtual” systems are described in U.S. Patent No. 8,126,255, assigned to the same applicant and published by Bhaskar et al. on 28 February 2012, U.S. Patent No. 9,222,895, published by Duffy et al. on 29 December 2015, and U.S. Patent No. 9,816,939, published by Duffy et al. on 14 November 2017, which are incorporated herein by reference as being fully described. Embodiments described herein can be further configured as described in those patents. For example, the computer systems described herein can be further configured as described in those patents.

[0049] The system includes one or more computer systems, which may include any of the computer systems further described herein, configured to cluster dies on a sample based on the color assigned to the dies according to the predicted defect density in the dies determined from measurements performed on the sample, thereby generating initial die clusters. The term “predicted” as used herein is synonymous with the terms simulated, estimated, and approximated, and “predicted defect density” is generally defined as the defect density determined, for example, by modeling, from data that is likely to be detected on the sample by the imaging system and is not related to defects on the sample.

[0050] In this way, measurements can be performed on the specimen by one of the imaging systems described herein or any other suitable measuring tool known to those skilled in the art. In another embodiment, the measurement is not performed for the detection of defects on the specimen. For example, any defects present on the specimen may affect the measurement of the specimen, but the measurement is not performed to detect defects on the specimen. Instead, the measurement may include any suitable type of measurement, such as the thickness of one or more films formed on the specimen, characteristics of the patterned mechanism such as dimensions, sidewall angles, roughness, shape, etc., other roughness on the specimen such as film roughness, overlay of mechanisms on one layer of the specimen and mechanisms on another layer of the specimen, and any variation of any such characteristics across the specimen.

[0051] Measurements may include measurements on the layer of the sample on which the process is performed and / or measurements on another layer of the sample. For example, measurements may include measurements of one or more structures formed on a layer below the upper layer, which may be performed in combination with measurements of one or more structures on the upper layer.

[0052] Any of such measurements may be performed using any of the imaging systems described herein or any other suitable measuring tools known to those skilled in the art, in any other appropriate manner known to those skilled in the art. In addition, measurements may be performed by embodiments described herein. However, measurements may be performed by other methods or systems and then acquired by the computer system of the embodiments described herein (for example, from a shared computer-readable medium accessible to both the computer system and other methods or systems).

[0053] Next, these measurements can be used to predict the defect density in the die on the sample. Then, based on the predicted defect density, the die can be assigned a color, and then, based on the assigned color, die clustering can be performed to generate initial die clusters. Each of these steps can be carried out as further described herein.

[0054] While some steps relating to a die on a specimen may be described herein, the steps described herein may also be performed on any other area of ​​the specimen. These areas may correspond to design areas on the specimen other than the die, such as fields. However, these areas may not have any correspondence with the design. For example, a wafer area may be artificially divided into a regular grid of tiles, which may or may not be independent of the design formed on the specimen. For example, a tile may be larger than a repeating pattern area (e.g., a die) on the wafer, or smaller than a repeating pattern area, or a tile may contain parts of two different repeating areas, and so on. A tile may also be a non-overlapping portion of the wafer, i.e., a mutually exclusive portion. In this case, the steps described herein may be performed on the tiles in the grid in the same manner as described herein for the die. The terms “die,” “tile,” and “grid cell” are used interchangeably herein.

[0055] In one embodiment, the measurement is performed on a sample using a measuring tool, and the process includes an inspection process. Thus, the embodiments described herein can determine one or more parameters of the inspection using the measurement results. In other words, the embodiments described herein can be configured for measurement-guided inspection (MGI). As shown in step 600 of Figure 6, for example, the input to the computer system may be a grid with predictions (i.e., predicted defect density) (e.g., a die-level wafer map as shown in Figure 3). Thus, this grid may also be called an MGI grid.

[0056] In one embodiment, the predicted defect density includes a defect density probability distribution. For example, measurement may be performed at a desired number of measurement points on the sample. The measurement results may then be used to determine various distributions of measurement results across the sample, i.e., measurement values ​​as a function of space across the sample. In this way, measurement values ​​can also be used to determine various distributions of defect density probabilities determined from measurement values ​​in the spatial domain of the sample. For example, defect density probabilities can be determined from measurement at various locations on the sample, and then the distribution of defect density probabilities can be determined based on the defect density probabilities as a function of location. The predicted defect density probability distribution can then be used for clustering as described herein. Embodiments may combine clustering and image processing steps, as further described herein, to extract local maximums of the defect density probability distribution on the wafer.

[0057] In another embodiment, the predicted defect density is determined as a continuous distribution of values ​​across the sample. This is a key difference between the embodiments described herein and currently used methods and systems. For example, several clustering algorithms currently in use exist, but none of them are designed to extract local maximums from a continuous distribution (rather than clustering isolated points such as the discrete locations of defects detected on the sample).

[0058] In some embodiments, the predicted defect density is generated by a machine learning (ML) model trained with training data that includes ground truth defect information for one or more training samples and measurements performed on one or more training samples. For example, the defect probability distribution may be calculated using an ML model trained with inspection results as ground truth and one or more measurement data sources as features, such as optical critical dimension (OCD) and patterned wafer shape (PWG) data generated for one or more layers on one or more wafers. In this way, after the ML model has been trained with measurement and ground truth defect information for one or more training samples, the measurements described herein can be input into the ML model to generate the predicted defect density on the samples.

[0059] ML models can have any suitable configuration known in the art and can be trained in any suitable method known to those skilled in the art. Computer systems may or may not be configured to train ML models. For example, as described above, a computer system or another method or system may be configured to train ML models. Training an ML model may include inputting measured values ​​of a training sample into the ML model and modifying one or more parameters of the ML model until the output of the ML model matches (strictly or within some predetermined tolerance) the defect density of ground truth determined for the training sample. The parameters of the ML model that are modified may include any modifiable parameters of the ML model, such as the weights and biases of one or more convolutional layers.

[0060] The embodiments described herein may or may not include an ML model. For example, the ML model may or may not be included in one or more components (not shown) executed by a computer system. When the ML model is executed by a computer system, the computer system can generate a predicted defect density by inputting measurement results, as further described herein, into the ML model.

[0061] In this way, the embodiments described herein can generate predicted defect densities. However, the embodiments described herein may obtain or receive the predicted defect densities simply from another method or system that generated them. In other words, one method or system can be configured to generate predicted defect densities, and another method or system can be configured to use the predicted defect densities as described herein.

[0062] The embodiments described herein can also be performed using predicted defect densities generated by any other method. In other words, the embodiments described herein are not limited to the manner in which the predicted defect density is generated or obtained. For example, as long as the predicted defect density is correct, the embodiments described herein can be used to improve results by pointing out the location of local maximums in the predicted defect density.

[0063] In a further embodiment, the computer system is configured to assign a color to a die by determining a predicted defect density for each die, determining the initial color of the die based on the predicted defect density for each die, and determining the final color of the die by histogram equalization. For example, the computer system may plot the defect density in RGBA color and then enhance the contrast using histogram equalization. RGBA = RBGAlpha, where the alpha parameter is a number from 0.0 (completely transparent) to 1.0 (completely opaque). As shown in step 602, for example, the computer system may be configured to calculate predictions at the die level. Step 602 may include aggregating predictions at the die level by calculating the average of all values ​​belonging to the same die. Thus, the input to this step may be predictions initially calculated on a high-density grid (thousands of values ​​per die), and after this step, each die may have a single aggregated value (e.g., average). The computer system may also be configured to calculate a color vector, e.g., an RGBA vector, from a histogram equalized wafer map, as shown in step 604, for example. Histogram equalization can be performed by any suitable method known in the art.

[0064] The wafer map 300 shown in Figure 3 is a gray-level version of the color die-level wafer map of defect density probabilities, in which four localized clusters (302, 304, 306, and 308) are easily detectable to the human eye. For example, in a black and white diagram, it is difficult to convey how easily these clusters can be detected by a user in a color wafer map. For example, to give an example of how these clusters might appear to a user in a color map, cluster 302 may be bright yellow to yellowish-green, cluster 304 may be yellowish-green and greener than cluster 302, cluster 306 may be bright pale green, and cluster 308 may be a darker pale green than cluster 306. The remaining dies in the map have a relatively dark blue hue, and some dies around clusters 302 and 304 may have a slightly greenish-blue tint.

[0065] Since it is impossible to perform an arbitrary clustering algorithm based on die / grid position, the clustering step of the embodiments described herein focuses on the color space (or defect density probability value) for each die within the grid. As further described herein, embodiments can individually identify all four clusters shown in Figure 3. The wafer map shown in Figure 3 and other wafer maps described herein are included only as examples of wafer maps to aid in understanding the embodiments described herein. Obviously, the wafer maps generated by any embodiment of the systems and methods described herein will vary depending on the specimens on which the steps described herein are performed.

[0066] In one embodiment, clustering is not performed based on the location of the dies on the sample. For example, the clustering step may include obtaining a die-level map of defect density probabilities and assigning the dies to clusters based on the color assigned to each die, rather than on the location information associated with the dies. Each cluster identifies a region of substantially equal levels of predicted defect density, i.e., regions of predicted defect density values ​​that do not differ from one another beyond a predetermined value. In a further embodiment, clustering includes applying the K-means method to the colors assigned to the dies. For example, as shown in step 606 of Figure 6, the computer system may be configured to perform KMeans using color vectors as features. The K-means (or KMeans) method may include any suitable K-means method known in the art, which is performed in any suitable manner known in the art.

[0067] In another embodiment, clustering includes applying K-means to the colors assigned to the dies and selecting the one with the highest silhouette score among the K-means results as one of the initial die clusters. The “silhouette score” is a measure of the “quality” of the clusters generated by the algorithm. A “good quality” set of clusters includes clearly separated clusters, and the components (dies in this example) contained in each of these clusters are relatively close to each other. Thus, the higher the silhouette score, the better the clusters generated by the algorithm. For example, a computer system can use KMeans in this color space to identify clusters by the highest silhouette score. In one such example, as shown in step 608, the computer system can be configured to select the number of KMeans clusters with the highest silhouette score.

[0068] As shown in step 610, the computer system can be configured to loop through steps 606 and 608 across a given number of clusters. The number of clusters can be a predetermined number, such as 2 to 6 clusters. This predetermined number of clusters may be a target number of clusters, which can be determined or set by the user or by any other method known in the art. For example, K-means supports specifying a cluster count, i.e., a target cluster count and / or a maximum cluster count. The predetermined number of clusters can be determined based on how many clusters are expected and how much time is allowed to find them (the more clusters there are, the more time it takes), and this number can be changed as appropriate based on the data.

[0069] Unlike currently used distance-based clustering algorithms, clustering in color space may not be able to decompose spatially separated clusters. This can be seen in the clusters identified by the KMeans algorithm in wafer map 400 shown in Figure 4. The KMeans method identifies two clusters in the wafer map, represented by different shaded areas of the die. For example, one cluster is identified as the die contained in shaded areas 402 and 404, and the other cluster is identified as the die contained in shaded areas 406, 408, and 410. Thus, as can be seen in this figure, color-based clustering may result in several clusters containing dies that are spatially separated from each other on the map. The cluster containing areas 406, 408, and 410 corresponds to similar colors (probabilities) in the original wafer map, but should be identified as different clusters if the goal is to identify local maximums for diversity sampling. Similarly, the cluster containing areas 402 and 404 in color space is actually two clusters in position space.

[0070] The computer system is also configured to analyze the initial die clusters in position space to determine whether any of the initial die clusters contains two or more die clusters. In this way, each of the initial die clusters can be analyzed separately to determine whether it contains two or more die clusters. For example, as shown in step 612 of Figure 6, the computer system may be configured to filter grid cells (or dies) into one cluster at a time by KMeans cluster ID. This step may be performed solely to separate the grid cells into their initial die clusters for analysis purposes. In addition, the computer system may be configured to subsample the filtered grid cells, as shown in step 614. This subsampling may be performed randomly or when time is a factor. For example, if the algorithm used in the analysis step does not scale linearly and there are time constraints (runtime of several seconds to several minutes per iteration), the subsampling step may be performed to help the analysis step run fast enough. Therefore, the subsampling step is optional.

[0071] In one embodiment, the analysis step includes applying a noisy application density-based spatial clustering (DBScan) method to the initial die clusters. Thus, the computer system can use DBScan in spatial space to identify clusters within each previously identified cluster. For example, as shown in step 616, the computer system may be configured to perform DBScan using grid positions (e.g., x and y positions) as features. Therefore, the embodiments described herein may use DBScan in spatial space after KMeans in color space (which may be contrast-enhanced). The DBScan method can include any suitable data clustering algorithm or technique known in the art. In addition, while DBScan may be one particularly advantageous method for performing the analysis step, the analysis of the initial die clusters can be performed using any other spatial-based clustering algorithm or method.

[0072] In another embodiment, the analysis step includes applying the DBScan method to the initial die cluster multiple times using different parameters of the DBScan method, identifying the result with the highest silhouette score among the results of the DBScan method applied with each of the different parameters, and determining, based on the identified result, whether any of the initial die clusters contain two or more die clusters. For example, a computer system can select the DBScan parameter that has the highest silhouette, as shown in step 618. The computer system can also perform a loop on the DBScan parameters, as shown in step 620, thereby performing steps 616 and 618 with different DBScan parameters. These steps may be performed to iteratively find the parameter that gives the best cluster (the cluster with the highest silhouette score) among the DBScan clustering algorithms by brute force. Thus, for each parameter set, DBScan can be performed and the cluster quality can be calculated. Next, using the results generated for each parameter set, the parameter with the best score can be selected, and then the analysis step can be performed using the results generated with the selected parameter.

[0073] After the DBScan step has been performed for one cluster, the computer system can loop through the KMeans clusters, as shown in step 622, thereby performing steps 612, 614, 616, 618, and 620 for the next cluster identified by KMeans. In this way, steps 612, 614, 616, 618, and 620 may be performed independently for each cluster found by KMeans or other initial clustering performed in the color space. For example, if the KMeans execution finds six clusters, the loop can be performed six times. In this way, these steps are performed for each KMeans cluster found in the color space, thereby allowing each KMeans cluster to be partitioned in the position space by DBScan as appropriate. In the example shown in Figure 4, the kidney-shaped cluster 402 and the shrimp-shaped cluster 404 may be the same cluster based on color, but are quite far apart in position space and therefore ultimately two different clusters. Thus, they may be in the same cluster after KMeans but become different clusters after DBScan.

[0074] In one such embodiment, prior to the determination step, the analysis includes excluding any of two or more die clusters having fewer dies than a predetermined die count from the results identified for any of the initial die clusters. For example, as shown in step 624 of Figure 6, the computer system can filter out clusters with a relatively small grid cell count. This step can be performed to avoid “background clusters”. The predetermined die count or relatively small grid cell count can be determined in any suitable manner, for example, based on user input, or can be determined dynamically, for example, based on the number of dies or grid cells in the initial die cluster. In this way, the DBScan method results with the highest silhouette score for the initial die cluster can be examined to determine whether any of the clusters in the results have fewer dies or grid cells than a predetermined die or grid cell count. Then, any of these clusters can be excluded from the die cluster in the DBScan results for that initial die cluster. In this way, any relatively small clusters identified by the DBScan results may not be separated from the initial die cluster into a different die cluster, or they may be assigned to background clusters or groups that are not designated as the final die cluster. By excluding extremely small clusters identified by DBScan, the inspection process can be simplified, and throughput can be improved with relatively little or no negative impact on defect detection performance.

[0075] In one embodiment, clustering and analysis steps (e.g., KMeans and DBScan) are performed such that the initial die clusters, and any two or more die clusters contained within any of the initial die clusters, correspond to the local maximum values ​​of the predicted defect density. For example, the embodiments described herein adapt a clustering algorithm to find the local maximum value clusters in the defect density probability distribution of the wafer. The embodiments described herein enable a computer system to resolve the local maximum values ​​as shown in the wafer map 500 of Figure 5. In particular, all original clusters are precisely identified. More specifically, the final die clusters 502, 504, 506, and 508 shown in wafer map 500 correspond to the local maximum values ​​302, 304, 306, and 308 in wafer map 300 shown in Figure 3, respectively. In addition, the final die clusters may include clusters 510 that are not clearly visible in wafer map 300 but are more clearly visible in a color version of the die-level map of defect density probabilities and can be identified by the clustering and analysis steps described herein.

[0076] The computer system is further configured to designate initial die clusters that do not contain two or more die clusters, and two or more die clusters contained within any of the initial die clusters, as final die clusters. In other words, if DBScan determines that an initial die cluster does not contain two or more die clusters, it can identify that initial die cluster as a final die cluster. If DBScan determines that an initial die cluster contains two or more die clusters, these two or more die clusters are identified as final die clusters, and the original initial die cluster is not identified as a final die cluster. As shown in step 626 of Figure 6, the computer system can assign cluster IDs to all grid cells. Thus, the designation step can include assigning cluster IDs to all grid cells regardless of whether the grid cells are in the original initial die cluster or in different die clusters identified by DBScan. In addition, any non-clustered grid cells in the wafer map can also be assigned cluster IDs indicating that they are non-clustered or background cells. Cluster IDs can have any suitable form (e.g., numeric, alphanumeric, etc.) and format known to those skilled in the art.

[0077] The computer system may also generate an output, as shown in step 628, which may include the final cluster ID of the grid cell and any other information relating to the final die cluster generated by the embodiments described herein. The generated output may have any suitable form or format known in the art that allows the output to be used for process setup as further described herein. The computer system may also be configured to store information relating to the final die cluster for use in setting up a process to be performed on a sample, which may be carried out as further described herein.

[0078] In one embodiment, the computer system is configured to determine the median of the predicted defect density for a final die cluster and to exclude any final die clusters whose median predicted defect density is below a predetermined threshold. For example, the nature of the clustering algorithm described herein means that each die can be part of a cluster regardless of whether it belongs to the local maximum (e.g., cluster 510 shown in Figure 5, and another cluster that may include all dies without shading or patterns in Figure 5, i.e., solid white dies). In the embodiments described herein, it is easy to exclude one or more clusters, such as cluster 510 and the cluster of dies without shading or patterns, by considering the median of each cluster and discarding clusters with a median below a user-defined threshold.

[0079] In another embodiment, the computer system is configured to determine the median predicted defect density for the final die cluster and sort the final die cluster based on the median predicted defect density. For example, the computer system may be configured to rank the final clusters by their median cluster probability, for example, from the highest to the lowest. The ranking step may be performed after the final cluster IDs have been assigned. Thus, this step may be performed after all the other steps shown in Figure 6. This ranking can be used in the setup process described further herein.

[0080] In further embodiments, the computer system is configured to set up the process by determining process sampling based on information about the final die cluster. For example, diversity sampling can be supported using the local maximum cluster identified by the embodiments described herein. Thus, one advantage of the embodiments described herein is that they can be used to find local maximums in the defect density probability distribution on a wafer, which can then be used to help find defects that are not part of the global maximum. In this way, the user of the tool can bias the inspection to search for areas that would normally be ignored due to stronger signals in other areas.

[0081] Therefore, the diversity sampling that can be performed in the embodiments described herein is fundamentally different from conventionally used so-called diversity sampling. For example, conventionally developed diversity sampling methods have worked on a set of defects to find the defects that are most different from each other. In this way, diversity sampling aimed to find the most interesting and most different defects. Such diversity sampling may be useful for many applications, such as defect discovery.

[0082] In contrast, diversity sampling performed by or using the embodiments described herein is based on the diversity of predicted defect densities in different regions of the wafer. For example, the final die cluster may correspond to a local maximum in the predicted defect density, which is the main objective of the embodiments described herein. The sampling for each of the final die clusters can then be determined independently based on their respective predicted defect densities. For example, for a final die cluster expected to have a fairly high defect density, a relatively dense inspection sampling plan may be selected to allow for a more detailed examination of the final die cluster, for example, for diagnostic purposes. In contrast, for a final die cluster expected to have a fairly low defect density, a fairly dense inspection sampling plan may be selected because a less dense sampling plan might miss all or too many relatively sparse defects. Thus, the sampling plans independently selected or generated for each final die cluster may vary, at least in part, depending on the purpose of the inspection (e.g., well-established process monitoring for possible process deviations, versus defect detection for less established processes).

[0083] The embodiments described herein may be performed in post-measurement processing as part of an MGI analysis flow. In addition, this process may be configured to consume cluster IDs and use them for inspection sampling. Thus, the primary application for which the embodiments described herein may be used is to calculate defect probability distributions (from the measurements of the preceding measurement steps) and combine them with the sampling planning capabilities of the inspection tool to help derive better dynamic inspection plans.

[0084] Therefore, the final die cluster can function as a type of CA on the specimen that can be used in place of or in addition to other care areas (CAs) on the specimen. For example, the process parameters used in different areas on the wafer corresponding to the final die cluster may be different as well as the parameters of the CA. In this way, one or more computer systems may be configured to designate some or all of the final die cluster as a CA for processes performed on the specimen using an imaging system. Designating some or all of the final die cluster as a CA may involve generating some CA identification information for the final die cluster, which may include any alphanumeric or other appropriate identification information known in the art. The output of this step may include the CA designation and any other information generated for the CA, such as the location and size of the specimen.

[0085] Instead of (or in addition to) sampling, one or more other parameters of the process can be selected independently for each of the final die clusters. For example, the parameters may include information indicating which final die clusters the process should run on, e.g., which final die cluster inspections are performed on and which are not. The other parameter values ​​selected independently for each final die cluster may include different sensitivities used in the inspection recipe for detecting defects on the sample. Different sensitivities can be controlled, for example, by one or more thresholds used to separate the image into pixels corresponding to defects and pixels that do not correspond to defects. In addition, or instead, the computer system may be configured to optimize the thresholds for each final die cluster to obtain a desired number of defects. Such sensitivities can be set using pre-defined capture rates ("cap rate" or "caprate") for different final die clusters. For example, an inspection target (e.g., desired false alarm rate + desired defect capture rate per defect type) can be specified for each final die cluster.

[0086] The selection of test recipe parameters may be performed separately for different final die clusters, but each of the test recipe parameters selected separately and independently for a final die cluster may be combined into a single test recipe performed on the sample. In addition, the selection of test recipe parameters can be performed simultaneously for multiple final die clusters, thereby optimizing the test recipe parameters for multiple final die clusters.

[0087] In further embodiments, measurements and processes are performed on a specimen using different tools. For example, generally, the systems described herein are configured for either measurement or inspection, but not for both. Thus, the measurements input to the steps described herein may be generated by a tool configured for measurement, and the processes performed using the results of the steps described herein may be performed by different tools configured for inspection. The steps described herein may be performed by any of these tools or without any of them. For example, the embodiments described herein may be performed on the tools by one or more computer systems included in or connected to the measurement tools and / or inspection tools (e.g., computer system 36 shown in Figure 1 or computer system 124 shown in Figure 2), or they may be performed outside the tools by a computer system not included in or directly connected to any of such tools (e.g., computer system 102 shown in Figure 1).

[0088] Thus, the systems described herein may or may not include an imaging system as shown in Figures 1 and 2. If the system includes an imaging system, the same computer system that generates the final die cluster information and, optionally, sets up the process based on that information, may also be configured to perform the setup process on the specimen. If the system does not include an imaging system, another system or method may be configured to perform the setup process on the specimen. In addition, one system may be configured to generate and store the final die cluster information as described herein, another system may be configured to set up the process using the stored information, and yet another system may be configured to perform the setup process.

[0089] A computer system may be configured to store information about the final die cluster for use when setting up processes to be performed on a specimen, such as specimen inspection, and / or for use in executing processes on the specimen after setup. For example, a computer system may be configured to store information by storing it in a recipe or by generating a recipe for a process in which the final die cluster is used. As used herein, the term “recipe” can generally be defined as a set of instructions that a tool can use to execute a process on a specimen. Thus, generating a recipe may involve generating information about how to execute a process, which can then be used to generate instructions for executing that process. Information about the final die cluster stored by a computer system may include any information that can be used to identify and / or use the final die cluster (e.g., a file name and its storage location, the file may include information about the final die cluster such as the final die cluster ID and the location of the final die cluster).

[0090] A computer system can be configured to store information about the final die cluster in any suitable computer-readable storage medium. The information can be stored together with any of the results described herein and can 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 information is stored, it can be accessed on the storage medium, used by any of the methods or system embodiments described herein, formatted for display to a user, used by another software module, method or system, etc. For example, an embodiment described herein can generate the inspection recipe described above. Then, a system or method (or another system or method) can store and use this inspection recipe to inspect a specimen and thereby generate information about the specimen (e.g., defect information).

[0091] A computer system and / or imaging system may be configured to perform an inspection process on a specimen using the results of one or more steps described herein. The inspection process can generate results regarding any defects detected on the specimen, such as information on the bounding box of the detected defect, such as its location, information on defect classification such as a detection score, class label, or ID, or any other appropriate information known in the art. These defect results can be generated by the computer system and / or imaging system in any appropriate manner. The defect results can have any appropriate format or type, such as a standard file type. The computer system and / or imaging system may generate and store these results so that they can be used by the computer system and / or another system or method to perform one or more functions on the specimen or another specimen of the same kind. For example, this information can be used by the computer system or another system or method for defect review or sampling of defects for other analysis, identification of the root cause of defects, etc.

[0092] The functions that can be performed using such information include, but are not limited to, modifying processes such as manufacturing processes or steps that have been performed or will be performed on a specimen being inspected or another specimen in a feedback or feedforward manner. For example, a computer system may be configured to identify one or more changes to processes that have been performed on a specimen inspected as described herein, and / or processes that will be performed on the specimen based on the defects detected. Changes to processes may include any appropriate changes to one or more parameters of the processes. Preferably, the computer system identifies changes that can reduce or prevent defects in other specimens on which the modified process is performed, the defects can be corrected or eliminated on the specimen in another process performed on the specimen, the defects can be compensated for in another process performed on the specimen, and so on. The computer system may determine such changes in any appropriate manner known in the art. Such changes may be determined using the results of other processes as described herein.

[0093] These changes can then be transmitted to a semiconductor manufacturing system (not shown) or storage medium (not shown) that has access to the computer system and the semiconductor manufacturing system. The semiconductor manufacturing system may or may not be part of the system embodiments described herein. For example, the computer system and / or imaging system described herein may be coupled to the semiconductor manufacturing system via one or more common elements such as a housing, power supply, sample handling device, or mechanism. The semiconductor manufacturing system may include any semiconductor manufacturing system known in the art, such as lithography tools, etching tools, chemical mechanical polishing (CMP) tools, and deposition tools.

[0094] Therefore, as described herein, new processes or recipes can be set up using the embodiments. The embodiments can also be used to modify existing processes or recipes, whether they are processes or recipes used for a specimen, or processes or recipes created for one specimen and adapted for another specimen.

[0095] Each of the above embodiments of the system can be combined into a single embodiment.

[0096] Another embodiment relates to a computer implementation method for generating information used to set up a process performed on a specimen. The method includes the clustering, analysis, designation, and storage steps described above. Each step of the method may be performed as further described herein. The method may also include any other steps that can be performed by the imaging system and / or computer system described herein. The clustering, analysis, designation, and storage steps are performed by one or more computer systems, which may be configured according to any of the embodiments described herein. In addition, the method described above may be performed by any of the system embodiments described herein.

[0097] Additional embodiments relate to a non-temporary computer-readable medium that stores program instructions executable on a computer system for executing a computer implementation method for generating information used to set up a process to be performed on a specimen. One such embodiment is shown in Figure 7. In particular, as shown in Figure 7, the non-temporary computer-readable medium 700 includes program instructions 702 executable on a computer system 704. The computer implementation method may include any step of any method described herein.

[0098] Program instructions 702 that perform the methods described herein may be stored in a computer-readable medium 700. The computer-readable medium may be a storage medium such as a magnetic disk or optical disk, a magnetic tape, or any other suitable non-temporary computer-readable medium known in the art.

[0099] Program instructions can be implemented in any of the many methods available, including procedure-based techniques, component-based techniques, and / or object-oriented techniques. For example, program instructions can be implemented using ActiveX controls, C++ objects, JavaBeans, Microsoft Foundation Classes ("MFC"), SSE (Streaming SIMD Extensions), or other techniques or methods, as desired.

[0100] The computer system 704 can be configured according to any of the embodiments described herein.

[0101] By taking into consideration this specification, further modifications and alternative embodiments of various aspects of the invention will become apparent to those skilled in the art. For example, methods and systems for generating information for use in setting up processes to be performed on specimens are provided. Thus, this specification should be construed as merely illustrative and is intended to teach those skilled in the art general ways of carrying out the invention. It is understood that the forms of the invention depicted and described herein are preferred embodiments at present. Elements and materials may be replaced with those illustrated and described herein, parts and processes may be reversed, and certain features of the invention may be used independently, all of which will be apparent to those skilled in the art who have benefited from the description of the specification of the invention. Modifications to the components described herein may be made without departing from the spirit and scope of the invention, as described in the following claims.

Claims

1. A system configured to generate information used to set up a process to be performed on a sample, The dies on the sample are clustered based on the color assigned to the die according to the predicted defect density within the die determined from measurements performed on the sample, thereby generating initial die clusters. In order to determine whether any of the initial die clusters contains two or more die clusters, the initial die clusters in spatial space are analyzed, The initial die cluster that does not contain two or more die clusters, and the two or more die clusters that are included in any of the initial die clusters, are designated as the final die cluster. To store the information of the final die cluster for use in setting up the process to be executed on the aforementioned sample, One or more computer systems configured to perform A system characterized by comprising the following features.

2. The system according to claim 1, wherein the predicted defect density includes a defect density probability distribution.

3. The system according to claim 1, characterized in that the predicted defect density is determined as a continuous distribution of values ​​across the sample.

4. The system according to claim 1, characterized in that the clustering and analysis are performed such that the initial die cluster and any two or more die clusters included in any of the initial die clusters correspond to the local maximum value of the predicted defect density.

5. A system according to claim 1, wherein the measurement is performed on the specimen using a measuring tool, and the process includes an inspection process.

6. The system according to claim 1, wherein the measurement is not performed for the purpose of detecting defects on the sample.

7. The system according to claim 1, characterized in that the predicted defect density is generated by a machine learning model trained using training data including ground truth defect information relating to one or more training samples and the measurements performed on the one or more training samples.

8. A system according to claim 1, wherein the one or more computer systems are further configured to assign a color to the die by determining a predicted defect density for each die, determining the initial color of the die based on the predicted defect density for each die, and determining the final color of the die by histogram equalization.

9. The system according to claim 1, wherein the clustering includes applying the K-means method to the colors assigned to the die.

10. A system according to claim 1, wherein the clustering includes applying the K-means method to the colors assigned to the dies, and selecting the one with the highest silhouette score among the results of the K-means method as one of the initial die clusters.

11. The system according to claim 1, characterized in that the clustering is not performed based on the position of the die on the sample.

12. The system according to claim 1, wherein the analysis includes applying a density-based spatial clustering (DBScan) method for noisy applications to the initial die cluster.

13. A system according to claim 1, wherein the analysis includes applying a density-based spatial clustering (DBScan) method for a noisy application to the initial die cluster multiple times using different parameters of the DBScan method; identifying the result having the highest silhouette score among the results of the DBScan method applied using each of the different parameters; and determining whether any of the initial die clusters contain two or more die clusters based on the identified result.

14. The system according to claim 13, wherein, prior to the determination, the analysis further comprises excluding from the identified results for any of the initial die clusters any of the two or more die clusters having fewer dies than a predetermined die count.

15. A system according to claim 1, wherein one or more computer systems are further configured to determine the median of the predicted defect density for the final die cluster and to exclude any of the final die clusters whose median of the predicted defect density is less than a predetermined threshold.

16. A system according to claim 1, wherein one or more computer systems are further configured to determine the median of the predicted defect density for the final die cluster and to sort the final die cluster based on the median of the predicted defect density.

17. A system according to claim 1, wherein one or more computer systems are further configured to set up the process by determining the sampling of the process based on the information relating to the final die cluster.

18. A system according to claim 1, wherein the measurement and the process are performed on the specimen using different tools.

19. A non-temporary computer-readable medium storing program instructions executable on a computer system for executing a computer implementation method for generating information used to set up a process to be executed on a specimen, wherein the computer implementation method is The dies on the sample are clustered based on the color assigned to the die according to the predicted defect density within the die determined from measurements performed on the sample, thereby generating initial die clusters. In order to determine whether any of the initial die clusters contains two or more die clusters, the initial die clusters in spatial space are analyzed, The initial die cluster that does not contain two or more die clusters, and the two or more die clusters that are included in any of the initial die clusters, are designated as the final die cluster. To store the information of the final die cluster for use in setting up the process to be executed on the aforementioned sample, A non-temporary computer-readable medium characterized by including [a specific element].

20. A computer implementation method for generating information used to set up a process to be performed on a sample, The dies on the sample are clustered based on the color assigned to the die according to the predicted defect density within the die determined from measurements performed on the sample, thereby generating initial die clusters. In order to determine whether any of the initial die clusters contains two or more die clusters, the initial die clusters in spatial space are analyzed, The initial die cluster that does not contain two or more die clusters, and the two or more die clusters that are included in any of the initial die clusters, are designated as the final die cluster. To store the information of the final die cluster for use in setting up the process to be executed on the aforementioned sample, A method comprising the clustering, analysis, designation, and storage, wherein the clustering, analysis, designation, and storage are performed by one or more computer systems.

Images