Method for calibrating, predicting and controlling random defects in extreme ultraviolet lithography

Abstract

One or more defects within an inspection area of a workpiece are imaged using an optical tool or an electron beam tool, using an initial probability of occurrence of a random defect within the inspection area. The model is used to generate a probability of occurrence of a random defect at each of the defect locations. And grouping the defect positions into probability grids. Consistency between the initial probability and observations is determined, and the model may be tuned based on the consistency.

Description

[0001] Cross-reference of related applications

[0002] This application claims priority to provisional patent application No. 63 / 540,605, filed and assigned on September 26, 2023, the disclosure of which is hereby incorporated by reference. Technical Field

[0003] This disclosure relates to the measurement of random defects during semiconductor manufacturing. Background Technology

[0004] The evolution of the semiconductor manufacturing industry places increasingly higher demands on yield management, and specifically on metrology and inspection systems. Critical dimensions continue to shrink, but the industry needs to reduce the time spent achieving high-yield, high-value production. Minimizing the total time from detecting a yield problem to resolving it maximizes the semiconductor manufacturer's return on investment.

[0005] Semiconductor devices, such as logic and memory devices, typically involve processing semiconductor wafers using numerous manufacturing processes to form the various features and multiple layers of the semiconductor device. For example, photolithography is a semiconductor manufacturing process involving transferring a pattern from a photomask to a photoresist disposed on a workpiece, such as a semiconductor wafer. Additional examples of semiconductor manufacturing processes include, but are not limited to, chemical mechanical polishing (CMP), etching, deposition, and ion implantation. An arrangement of multiple semiconductor devices fabricated on a single semiconductor wafer can be separated into individual semiconductor devices.

[0006] Inspection processes are used at various stages of semiconductor manufacturing to detect defects on wafers, thereby promoting higher yields and, consequently, higher profits. Inspection has always been a crucial part of manufacturing semiconductor devices such as integrated circuits (ICs). However, as the size of semiconductor devices shrinks, inspection becomes even more critical for the successful manufacture of acceptable semiconductor devices because even small defects can cause device failure. For example, as the size of semiconductor devices decreases, the detection of smaller defects has become necessary because even relatively small defects can still cause undesirable aberrations in the semiconductor device.

[0007] Defect re-inspection typically involves re-inspecting defects detected during the inspection process and using high-magnification optical systems or scanning electron microscopy (SEM) to generate additional information about the defects at higher resolution. Defect re-inspection is usually performed at discrete locations on samples that have already passed inspection and been found to have defects. The higher-resolution data of defects generated by defect re-inspection is better suited for determining defect properties such as profile, roughness, or more accurate size information.

[0008] Photolithography can exhibit defects driven by the quantization properties of light and materials. For example, light is quantized into photons, and the chemical reactants in the photoresist are discrete molecules. These are often referred to as shot noise defects or random defects. These random defects may be prevalent in extreme ultraviolet (EUV) lithography but can also occur at exposure wavelengths used in other lithography processes, such as ArF immersion. “Random” means that the average behavior can be within desired specifications (e.g., photoresist width, tip-to-tip measurement of line ends, or photoresist thickness) while exhibiting fluctuations that cause the pattern to fail with a non-zero probability (e.g., bridging or breakage of line / space patterns). Given that the workpiece contains billions of transistors, even a small probability of failure can lead to huge yield losses.

[0009] Random defects can present multiple challenges in a manufacturing environment. Typically, defects can be assumed to be deterministic, meaning that known defects will always be present when manufacturing is carried out according to a known production recipe containing the pattern and exposure parameters of the element to be manufactured on the sample. For example, process window compliance (PWQ) typically identifies process-limiting defects that always occur when exposure conditions exceed the process window. In this example, the process window may define a limit on defocusing associated with the sample's position along the optical axis of the lithography tool (e.g., the sample's focal point) or on the energy dose from the illumination source incident on the sample during exposure.

[0010] EUV lithography, used in the manufacture of high-end semiconductor devices, can lead to defects in the workpiece, which tend to be random in nature. These defects can occur immediately after the development of the photoresist image or further downstream, such as during etching processes. Mathematical models have been used to predict the behavior of these random defects to aid in manufacturing process optimization. Generalized empirical models were previously built primarily based on experimental data and sometimes with physical constraints. These generalized empirical models rely on extensive defect rate data from optical inspection and SEM re-inspection to build and calibrate the models.

[0011] Experimental SEM data is also used to build empirical models. These typically build models with less data by leveraging geometric symmetry on design features (e.g., 1D line / space features or contact hole arrays). However, these empirical models tend to be applicable only to specific design features and are therefore impractical for general use across complex design layouts.

[0012] Accelerated rigorous models are built to simulate the results of a full 3D rigorous physics simulator. These models typically require less experimental data to build. Often, lower-fidelity / high-precision experimental data is sufficient to build these models, rather than high-resolution SEM data. Additional experimental data is typically used to calibrate / fine-tune the model to align it with the observed defect rate.

[0013] Empirical models are often applicable to specific design features, making their use across complex design layouts impractical. Collecting the experimental datasets needed to build and calibrate these models is time-consuming and expensive. Calibration of these models may require extensive curation of the experimental datasets to improve prediction quality, making it difficult to implement. Building large, curated datasets capable of calibrating models to predict sufficiently low-probability hotspots across a sufficiently wide range of complex designs can also be challenging or impractical. Techniques relying on experimental observations of defects under non-nominal operating conditions and extrapolating them to nominal operating conditions may be potentially erroneous and carry the risk of under- or over-predicting defect-free process windows. Improved systems and techniques are needed. Summary of the Invention

[0014] In a first embodiment, a method is provided. One or more steps of the method may use a processor. The method includes receiving, at the processor, an initial probability of random defects occurring within an inspection area of ​​a workpiece. A model is used to generate the initial probability of random defects occurring. One or more defects within the inspection area are imaged using an optical tool or an electron beam tool. Each defect is located at a defect location. The model is used to generate the probability of random defects occurring at each of the defect locations. The defect locations are grouped into probability grids. The consistency between the initial probabilities and the observations is determined. The model is tuned based on the consistency.

[0015] The method may include determining the desired resolution of the probability prediction for each of the probability lattices before using the grouping of the processor.

[0016] The method may include using the processor to determine the expected defect count within each of the probability lattices.

[0017] The consistency can be determined using binary cross-entropy, the expected count of observations (RMSe), a binomial test of significance, or the Brier score.

[0018] In this example, the imaging uses the electron beam tool. The imaging can occur over multiple workpiece exposures. The method may include determining the defect frequency based on defect counts. The defect locations may be grouped according to the geometric pattern shape on the workpiece.

[0019] The method may include using the processor to apply the model to generate the initial probability of random defects occurring.

[0020] An embodiment of a non-transitory computer-readable medium that stores a program may be configured to instruct the processor to perform the method.

[0021] In a second embodiment, a system is provided. The system includes an inspection tool configured to image a workpiece and a processor in electronic communication with the inspection tool. The processor is configured to: receive an initial probability of random defects occurring within an inspection area of ​​the workpiece; send instructions to use the inspection tool to image one or more defects within the inspection area; use a model to generate a probability of random defects occurring at each defect location; group the defect locations into probability grids; determine the consistency between the initial probabilities and observations; and tune the model based on the consistency. The initial probability of random defects occurring is generated using the model. Each defect is located at a defect location.

[0022] The testing tool can be an optical tool or an electron beam tool.

[0023] The processor may be further configured to determine the desired resolution of the probability prediction for each of the probability grids prior to the grouping.

[0024] The processor can be further configured to determine the expected defect count within each of the probability lattices.

[0025] The consistency can be determined using binary cross-entropy, the expected count of observations (RMSe), a binomial test of significance, or a Blair score.

[0026] In this example, the inspection tool is an electron beam tool. The imaging can occur within multiple workpiece exposures. The processor can be further configured to determine the defect frequency based on defect counts. The defect locations can be grouped according to the geometric pattern shape on the workpiece.

[0027] The processor can be further configured to use the model to generate the initial probability of random defects occurring.

[0028] A non-transitory computer-readable storage medium may contain one or more programs for performing the following steps on one or more computing devices: Receiving an initial probability of random defects occurring within a detection area of ​​a workpiece; Generating the initial probability of random defects occurring using a model; Sending instructions to image one or more defects within the detection area using an optical tool or an electron beam tool, each defect being at a defect location; Generating the probability of random defects occurring at each of the defect locations using the model; Grouping the defect locations into probability grids; Determining the consistency between the initial probabilities and observations; Tuning the model based on the consistency. Attached Figure Description

[0029] For a fuller understanding of the nature and purpose of this disclosure, reference should be made to the following detailed description taken in conjunction with the accompanying drawings, wherein:

[0030] Figure 1 Illustrate embodiments of the method according to this disclosure;

[0031] Figure 2 It is a diagram of an exemplary optical tool;

[0032] Figure 3 A diagram of an exemplary electron beam tool; and

[0033] Figure 4 This is an exemplary sample fraction defect frequency map extracted from an electron beam tool image. Detailed Implementation

[0034] While the claimed objectives will be described with reference to specific embodiments, other embodiments (including those not providing all the advantages and features set forth herein) are also within the scope of this disclosure. Various structural, logical, procedural, and electronic changes may be made without departing from the scope of this disclosure. Therefore, the scope of this disclosure is defined only by reference to the appended claims.

[0035] Random defects are typically detected using inspection tools and then characterized by top-down SEM. Inspection and characterization can occur after photolithography (ADI) or after subsequent etching and cleaning steps (AEI or ACI). AEI / ACI results are likely most relevant to yield and depend on the remaining photoresist thickness under ADI. The embodiments disclosed herein can calibrate random defect prediction models, such as accelerated rigorous models requiring fine-tuning. Calibration can be performed with less experimental data and avoids geometric grouping of images. Alternatively, embodiments can utilize grouping of hotspot locations according to the predicted probability of defects. Predicted probabilities across layouts can be handled in a manner best suited to downstream applications such as hotspot inspection / monitoring or hotspot repair.

[0036] Figure 1 An embodiment of method 100 is illustrated. At least some steps of method 100 can be performed using a processor. At 101, an initial probability of random defects occurring within an inspection area of ​​the workpiece is received. This may be based on one or more images of the workpiece. A model can be used to generate the initial probability of random defects occurring. Figure 1 In this diagram, the inspection area is represented by a square indicated by dashed lines. The workpiece can be, for example, a semiconductor wafer or another type of workpiece. Although only a portion of the workpiece is shown, the entire workpiece surface can be analyzed. The inspection area may include tunable parameters.

[0037] The model may comprise a rigorously stochastic lithographic simulation model and / or a transformed model (e.g., through the application of machine learning) that empirically reproduces the stochastic defect probability predictions of the rigorous model while improving computational speed at the cost of a slight decrease in accuracy for certain input samples. The model can be tuned to accommodate the desired trade-offs between the scope of the input space, speed, and accuracy. An exemplary model is described in U.S. Publication No. 2022 / 0129775, the entire contents of which are incorporated herein by reference. Tunable parameters may be part of the model. While the model contains several parameters learned during the training phase, these tunable parameters typically affect the global characteristics of the model's output, such as the number of detected hotspots and the overall magnitude of the predicted defect probabilities.

[0038] In section 102, one or more defects within the inspection area can be imaged using optical or electron beam tools. Each defect is located at its specific location. The processor can send instructions to perform this imaging. Figure 1 In the model, defects are indicated by black circles within the inspection area. Imaging can be performed under nominal or non-nominal dose or focal conditions. Although inspection areas are displayed as having the same size, they can have different sizes. One inspection area can be a sub-region of another area analyzed by the model.

[0039] In step 103, the model is used to determine the probability of a random defect occurring at each of the defect locations. The desired pattern on the photomask can be used as input to the model. In addition to the mask pattern, the model input may also include process parameters such as exposure dose, focal length, source shape, photoresist thickness, etch resistance, etc.

[0040] Next, at step 104, defect locations are grouped into probabilistic grids. Each grid can have a desired grid width, which is the desired resolution for the probabilistic prediction. This grid width can be evenly or non-uniformly spaced within the probability range. An exemplary grid width could be, for example, 0.027 decimal. In one embodiment, if a lower resolution for the probabilistic prediction is required, a larger grid width, for example, 0.33 decimal (log10 probability value), can be used.

[0041] Optionally, the desired resolution for the probability prediction of each in the probability grid can be determined before grouping into 104.

[0042] The expected number of defects in each of the probability grids can be determined. The expected defect count in each grid can be the count multiplied by the probability.

[0043] In step 105, the consistency between the initial probability of using method 100 on a workpiece or a batch of workpieces and the observed results is determined. Figure 1The example graph shows the global ridge count, expected defect count (dashed line), and observed defect count (triangle) with various tolerances. At 106, the model can be tuned based on this consistency. At 106, one or more model parameters can be adjusted to, for example, improve accuracy. After each adjustment of the model parameters, the updated model predictions can be checked. Optimization algorithms can be used to find the optimal parameters that maximize consistency. Consistency can be determined using, for example, binary cross-entropy, the RMSe of expected versus observed counts, a binomial test of significance, or a statistical test measure of Blair scores. Binary cross-entropy can be considered a measure of the difference between two probability distributions. The RMSe of expected versus observed counts is the root mean square of the error. For the RMSe, this error is defined as the difference between the expected count and the observed count.

[0044] In this embodiment, imaging uses an electron beam tool. Imaging can be performed, for example... Figure 4 This occurs during multiple workpiece exposures, as shown in the diagram. Multiple workpiece exposure tests are possible. Defects can be detected from these electron beam tool images, for example, using automatic image alignment, reference image determination, defect detection (e.g., using image subtraction with a reference image), and noise filtering. Defect locations are grouped according to the geometric pattern shape on the workpiece. Repeated tests can be performed by grouping multiple locations with geometric pattern shapes similar to selected locations. This may include automatic pattern recognition to detect and align similar patterns at different locations on the workpiece. The defect frequency of defects can be determined based on defect counts.

[0045] In another embodiment of calibration and verification, predictions from the model (e.g., the probability of random defect occurrence) are made within an inspection region having tunable parameters. Defects are then collected within the inspection region, for example, using an optical system or an electron beam system (e.g., SEM). This can occur under specific dose and focal point conditions (e.g., nominal dose and nominal focal point conditions used during manufacturing). The inspection region may be the same as the modeled inspection region or a subregion of the modeled inspection region.

[0046] The number of defects observed within an inspection region or subregion under specific dose / focus conditions can be counted. The model then provides a predicted defect count within the inspection region or subregion under those specific dose-focus conditions. Defects can be repeatedly collected, counted, and predicted under various dose / focus conditions (e.g., non-nominal conditions that trigger hotspots more frequently). One or more statistical test measures can be used to estimate the consistency between the model and observations using both observed and predicted counts. These measures may include, for example, Pearson correlation coefficient, slope and intercept of least-squares fit, binary cross-entropy, RMSe of expected defect counts, binomial test of significance, or Blair score. The model can then be tuned to obtain an improved or optimal score.

[0047] In another embodiment of calibration and verification, predictions from the model (e.g., the probability of random defect occurrence) are made within an inspection region having tunable parameters. Defects are then collected within the inspection region, for example, using an optical system or an electron beam system (e.g., SEM). This can occur under specific dose and focus conditions (e.g., nominal dose and focus conditions used during manufacturing and / or nominal dose and focus conditions). The inspection region may be the same as the modeled inspection region or a subregion of the modeled inspection region.

[0048] Next, the defect locations and the associated pattern shapes for each defect location can be determined. For each of these pattern shapes, the number of matching examples within the tested region (which may be a sub-region) of the layout is determined. Subsequently, the observed frequency or probability of the defect is determined as the ratio of the number of defects to the number of matching examples for each pattern shape. The average predicted probability of the defect from the model can also be determined for each pattern shape. Where appropriate, all other pattern shapes present within the tested region of the layout where no defects were observed can be assigned a zero observed frequency, while the predicted probabilities of the defects from the model are also extracted and assigned. One or more statistical test measures can be used to estimate the consistency between the model and the observations using counts from observations and predictions. These measures may include, for example, binary cross-entropy, the RMSe of the expected frequency of observed defects for all pattern shapes, a binomial test of significance, or a Blair score. The model can then be tuned to obtain an improved or optimal score.

[0049] Method 100 may use optical tools or electron beam tools (e.g., a BBP tool followed by SEM re-examination). However, certain embodiments of method 100 may use electron beam tools. For example, around Figure 4 The described steps can be performed using an electron beam tool.

[0050] Figure 2 An embodiment of system 200 is shown. System 200 includes an optically based subsystem 201. Generally, the optically based subsystem 201 is configured to generate an optically based output of workpiece 202 by directing light to workpiece 202 (or scanning light across workpiece 202) and detecting the light from workpiece 202. In one embodiment, workpiece 202 includes a wafer. The wafer may include any wafer known in the art. In another embodiment, workpiece 202 includes a photomask. The photomask may include any photomask known in the art.

[0051] exist Figure 2 In the embodiment of system 200 shown, the optical-based subsystem 201 includes an illumination subsystem configured to direct light to workpiece 202. The illumination subsystem includes at least one light source. For example, such as... Figure 2As shown, the lighting subsystem includes a light source 203. In one embodiment, the lighting subsystem is configured to direct light to the workpiece 202 at one or more incident angles, which may include one or more tilt angles and / or one or more normal angles. For example, as Figure 2 As shown, light from light source 203 is guided through optical element 204 at an angle of incidence and then through lens 205 to workpiece 202. The angle of incidence may include any suitable angle of incidence that can vary depending on, for example, the characteristics of workpiece 202.

[0052] The optical-based subsystem 201 can be configured to guide light to the workpiece 202 at different incident angles at different times. For example, the optical-based subsystem 201 can be configured to change one or more characteristics of one or more elements of the illumination subsystem, such that it can guide light to the workpiece 202 at different times. Figure 2 The incident angle shown guides the light to the workpiece 202. In one example, the optical subsystem 201 can be configured to move the light source 203, optical element 204, and lens 205 such that the light is guided to the workpiece 202 at different tilting incident angles or normal (or near-normal) incident angles.

[0053] In some examples, the optical-based subsystem 201 can be configured to guide light to the workpiece 202 simultaneously at more than one incident angle. For example, the illumination subsystem may include more than one illumination channel, one of which may include, for example, Figure 2 The light source 203, optical element 204, and lens 205 shown herein, and another of the illumination channels (not shown), may contain similar elements in different or identical configurations, or may contain at least one light source and possibly one or more other components (such as those further described herein). If this light and another light are simultaneously directed to the sample, then one or more characteristics (e.g., wavelength, polarization, etc.) of the light directed to the workpiece 202 at different incident angles may be different, such that the light generated by illuminating the workpiece 202 at different incident angles can be distinguished from each other at the detector.

[0054] In another example, the lighting subsystem may contain only one light source (e.g., Figure 2The light source 203 shown in the diagram can be split 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 from each of these different optical paths can then be directed to the workpiece 202. Multiple illumination channels can be configured to direct light to the workpiece 202 simultaneously or at different times (e.g., when using different illumination channels to sequentially illuminate a sample). In another example, the same illumination channel can be configured to direct light with different characteristics to the workpiece 202 at different times. For example, in some examples, optical element 204 can be configured as a spectral filter, and the properties of the spectral filter can be changed in various different ways (e.g., by swapping out the spectral filter) so that light of different wavelengths can be directed to the workpiece 202 at different times. The illumination subsystem can have any other suitable configuration known in the art for sequentially or simultaneously directing light with different or the same characteristics to the workpiece 202 at different or the same incident angles.

[0055] In one embodiment, light source 203 may comprise a broadband plasma (BBP) source. In this way, the light generated by light source 203 and directed to workpiece 202 may comprise broadband light. However, the light source may comprise any other suitable light source, such as a laser. The laser may comprise any suitable laser known in the art and may be configured to generate light at any one or more suitable wavelengths known in the art. Additionally, the laser may be configured to generate monochromatic or near-monochromatic light. In this way, the laser may be a narrowband laser. Light source 203 may also comprise a multicolor light source that generates light at multiple discrete wavelengths or bands.

[0056] Light from optical element 204 can be focused onto workpiece 202 by lens 205. Although lens 205... Figure 2 While shown as a single refractive optical element, it should be understood that, in practice, lens 205 may comprise several refractive and / or reflective optical elements that collectively focus light from the optical element onto the sample. Figure 2 The illumination subsystem shown and described herein may include any other suitable optical elements (not shown). Examples of such optical elements include, but are not limited to, polarizing components, spectral filters, spatial filters, reflective optics, apodizers, beam splitters (e.g., beam splitter 213), apertures, and the like, which may include any such suitable optical elements known in the art. Furthermore, the optical-based subsystem 201 may be configured to vary one or more of the elements of the illumination subsystem based on the type of illumination to be used to generate the optical-based output.

[0057] The optical-based subsystem 201 may also include a scanning subsystem configured to cause an optical scan across the workpiece 202. For example, the optical-based subsystem 201 may include a stage 206 on which the workpiece 202 is placed during optical output generation. The scanning subsystem may include any suitable mechanical and / or robotic assembly (which includes the stage 206) that can be configured to move the workpiece 202 so that light can scan across the workpiece 202. Alternatively or additionally, the optical-based subsystem 201 may be configured such that one or more optical elements of the optical-based subsystem 201 perform an optical scan across the workpiece 202. The optical scan across the workpiece 202 may be performed in any suitable manner (e.g., in a serpentine or helical path).

[0058] The optical-based subsystem 201 further includes one or more detection channels. At least one of the detection channels includes a detector configured to detect light from the workpiece 202 due to illumination of the workpiece 202 by the subsystem, and to generate an output in response to the detected light. For example, Figure 2 The optical-based subsystem 201 shown includes two detection channels, one formed by a light collector 207, an element 208, and a detector 209, and the other formed by a light collector 210, an element 211, and a detector 212. Figure 2 As shown, two detection channels are configured to collect and detect light at different focusing angles. In some examples, the two detection channels are configured to detect scattered light, and the detection channel is configured to detect light scattered from the workpiece 202 at different angles. However, one or more of the detection channels may be configured to detect another type of light (e.g., reflected light) from the workpiece 202.

[0059] like Figure 2 The diagram further illustrates that two detection channels are positioned within the plane of the paper, and the illumination subsystem is also positioned within the plane of the paper. Therefore, in this embodiment, the two detection channels are positioned within the plane of incidence (e.g., centered on the plane of incidence). However, one or more of the detection channels may be positioned outside the plane of incidence. For example, the detection channel formed by the light collector 210, element 211, and detector 212 may be configured to collect and detect light scattered from the plane of incidence. Therefore, this detection channel may generally be referred to as a "side" channel, and this side channel may be centered on a plane substantially perpendicular to the plane of incidence.

[0060] although Figure 2An embodiment of an optically based subsystem 201 comprising two detection channels is shown, but the optically based subsystem 201 may comprise a different number of detection channels (e.g., only one detection channel or two or more detection channels). In one example, the detection channel formed by the light collector 210, element 211, and detector 212 may form a side channel as described above, and the optically based subsystem 201 may include an additional detection channel (not shown) formed as another side channel positioned on the opposite side of the incident plane. Thus, the optically based subsystem 201 may include a detection channel comprising a light collector 207, element 208, and detector 209 centered on the incident plane and configured to collect and detect light at a scattering angle normal to or near normal to the surface of the workpiece 202. Therefore, this detection channel may generally be referred to as the “top” channel, and the optically based subsystem 201 may also include two or more side channels configured as described above. Therefore, the optical-based subsystem 201 may include at least three channels (i.e., a top channel and two side channels), and each of the at least three channels has its own light collector, each of the light collectors being configured to collect light at a different scattering angle than each of the other light collectors.

[0061] As further described above, each of the detection channels included in the optical-based subsystem 201 can be configured to detect scattered light. Therefore, Figure 2 The optical-based subsystem 201 shown herein can be configured for generating dark-field (DF) output from sample 202. However, the optical-based subsystem 201 may also, or alternatively, include a detection channel configured for generating bright-field (BF) output from sample 202. In other words, the optical-based subsystem 201 may include at least one detection channel configured to detect light reflected from the specular surface of workpiece 202. Therefore, the optical-based subsystem 201 described herein can be configured for DF-only, BF-only, or both DF and BF imaging. Although in Figure 2 Each of the light collectors is shown as a single refractive optical element, but it should be understood that each of the light collectors may contain one or more refractive optical blanks and / or one or more reflective optical elements.

[0062] One or more detection channels may contain any suitable detector known in the art. For example, the detector may include a photomultiplier tube (PMT), a charge-coupled device (CCD), a time-delay integration (TDI) camera, and any other suitable detector known in the art. The detector may also include non-imaging detectors or imaging detectors. In this way, if the detector is a non-imaging detector, each of the detectors may be configured to detect specific characteristics (e.g., intensity) of the scattered light, but may not be configured to detect such characteristics that vary depending on the position in the imaging plane. Thus, the output generated by each of the detectors in each of the detection channels of the optical-based subsystem may be a signal or data, rather than an image signal or image data. In such examples, a processor (e.g., processor 214) may be configured to generate an image of the workpiece 202 from the non-imaging output of the detector. However, in other examples, the detector may be configured as an imaging detector configured to generate an imaging signal or image data. Thus, the optical-based subsystem may be configured to generate optical images or other optical-based outputs as described herein in several ways.

[0063] It should be noted that the information provided in this article... Figure 2 The configuration of an optically based subsystem 201, which may be included in or may produce optically based output used by the system embodiments described herein, is described in general terms. The configuration of the optically based subsystem 201 described herein may be modified to optimize the performance of the optically based subsystem 201 as typically performed when designing a commercial output acquisition system. Alternatively, the system described herein may be implemented using existing systems (e.g., by adding the functionality described herein to an existing system). For some such systems, the methods described herein may be provided as optional functionality of the system (e.g., in addition to other system functionalities). Alternatively, the system described herein may be designed as a completely new system.

[0064] Figure 3 This is a block diagram of an embodiment of system 300. System 300 includes a wafer inspection tool (which includes an electron column 301) configured to generate an image of a workpiece 304.

[0065] The wafer inspection tool includes an output acquisition subsystem, which comprises at least one energy source and a detector. The output acquisition subsystem may be an electron beam-based output acquisition subsystem. For example, in one embodiment, the energy directed to workpiece 304 contains electrons, and the energy detected from workpiece 304 contains electrons. In this way, the energy source may be an electron beam source. Figure 3 In one embodiment shown, the output acquisition subsystem includes an electronic column 301 coupled to the computer subsystem 302. A stage 310 holds the workpiece 304.

[0066] For example Figure 3 As shown, electron column 301 includes electron beam source 303, configured to generate electrons that are focused onto workpiece 304 through one or more components 305. Electron beam source 303 may include, for example, a cathode source or an emitter tip. One or more components 305 may include, for example, a gun lens, anode, beam-limiting aperture, gate valve, beam current selection aperture, objective lens, and scanning subsystem, all of which may include any such suitable components known in the art.

[0067] Electrons returning from workpiece 304 (e.g., secondary electrons) can be focused onto detector 307 by one or more elements 306. One or more elements 306 may include, for example, a scanning subsystem, which may be the same scanning subsystem included in element 305.

[0068] The electronic column 301 may also include any other suitable elements known in the art.

[0069] Although electron column 301 is in Figure 3 The image is shown as being configured such that the electron beam is guided to workpiece 304 at an oblique incident angle and scattered from workpiece 304 at another oblique angle, but the electron beam can be guided to and scattered from workpiece 304 at any suitable angle. Furthermore, the electron beam-based output acquisition subsystem can be configured to generate images of workpiece 304 using multiple modes (e.g., employing different illumination angles, focusing angles, etc.). These multiple modes of the electron beam-based output acquisition subsystem can differ in terms of any image generation parameters of the output acquisition subsystem.

[0070] It should be noted that the information provided in this article... Figure 3 The configuration of the electron beam-based output acquisition subsystem that can be used in the embodiments described herein is explained in general terms. The configuration of the electron beam-based output acquisition subsystem described herein can be modified to optimize the performance of the output acquisition subsystem as is typically done when designing a commercial output acquisition system. Alternatively, the system described herein can be implemented using existing systems (e.g., by adding the functionality described herein to an existing system). For some such systems, the methods described herein may be provided as optional functionality of the system (e.g., in addition to other system functionalities). Alternatively, the system described herein may be designed as a completely new system.

[0071] Although the output acquisition subsystem has been described above as an electron beam-based output acquisition subsystem, it can also be an ion beam-based output acquisition subsystem. This output acquisition subsystem can be as follows: Figure 3The configuration shown is such that, except that the electron beam source can be replaced by any suitable ion beam source known in the field, the output acquisition subsystem can be any other suitable ion beam-based output acquisition subsystem, such as those included in commercially available focused ion beam (FIB) systems, helium ion microscopy (HIM) systems, and secondary ion mass spectrometry (SIMS) systems.

[0072] Embodiments of system 200 and system 300 include a processor and an electronic data storage unit, such as processor 214 or processor 308 and electronic data storage unit 215 or electronic data storage unit 309. Processor 214 or processor 308 may include a microprocessor, microcontroller, or other device. In this example, processor 309 is part of a computer subsystem (e.g., computer subsystem 302).

[0073] The processor 214 or processor 308, other systems, or other subsystems described herein may be part of various systems, including personal computer systems, graphics computers, mainframe computer systems, workstations, network appliances, internet appliances, or other devices. The subsystem or the system may also include any suitable processor known in the art, such as a parallel processor. Additionally, the subsystem or the system may include a platform with high-speed processing and software as a standalone or network-linked tool.

[0074] Processor 214 or processor 308 may be coupled to elements of system 200 or system 300 in any suitable manner (e.g., via one or more transmission media, which may include wired and / or wireless transmission media) such that processor 214 or processor 308 can receive output. Processor 214 or processor 308 may be configured to use the output to perform certain functions. A wafer inspection tool may receive instructions or other information from processor 214 or processor 308. Processor 214 or processor 308 may optionally communicate electronically with another wafer inspection tool, wafer metrology tool, or wafer re-inspection tool (not described) to receive additional information or send instructions.

[0075] Processor 214 or processor 308 communicates electronically with a wafer inspection tool (e.g., detector 209, detector 212, or detector 307). Processor 214 or processor 308 may be configured to process images generated using measurements from the detectors or to perform other functions. For example, the processor may perform embodiments of method 100 or method 200.

[0076] Processor 214 or processor 308 and its associated electronic data storage unit may be housed in system 200 or system 300 or otherwise be part of system 200 or system 300 or may be part of another device. In examples, processor 214 or processor 308 and its associated electronic data storage unit may be part of a separate control unit or in a centralized quality control unit. Multiple processors 214 or processor 308 or electronic data storage units may be used.

[0077] Processor 214 or processor 308 may be implemented in any combination of hardware, software, and firmware. Furthermore, its functionality, as described herein, may be executed by a single unit or partitioned among different components, each of which may in turn be implemented in any combination of hardware, software, and firmware. Program code or instructions for processor 214 or processor 308 to implement various methods and functions may be stored in a readable storage medium (e.g., the memory in electronic data storage unit 215, electronic data storage unit 309, or other memory).

[0078] Processor 214 or processor 308 may be configured to perform certain functions using the outputs of system 200, system 300, or other outputs. For example, processor 214 or processor 308 may be configured to send outputs to its corresponding electronic data storage unit or another storage medium. Processor 214 or processor 308 may be further configured as described herein.

[0079] Processor 214 or processor 308 may be configured according to any of the embodiments described herein. For example, processor 214 or processor 308 may be configured to perform an embodiment of method 100. Processor 214 or processor 308 may also be configured to perform other functions or additional steps using the output of system 200, system 300, or using images or data from other sources.

[0080] If system 200 or system 300 includes more than one processor 214 or processor 308, then different subsystems can be coupled to each other so that images, data, information, instructions, etc., can be transmitted directly between the subsystems. For example, a subsystem can be coupled to an additional subsystem via any suitable transmission medium known in the art that is suitable for wired and / or wireless transmission. Two or more such subsystems can also be effectively coupled by sharing a computer-readable storage medium (not shown).

[0081] The processor (e.g., processor 214 or processor 308) used to perform the embodiments disclosed herein may be part of a single computer or a computer network. The processor may include one or more CPU processors or a combination of CPU and GPU processors. In this example, the network has 15 computer nodes, and each node includes 44 CPUs and 2 GPUs with 385 GB of RAM and 500 GB of hard disk space.

[0082] The various steps, functions, and / or operations of the systems 200, 300, or methods disclosed herein may be implemented by one or more of the following: electronic circuits, logic gates, multiplexers, programmable logic devices, ASICs, analog or digital controls / switches, microcontrollers, or computing systems. Program instructions for implementing methods such as those described herein may be transmitted via or stored on a carrier medium. The carrier medium may include storage media, such as read-only memory, random access memory, magnetic disks or optical disks, non-volatile memory, solid-state memory, magnetic tape, and the like. The carrier medium may include transmission media, such as wires, cables, or wireless transmission links. Therefore, the foregoing description should not be construed as a limitation of this disclosure but is merely illustrative.

[0083] Additional embodiments relate to a non-transitory computer-readable medium storing program instructions executable on a controller for performing computer-implemented methods as disclosed herein. Specifically, electronic data storage unit 215, electronic data storage unit 309, or other storage media may contain a non-transitory computer-readable medium comprising program instructions executable on a processor. Computer-implemented methods may include any steps of any method described herein (including embodiments of method 100).

[0084] The program instructions can be implemented in any of a variety of ways, including process-based techniques, component-based techniques, and / or object-oriented techniques. For example, ActiveX controls, C++ objects, JavaBeans, Microsoft Foundation Classes (MFC), Streaming SIMD Extensions (SSE), or other techniques or methodologies may be used as needed to implement the program instructions.

[0085] Each of the steps of the method may be performed as described herein. The method may also include any other steps that can be performed by a processor and / or computer subsystem or system as described herein. The steps may be performed by one or more computer systems that can be configured according to any of the embodiments described herein. Additionally, the methods described above may be performed by any of the system embodiments described herein.

[0086] Although this disclosure has been described with respect to one or more specific embodiments, it should be understood that other embodiments of this disclosure may be made without departing from the scope of this disclosure. Therefore, this disclosure is to be considered limited only by the appended claims and their reasonable interpretation.

Claims

1. A method comprising: The processor receives the initial probability of random defects occurring within the inspection area of ​​the workpiece, wherein the initial probability of random defects occurring is generated using a model; One or more defects within the inspection area are imaged using optical or electron beam tools, wherein each defect is located at a defect location. Using the processor, the model is used to generate the probability of a random defect occurring at each of the defect locations; The processor is used to group the defect locations into probability grids; The processor is used to determine the consistency between the initial probability and the observation; and The model is tuned based on the aforementioned consistency.

2. The method of claim 1, further comprising determining the desired resolution of the probability prediction for each of the probability lattices prior to using the grouping by the processor.

3. The method of claim 1, further comprising using the processor to determine the expected defect count within each of the probability lattices.

4. The method of claim 1, wherein the consistency is determined using binary cross-entropy, the expected RMSe of the observation count, a binomial test of significance, or a Blair score.

5. The method of claim 1, wherein the imaging uses the electron beam tool.

6. The method of claim 5, wherein the imaging occurs during multiple workpiece exposures.

7. The method of claim 6, further comprising determining the defect frequency of the defect based on defect count.

8. The method of claim 5, wherein the defect locations are grouped according to the geometric pattern shape on the workpiece.

9. The method of claim 1, further comprising using the processor to apply the model to generate the initial probability of random defect occurrence.

10. A non-transitory computer-readable medium storing a program, configured to instruct the processor to perform the method according to claim 1.

11. A system comprising: Inspection tools, configured to image the workpiece; and A processor that communicates electronically with the inspection tool, wherein the processor is configured to: Receive the initial probability of random defects occurring within the inspection area of ​​the workpiece, wherein a model is used to generate the initial probability of random defects occurring; Send instructions to use the detection tool to image one or more defects within the detection area, wherein each defect is located at a defect location; Use a model to generate the probability of a random defect occurring at each defect location; The defect locations are grouped into probability grids; Determine the consistency between the initial probability and the observation results; and The model is tuned based on the aforementioned consistency.

12. The system of claim 11, wherein the inspection tool is an optical tool or an electron beam tool.

13. The system of claim 11, wherein the processor is further configured to determine the desired resolution of the probability prediction for each of the probability lattices prior to the grouping.

14. The system of claim 11, wherein the processor is further configured to determine the expected defect count within each of the probability lattices.

15. The system of claim 11, wherein the consistency is determined using binary cross-entropy, the expected RMSe of the observation count, a binomial test of significance, or a Blair score.

16. The system of claim 11, wherein the inspection tool is an electron beam tool, and wherein the imaging occurs within multiple workpiece exposures.

17. The system of claim 16, wherein the processor is further configured to determine the defect frequency of the defects based on defect counts.

18. The system of claim 16, wherein the defect locations are grouped according to the geometric pattern shape on the workpiece.

19. The system of claim 11, wherein the processor is further configured to use the model to generate the initial probability of random defect occurrence.

20. A non-transitory computer-readable storage medium comprising one or more programs for performing the following steps on one or more computing devices: Receive an initial probability of random defects occurring within the inspection area of ​​the workpiece, wherein a model is used to generate said initial probability of random defects occurring; and Send instructions to image one or more defects within the detection area using optical or electron beam tools, wherein each defect is located at a defect location; The model is used to generate the probability of a random defect occurring at each of the defect locations; The defect locations are grouped into probability grids; Determine the consistency between the initial probability and the observation results; and The model is tuned based on the aforementioned consistency.

Images