Correcting target location for temperature in semiconductor applications

Abstract

Methods and systems for determining information of a sample are provided. In one embodiment, a system includes an output acquisition subsystem configured to generate an output of a sample at one or more target locations on the sample; and one or more temperature sensors configured to measure one or more temperatures within the system. The system also includes a deep learning model configured to predict an error of at least one of the one or more target locations based on at least one of the one or more measured temperatures input to the deep learning model by a computer subsystem. The computer subsystem is configured to determine a corrected target position of the at least one of the one or more target positions by applying the prediction error to the at least one of the one or more target positions.

Description

Technical Field

[0001] The present invention generally relates to methods and systems for determining information of a sample. Particular embodiments relate to calibrating a target position for temperature on a sample in a system performing a process on the sample. Background Art

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

[0003] Fabricating semiconductor devices, such as logic devices and memory devices, typically includes processing a substrate, such as a semiconductor wafer, using a number of semiconductor fabrication processes to form various features and multiple levels of the semiconductor device. For example, photolithography is a semiconductor fabrication process that involves transferring a pattern from a mask to a resist disposed on a semiconductor wafer. Additional examples of semiconductor fabrication processes include, but are not limited to, chemical mechanical polishing (CMP), etching, deposition, and ion implantation. Multiple semiconductor devices may be fabricated in an arrangement on a single semiconductor wafer and then separated into individual semiconductor devices.

[0004] Inspection processes are used at various steps during the semiconductor manufacturing process to detect defects on samples to facilitate higher yields and therefore higher profits in the manufacturing process. Inspection has always been an important part of making semiconductor devices. However, as semiconductor device sizes decrease, inspection becomes more important to successfully manufacturing acceptable semiconductor devices because even small defects can cause the device to fail.

[0005] Defect re-inspection generally involves re-inspecting defects as detected by the inspection process and generating additional information about the defects at a higher resolution using a high magnification optical system or a scanning electron microscope (SEM). Thus, defect re-inspection is performed at discrete locations on the sample where the defect was detected by inspection. The higher resolution data for the defect generated by defect re-inspection is more suitable for determining the properties of the defect, such as contour, roughness, more accurate size information, etc. Compared to inspection, defects can generally be more accurately classified into several defect types based on the information determined by defect re-inspection.

[0006] Metrology processes are also used at various steps during the semiconductor manufacturing process to monitor and control the process. Metrology processes differ from inspection processes in that, unlike inspection processes in which defects are detected on a sample, metrology processes are used to measure one or more characteristics of a sample that cannot be determined using currently used inspection tools. For example, a metrology process is used to measure one or more characteristics of a sample (e.g., the dimensions of features formed on the sample during a process (e.g., line width, thickness, etc.)) so that the performance of the process can be determined based on the one or more characteristics. In addition, if one or more characteristics of a sample are unacceptable (e.g., outside a predetermined range of characteristics), the measurement of the one or more characteristics of the sample can be used to change one or more parameters of the process so that additional samples manufactured by the process have acceptable characteristics.

[0007] The metrology process also differs from the defect re-inspection process in that, unlike the defect re-inspection process in which defects detected by inspection are revisited in the defect re-inspection, the metrology process can be performed at a location where no defects have been detected. In other words, unlike defect re-inspection, the location at which the metrology process is performed on the sample can be independent of the results of the inspection process performed on the sample. In particular, the location at which the metrology process is performed can be selected independently of the inspection results. In addition, since the location on the sample at which the metrology is performed can be selected independently of the inspection results, unlike defect re-inspection in which the location on the sample at which the defect re-inspection is to be performed cannot be determined until the inspection results for the sample are generated and made available, the location at which the metrology process is performed can be determined before the inspection process has been performed on the sample.

[0008] One aspect of the methods and systems described above that may be difficult is knowing where on a sample a result, such as a measurement, a detected defect, a re-detected defect, etc., was generated. For example, the tools and processes described above are used to determine information about structures and / or defects on a sample. Since structures vary across a sample (such that the structures may form a functional device on the sample), a measurement, inspection, or defect re-detection result is generally useless unless it is precisely known where on the sample the measurement, inspection, or defect re-detection result was generated. In a metrology example, unless a measurement is performed at a known predetermined location on the sample, the measurement may fail if the measurement location does not contain the portion of the sample intended to be measured and / or a measurement of one portion of the sample is assigned to another portion of the sample. In the case of inspection, unless the defect location on the sample is determined substantially accurately, the defect location may be inaccurately determined relative to the sample and / or the design of the sample. In any case, an error in the location on the sample where a result is generated may render the result useless and may even be detrimental to the manufacturing process if the result is used to make changes to the manufacturing process.

[0009] Known alignment targets or marks on the sample can be used to align the sample relative to the tool before the sample is scanned or measured. Based on information about the sample (e.g., where the pattern is formed on the sample and / or where the intended target is on the sample), the tool can then perform a measurement or scan of the intended target on the sample. Even if the alignment before the measurement or scan is successful, there are still several factors that can cause the reported position in the results to be inaccurate. Therefore, many processes involving navigating across the sample include measuring and correcting positioning errors during runtime. In some such examples, the scanning subsystem of the tool (which may include a robot and any other associated hardware) can move the sample into position and then the tool can perform alignment on a site-by-site basis.

[0010] However, there are several disadvantages to such site-by-site alignment methods. For example, such methods are generally substantially slow. Site-by-site alignment also requires recording more images and image processing to be performed when performing the alignment. In addition, site-by-site alignment methods do not take into account temperature-induced positioning errors on the tool.

[0011] Therefore, it would be advantageous to develop systems and methods for determining information of a sample that do not suffer from one or more of the disadvantages described above. Summary of the invention

[0012] The following description of various embodiments should in no way be considered to limit the subject matter of the appended claims.

[0013] One embodiment relates to a system configured to determine information of a sample. The system includes: an output acquisition subsystem configured to generate an output of the sample at one or more target locations on the sample; and one or more temperature sensors configured to measure one or more temperatures within the system. The system also includes: a computer subsystem configured to acquire the one or more measured temperatures from the one or more temperature sensors; and one or more components executed by the computer subsystem. One or more components include a deep learning (DL) model, the DL model being configured to predict an error of at least one of the one or more target locations based on at least one of the one or more measured temperatures input to the DL model by the computer subsystem. The computer subsystem is configured to determine a corrected target position of at least one of the one or more target locations by applying the predicted error to at least one of the one or more target locations. The system may be further configured as described herein.

[0014] Another embodiment relates to a method for determining information of a sample. The method includes measuring one or more temperatures within a system using one or more temperature sensors. The system includes an output acquisition subsystem configured to generate an output of the sample at one or more target locations on the sample. The method also includes predicting an error of at least one of the one or more target locations by inputting at least one of the one or more measured temperatures into a DL model included in one or more components executed by a computer system. In addition, the method includes determining a corrected target location of at least one of the one or more target locations by applying the prediction error to at least one of the one or more target locations. The input and determination are performed by a computer system.

[0015] Each of the steps of the method may be performed as further described herein. The method may include any other steps of any other method described herein. The method may be performed by any of the systems described herein.

[0016] Another embodiment relates to a non-transitory computer-readable medium storing program instructions that can be executed on a computer system to perform a computer-implemented method for determining information of a sample. The computer-implemented method includes obtaining one or more temperatures measured within the system using one or more temperature sensors. The system is configured as described above, and the computer-implemented method includes the prediction and determination steps of the method described above. The computer-readable medium may be further configured as described herein. The steps of the computer-implemented method may be performed as further described herein. In addition, the computer-implemented method may include any other steps of any other method described herein, and the program instructions may be executed to implement the computer-implemented method. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other advantages of the present invention will become apparent to those skilled in the art upon having the benefit of the following detailed description of the preferred embodiments and upon reference to the accompanying drawings, in which:

[0018] Figures 1 to 3 is a schematic diagram illustrating a side view of an embodiment of a system configured as described herein;

[0019] Figures 4 to 5 are flow charts illustrating steps that may be performed by embodiments described herein during training and runtime, respectively; and

[0020] Figure 6 is a block diagram illustrating one embodiment of a non-transitory computer-readable medium storing program instructions for causing a computer system to perform the computer-implemented methods described herein.

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

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

[0023] In general, embodiments described herein are systems and methods for determining information about a sample. The information determined by the embodiments described herein includes positioning correction estimates for applications such as metrology and inspection. Embodiments may be advantageously used to generate positioning correction information to improve the accuracy of sample placement in equipment such as metrology and inspection tools. Additionally, embodiments described herein advantageously use the field of machine learning (ML) to estimate positioning corrections.

[0024] In some embodiments, the sample is a wafer. The wafer may include any wafer known in semiconductor technology. Although some embodiments may be described herein with respect to one or several wafers, the embodiments are not limited to the samples they may be used for. For example, the embodiments described herein may be used for samples such as masks, flat panels, personal computer (PC) boards, and other semiconductor samples.

[0025] exist Figure 1 One embodiment of a system configured for determining information of a sample is shown in FIG. The system includes an output acquisition subsystem 100. The output acquisition subsystem includes and / or is coupled to a computer subsystem, such as computer subsystem 36 and / or one or more computer systems 102.

[0026] In general, the output acquisition subsystem described herein includes at least an energy source, a detector, and a scanning subsystem. The energy source is configured to generate energy directed to the sample by the output acquisition subsystem. The detector is configured to detect energy from the sample and generate an output in response to the detected energy. The scanning subsystem is configured to change the location on the sample to which energy is directed and from which energy is detected. In one embodiment, Figure 1As shown in , the output acquisition subsystem is configured as a light-based output acquisition subsystem.

[0027] In the light-based output acquisition subsystem described herein, the energy directed to the sample comprises light, and the energy detected from the sample comprises light. Figure 1 In an embodiment of the system shown in , the output acquisition subsystem includes an illumination subsystem configured to direct light to the sample 14. The illumination subsystem includes at least one light source. For example, Figure 1 , the illumination subsystem includes a light source 16. The illumination subsystem is configured to direct light to the sample at one or more incident angles (which may include one or more oblique angles and / or one or more normal angles). Figure 1 , light from light source 16 is directed at an oblique angle of incidence through optical element 18 and then through lens 20 to sample 14. The oblique angle of incidence may include any suitable oblique angle of incidence, which may vary depending on, for example, the characteristics of the sample and the process performed on the sample.

[0028] The illumination subsystem may be configured to direct light to the sample at different incident angles at different times. For example, the output acquisition subsystem may be configured to alter one or more characteristics of one or more elements of the illumination subsystem so that the light can be directed at different incident angles. Figure 1 In one such example, the output acquisition subsystem may be configured to move the light source 16, optical element 18, and lens 20 so that light is directed to the sample at different oblique angles of incidence or normal (or near-normal) angles of incidence.

[0029] In some examples, the output acquisition subsystem can be configured to direct light to the sample at more than one incident angle simultaneously. For example, the illumination subsystem can include more than one illumination channel, one of which can include, for example, Figure 1 , and another of the illumination channels (not shown) may include similar elements that may be configured differently or identically, or may include at least a light source and possibly one or more other components (such as those described further herein). If this light is directed to the sample simultaneously with other light, one or more characteristics (e.g., wavelength, polarization, etc.) of the light directed to the sample at different angles of incidence may be different, so that light resulting from illumination of the sample at different angles of incidence may be distinguished from one another at the detector(s).

[0030] In another example, the illumination subsystem may include only one light source (e.g., Figure 1), and the light from the light source can be separated into different optical paths (e.g., based on wavelength, polarization, etc.) by one or more optical elements (not shown) of the illumination subsystem. The light in each of the different optical paths can then be directed to the sample. Multiple illumination channels can be configured to direct light to the sample simultaneously or at different times (e.g., when different illumination channels are used to sequentially illuminate the sample). In another example, the same illumination channel can be configured to direct light with different characteristics to the sample at different times. For example, the optical element 18 can be configured as a spectral filter and the properties of the spectral filter can be changed in a variety of different ways (e.g., by replacing one spectral filter with another spectral filter) so that different wavelengths of light can be directed to the sample at different times. The illumination subsystem can have any other suitable configuration known in the art for directing light with different or the same characteristics to the sample sequentially or simultaneously at different or the same angles of incidence.

[0031] 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 sample may include broadband light. However, the light source may include any other suitable light source, such as any suitable laser known in the art that is configured to generate light at any suitable wavelength. The laser may be configured to generate monochromatic or nearly monochromatic light. In this way, the laser may be a narrow-band laser. The light source may also include a polychromatic light source that generates light at multiple discrete wavelengths or bands.

[0032] Light from optical element 18 may be focused onto sample 14 by lens 20. Figure 1 The lens 20 is shown as a single refractive optical element in FIG, but in reality, the lens 20 may include several refractive and / or reflective optical elements that in combination focus the light from the optical element to the sample. Figure 1 The illumination subsystem shown in and described herein may include any other suitable optical elements (not shown). Examples of such optical elements include, but are not limited to, polarizing components, spectral filters, spatial filters, reflective optical elements, apodizers, beam splitters, stops, and the like, which may include any such suitable optical elements known in the art. Additionally, the system may be configured to alter one or more of the elements of the illumination subsystem based on the type of illumination to be used to produce the output.

[0033] The output acquisition subsystem may also include a scanning subsystem configured to change the location on the sample to which light is directed and detected from and may cause the light to scan across the sample. For example, the output acquisition subsystem may include a stage 22 on which the sample 14 is disposed during output generation. The scanning subsystem may include any suitable mechanical and / or robotic assembly (including the stage 22) that may be configured to move the sample so that light can be directed to or detected from different locations on the sample. Additionally or alternatively, the output acquisition subsystem may be configured so that one or more optical elements of the output acquisition subsystem perform some light scanning on the sample so that light can be directed to or detected from different locations on the sample. In instances where light is scanned across the sample, the light may be scanned across the sample in any suitable manner, such as in a serpentine path or in a spiral path.

[0034] The output acquisition subsystem further includes one or more detection channels. At least one of the detection channels includes a detector configured to detect light from the sample due to illumination of the sample by the output acquisition subsystem and generate an output in response to the detected light. For example, Figure 1 The output acquisition subsystem shown in FIG. 1 includes two detection channels, one detection channel formed by collector 24, element 26 and detector 28 and the other detection channel formed by collector 30, element 32 and detector 34. Figure 1 , the two detection channels are configured to collect and detect light at different collection angles. In some examples, the two detection channels are configured to detect scattered light, and the detection channels are configured to detect light scattered from the sample at different angles. However, one or more of the detection channels may be configured to detect another type of light from the sample (e.g., reflected light).

[0035] like Figure 1 , the two detection channels are shown as being positioned in the plane of the paper and the illumination subsystem is also shown as being positioned in the plane of the paper. Thus, in this embodiment, the two detection channels are positioned in (e.g., centered in) 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 collector 30, the element 32, and the detector 34 may be configured to collect and detect light that is scattered out of the plane of incidence. Thus, this detection channel may be generally referred to as a "side" channel, and this side channel may be centered in a plane that is substantially perpendicular to the plane of incidence.

[0036] Although Figure 1An embodiment of an output acquisition subsystem including two detection channels is shown, but the output acquisition subsystem may include a different number of detection channels (e.g., only one detection channel or two or more detection channels). In one such example, the detection channel formed by the collector 30, the element 32, and the detector 34 may form a side channel as described above, and the output acquisition subsystem may include an additional detection channel (not shown) formed as another side channel positioned on the opposite side of the incident plane. Therefore, the output acquisition subsystem may include a detection channel that includes the collector 24, the element 26, and the detector 28 and is centered in the incident plane and is configured to collect and detect light at a scattering angle that is at or close to the normal to the sample surface. This detection channel may therefore be collectively referred to as a "top" channel, and the output acquisition subsystem may also include two or more side channels configured as described above. In this way, the output acquisition subsystem may include at least three channels (i.e., one top channel and two side channels), and each of the at least three channels has its own collector, each collector being configured to collect light at a different scattering angle than each of the other collectors.

[0037] As further described above, each of the detection channels included in the output acquisition subsystem can be configured to detect scattered light. Figure 1 The output acquisition subsystem shown in can be configured for dark field (DF) imaging of the sample. However, the output acquisition subsystem may also or alternatively include a detection channel configured for bright field (BF) imaging of the sample. In other words, the output acquisition subsystem may include at least one detection channel configured to detect light specularly reflected from the sample. Thus, the output acquisition subsystem described herein may be configured for only DF imaging, only BF imaging, or both DF imaging and BF imaging. Although in Figure 1 Each of the collectors is shown in as a single refractive optical element, but each of the collectors may include one or more refractive optical elements and / or one or more reflective optical elements.

[0038] One or more detection channels may include any suitable detector known in the art, such as a photomultiplier tube (PMT), a charge coupled device (CCD), and a time delay integration (TDI) camera. The detector may also include a non-imaging detector or an imaging detector. If the detector is a non-imaging detector, each of the detectors may be configured to detect specific characteristics (such as intensity) of the scattered light but may not be configured to detect such characteristics based on positioning within the imaging plane. In this way, the output generated by each of the detectors included in each of the detection channels of the output acquisition subsystem may be a signal or data, rather than an image signal or image data. In such examples, a computer subsystem (such as computer subsystem 36) may be configured to generate an image of the sample from the non-imaging output of the detector. However, in other examples, the detector may be configured as an imaging detector, which is configured to generate an imaging signal or image data. Therefore, the output acquisition subsystem may be configured to generate an image in several ways.

[0039] It should be noted that this article provides Figure 1 The configuration of an output acquisition subsystem that may be included in an embodiment of the system described herein is generally illustrated. Obviously, the output acquisition subsystem configuration described herein may be modified to optimize the performance of the output acquisition subsystem, as is typically performed when designing a commercial system. In addition, the system described herein may be implemented using an existing system such as the 29xx / 39xx series of tools commercially available from KLA Corp., Milpitas, Calif. (e.g., by adding the functionality described herein to the 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 functionality of the system). Alternatively, the system described herein may be designed "from scratch" to provide an entirely new system.

[0040] The computer subsystem 36 may be coupled to the detector of the output acquisition subsystem in any suitable manner (e.g., via one or more transmission media, which may include "wired" and / or "wireless" transmission media) so that the computer subsystem may receive the output generated by the detector. The computer subsystem 36 may be configured to perform a number of functions, including the steps and functions further described herein, with or without the output of the detector. As such, the steps described herein may be performed "on tool" by a computer subsystem that is coupled to or is part of the output acquisition subsystem. Additionally or alternatively, the computer system 102 may perform one or more of the steps described herein. Thus, one or more of the steps described herein may be performed "off tool" by a computer system that is not directly coupled to the output acquisition subsystem. The computer subsystem 36 and the computer system 102 may be further configured as described herein.

[0041] The computer subsystem 36 (and other computer subsystems described herein) may also be referred to herein as a computer system. Each of the computer subsystems or systems described herein may take various forms, including a personal computer system, an image computer, a mainframe computer system, a workstation, a network appliance, an Internet appliance, or other devices. In general, the term "computer system" may be broadly defined to encompass any device having one or more processors that execute instructions from a storage medium. The computer subsystem or system may also include any suitable processor known in the art, such as a parallel processor. In addition, the computer subsystem or system may include a computer platform with high-speed processing and software as a stand-alone tool or a networked tool.

[0042] If the system includes more than one computer subsystem, the different computer subsystems may be coupled to each other so that images, data, information, instructions, etc. may be transmitted between the computer subsystems. For example, computer subsystem 36 may be coupled to computer system 102 via any suitable transmission medium (e.g., Figure 1 The transmission medium may include any suitable wired and / or wireless transmission medium known in the art. Two or more of such computer subsystems may also be effectively coupled by sharing a computer-readable storage medium (not shown).

[0043] Although the output acquisition subsystem is described above as an optical or light-based output acquisition subsystem, in another embodiment, the output acquisition subsystem is configured as an electron-based output acquisition subsystem. In an electron beam output acquisition subsystem, the energy directed to the sample comprises electrons, and the energy detected from the sample comprises electrons. Figure 2In one such embodiment shown in FIG. 1 , the output acquisition subsystem includes an electronic column 122, and the system includes a computer subsystem 124 coupled to the output acquisition subsystem. The computer subsystem 124 may be configured as described above. In addition, this output acquisition subsystem may be configured with the electronic column 122 and the computer subsystem 124 coupled to the output acquisition subsystem. Figure 1 The system can be coupled to another computer system or systems in the same manner as shown in FIG.

[0044] Also like Figure 2 , the electron column includes an electron beam source 126 configured to generate electrons that are focused to a sample 128 by one or more elements 130. The electron beam source may include, for example, a cathode source or emitter tip, and the one or more elements 130 may include, for example, a gun lens, an anode, a beam limiting aperture, a gate valve, a beam current selecting aperture, an objective lens, and a scanning subsystem, all of which may include any such suitable elements known in the art.

[0045] Electrons returning from the sample (eg, secondary electrons) may be focused to a detector 134 by one or more elements 132. One or more elements 132 may include, for example, a scanning subsystem, which may be the same scanning subsystem included in element 130.

[0046] The electron column may include any other suitable elements known in the art. In addition, the electron column may be further configured as described in the following U.S. Patents: U.S. Patent No. 8,664,594 to Jiang et al., issued on April 4, 2014, U.S. Patent No. 8,692,204 to Kojima et al., issued on April 8, 2014, U.S. Patent No. 8,698,093 to Gubbens et al., issued on April 15, 2014, and U.S. Patent No. 8,716,662 to MacDonald et al., issued on May 6, 2014, which are generally incorporated by reference as if fully set forth herein.

[0047] Although the electron column Figure 2 1 is shown as being configured so that electrons are directed to the sample at an oblique angle of incidence and scattered from the sample at another oblique angle, but the electron beam can be directed to the sample and scattered from the sample at any suitable angle. In addition, as further described herein, the electron beam output acquisition subsystem can be configured to use multiple modes to generate output for the sample (e.g., at different illumination angles, collection angles, etc.). The multiple modes of the electron beam output acquisition subsystem can differ in any output generation parameter of the output acquisition subsystem.

[0048] The computer subsystem 124 may be coupled to a detector 134, as described above. The detector may detect electrons returning from the surface of the sample, thereby forming an electron beam image (or other output) of the sample. The electron beam image may include any suitable electron beam image. The computer subsystem 124 may be configured to use the output generated by the detector 134 to determine information about the sample, which may be performed as further described herein. The computer subsystem 124 may be configured to perform any additional steps described herein. The computer subsystem 124 may be further configured to include as described herein. Figure 2 The output acquisition subsystem shown in the system.

[0049] It should be noted that this article provides Figure 2 The configuration of the electron beam output acquisition subsystem that may be included in the embodiments described herein is generally illustrated. As with the optical subsystem described above, the electron beam subsystem configuration described herein may be modified to optimize the performance of the output acquisition subsystem, as is typically performed when designing a commercial system. In addition, the systems described herein may be implemented using existing systems such as tools commercially available from KLA (e.g., by adding the functionality described herein to the 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 functionality of the system). Alternatively, the systems described herein may be designed "from the ground up" to provide an entirely new system.

[0050] Although the output acquisition subsystem is described above as a light beam or electron beam subsystem, the output acquisition subsystem may be an ion beam output acquisition subsystem. Figure 2 The output acquisition subsystem is configured as shown in , except that the electron beam source can be replaced with any suitable ion beam source known in the art. In addition, the output acquisition subsystem can include any other suitable ion beam output acquisition system, such as the ion beam output acquisition system included in a commercially available focused ion beam (FIB) system, a helium ion microscope (HIM) system, and a secondary ion mass spectrometry (SIMS) system.

[0051] Figure 3 Another embodiment of a system including various light-based output acquisition subsystems is illustrated. Figure 3 The output acquisition subsystem shown in FIG. 1 is incorporated by reference as if fully set forth herein. Figure 3 Some of the non-essential details of the system presented in this patent are omitted from the corresponding description. However, it will be understood that the system can be further configured as described in this patent. Figure 3In addition, it will be apparent after reading the description of several embodiments provided herein that the Figure 3 The system illustrated in is an improvement on the system described in US Pat. No. 6,515,746 to Opsahl et al. The modification includes compensating for temperature-induced changes in the location of the target.

[0052] One of the output acquisition subsystems is configured as a broadband reflectance spectrometer. The broadband reflectance spectrometer (BRS) 230 utilizes multiple wavelengths of light to simultaneously detect the sample 226. The BRS 230 uses a lens 232 and includes a broadband spectrometer 234, which can be any type commonly known and used in the art. The lens 232 can be a transmissive optical component formed of a material such as calcium fluoride (CaF2). This lens can be a spherical microscope objective with a high numerical aperture (approximately 0.90NA) to produce a large range of incident angles relative to the sample surface and produce a spot size of approximately one micron in diameter. Alternatively, the lens 232 can be a reflective optical component. This lens can have a lower numerical aperture (approximately 0.4NA) and can be capable of focusing light to a spot size of approximately 10 microns to 15 microns. Figure 3 The spectrometer 234 shown in FIG. 2 includes a lens 236, an aperture 238, a dispersive element 240, and a detector array 242. The lens 236 may be formed of CaF2.

[0053] During operation, a probe beam 244 from a light source 246 is collimated by a lens 245, directed by a mirror 243 through a mirror 266 to a mirror 286, which directs the light through a mirror 248 to a lens 232, which is then focused by the lens 232 onto the sample 226. The light source may include any of the light sources described above. The lens 245 may be formed of CaF2.

[0054] Light reflected from the surface of the sample passes through lens 232 and is directed by mirror 248 (through mirror 250) to spectrometer 234. Lens 236 focuses the probe beam through aperture 238, which defines a spot in the field of view on the sample surface for analysis. Dispersive element 240, such as a diffraction grating, prism, or hologram, scatters the beam angularly, depending on wavelength, to individual detector elements contained in detector array 242.

[0055] Different detector elements preferably measure the optical intensity of different wavelengths of light contained in the probe beam simultaneously. Alternatively, the detector 242 may be a charge coupled device ("CCD") camera or a photomultiplier with suitable dispersive or otherwise wavelength selective optics. It should be noted that a monochromator may be used to measure different wavelengths (one at a time) sequentially using a single detector element. In addition, the dispersive element 240 may also be configured to disperse the light in one direction as a function of wavelength and in an orthogonal direction as a function of the angle of incidence relative to the sample surface, so that simultaneous measurement as a function of both wavelength and angle of incidence is possible. The computer subsystem 252 processes the intensity information measured by the detector array 242.

[0056] Broadband spectroscopic ellipsometer (BSE) 254 is also configured to perform measurements on the sample using light. BSE 254 includes polarizer 256, focusing mirror 258, collimating mirror 260, rotation compensator 262, and analyzer 264. In some embodiments, BSE 254 may be configured to perform measurements on the sample using light provided by light source 246, light source 283, or another light source (not shown).

[0057] In operation, mirror 266 directs at least a portion of the probe beam 244 to polarizer 256, which produces a known polarization state, preferably linear polarization, for the probe beam. Mirror 258 focuses the beam onto the sample surface at an angle of inclination, ideally about 70 degrees to the normal to the sample surface. Based on the well-known principles of ellipsometry, the reflected beam will typically have a mixed linear and circular polarization state after interacting with the sample, based on the composition and thickness of the sample's film 268 and substrate 270.

[0058] The reflected beam is collimated by mirror 260, which directs the beam to a rotating compensator 262. Compensator 262 introduces a relative phase delay δ (phase retardation) between a pair of mutually orthogonal polarized beam components. Compensator 262 is preferably rotated at an angular velocity c about an axis substantially parallel to the direction of beam propagation by an electric motor 272. Analyzer 264, preferably another linear polarizer, mixes the polarization states incident thereon. By measuring the light transmitted by analyzer 264, the polarization state of the reflected probe beam can be determined.

[0059] Mirror 250 directs the beam to spectrometer 234, which simultaneously measures the intensity of different wavelengths of light in the reflected probe beam through the compensator / analyzer combination. Computer subsystem 252 receives the output of detector 242 and processes the intensity information measured by detector 242 as a function of wavelength and as a function of the azimuth (rotation) angle of compensator 262 about its rotation axis to solve for ellipsometric values ​​ψ and Δ as described in U.S. Pat. No. 5,877,859 to Aspnes et al., which is incorporated by reference as if fully set forth herein.

[0060] A system including the broadband reflectance spectrometer and broadband spectroscopic ellipsometry described above may also include additional output acquisition subsystems configured to perform additional measurements on the sample using light. For example, the system may include an output acquisition subsystem configured as a beam profile ellipsometry, a beam profile reflectometer, another optical subsystem, or a combination thereof.

[0061] A beam profile ellipsometry (BPE) is discussed in U.S. Pat. No. 5,181,080 to Fanton et al., which is incorporated by reference as if fully set forth herein. BPE 274 includes a laser 283 that produces a probe beam 284. Laser 283 may be a solid-state laser diode from Toshiba Corp. that emits a linearly polarized 3 mW beam at 673 nm. BPE 274 also includes a quarter wave plate 276, a polarizer 278, a lens 280, and a four-quadrant detector 282. In operation, the linearly polarized probe beam 284 is focused on sample 226 by lens 232. Light reflected from the sample surface passes upward through lens 232 and mirrors 248, 286, and 288, and is directed into BPE 274 by mirror 290.

[0062] The positioning of the rays within the reflected probe beam corresponds to a particular angle of incidence relative to the sample surface. The quarter wave plate 276 retards the phase of one of the polarization states of the beam by 90 degrees. The linear polarizer 278 causes the two polarization states of the beam to interfere with each other. For maximum signal, the axis of the polarizer 278 should be oriented at an angle of 45 degrees relative to the fast and slow axes of the quarter wave plate 276. The detector 282 is a four-element detector having four radially disposed quadrants, each quadrant intercepting a quarter of the probe beam and producing a separate output signal proportional to the power of the portion of the probe beam impinging on that quadrant.

[0063] The output signal from each quadrant is sent to the computer subsystem 252. By monitoring the change in the polarization state of the beam, elliptical polarization information such as ψ and Δ can be determined. To determine this information, the computer subsystem 252 takes the difference between the sums of the output signals of diametrically opposed quadrants, which is a value that varies linearly with film thickness for very thin films.

[0064] A beam profile reflectometer (BPR) is discussed in U.S. Pat. No. 4,999,014 to Gold et al., which is incorporated by reference as if fully set forth herein. The BPR 292 includes a laser 283, a lens 294, a beam splitter 296, and two linear detector arrays 298 and 300 to measure the reflectivity of a sample. In operation, a linearly polarized probe beam 284 is focused by lens 232 onto sample 226, and the various rays within the beam strike the sample surface at a range of incident angles. Light reflected from the sample surface passes upward through lens 232 and mirrors 248 and 286, and is directed by mirror 288 into the BPR 292. The positioning of the rays within the reflected probe beam corresponds to a particular angle of incidence relative to the sample surface. Lens 294 spatially expands the beam in two dimensions. The beam splitter 296 separates the S component of the beam from the P component, and the detector arrays 298 and 300 are oriented orthogonally to each other to isolate information about the S polarization and the P polarization. Higher incident angle rays will fall closer to the opposite end of the array. The output from each element in the diode array will correspond to a different incident angle. The detector arrays 298 and 300 measure the intensity across the reflected probe beam as a function of the incident angle relative to the sample surface. The computer subsystem 252 receives the output of the detector arrays 298 and 300 and derives the thickness and refractive index of the thin film layer 268 based on these angle-dependent intensity measurements by utilizing various types of modeling algorithms. An optimization routine using an iterative process, such as a least squares fitting routine, is typically employed.

[0065] Figure 3 The system shown in may also include additional components, such as a detector / camera 302. The detector / camera 302 is positioned above the mirror 290 and may be used to observe the reflected beam of the sample 226 for alignment and focusing purposes.

[0066] To calibrate BPE 274, BPR 292, BRS 230, and BSE 254, the system may include a wavelength-stable calibration reference ellipsometer 304 used in conjunction with a reference sample (not shown). For calibration purposes, the reference sample ideally consists of a thin oxide layer having a thickness d formed on a silicon substrate. However, in general, the sample may be any suitable substrate of known composition, including bare silicon wafers and silicon wafer substrates having one or more thin films thereon. The thickness d of the layer need not be known or consistent between periodic calibrations.

[0067] Ellipsometer 304 includes light source 306, polarizer 308, lenses 310 and 312, rotating compensator 314, analyzer 316, and detector 318. Compensator 314 is preferably rotated by electric motor 322 at an angular velocity ψ about an axis substantially parallel to the direction of propagation of beam 320. It should be noted that the compensator can be located between the sample and the analyzer (e.g., Figure 3 ) or between the sample and polarizer 308. It should also be noted that the construction of polarizer 308, lenses 310 and 312, compensator 314, and polarizer 316 are optimized for the specific wavelength of light produced by light source 306, which maximizes the accuracy of the ellipsometer.

[0068] Light source 306 produces a quasi-monochromatic probe beam 320 having a known stable wavelength and a stable intensity. This can be done passively, where light source 306 produces a very stable output wavelength that does not vary over time (i.e., varies by less than 1%). An example of a passive stable light source is a helium-neon laser or other gas discharge laser system. Alternatively, a non-passive system can be used, where the light source includes a light generator (not shown) that produces light with a wavelength that is not precisely known or stable over time, and a monochromator (not shown) that accurately measures the wavelength of the light produced by the light generator. Examples of such light generators include laser diodes, or polychromatic light sources used in conjunction with color filters such as gratings. In either case, the wavelength of beam 320, which is either a known constant or measured by a monochromator, is provided to computer subsystem 252 so that ellipsometer 304 can accurately calibrate the optical measurement devices in the system.

[0069] The operation of the ellipsometer 304 during calibration is further described in U.S. Patent No. 6,515,746. Briefly, beam 320 enters detector 318, which measures the intensity of the beam that has passed through the compensator / analyzer combination. Computer subsystem 252 processes the intensity information measured by detector 318 to determine the polarization state of the light after it interacts with the analyzer, and therefore determines the ellipsometric parameters of the sample. This information processing includes measuring the intensity of the beam as a function of the azimuth (rotation) angle of the compensator about its axis of rotation. This measurement of the intensity as a function of the compensator rotation angle is actually a measurement of the intensity of beam 320 as a function of time, because the compensator angular velocity is typically known and constant.

[0070] By knowing the composition of the reference sample and by knowing the exact wavelength of light produced by light source 306, the optical properties of the reference sample, such as film thickness d, refractive index, and extinction coefficient, can be determined by ellipsometer 304. Once the thickness d of the film has been determined by ellipsometer 304, the same sample is probed by other optical measurement devices BPE 274, BPR 292, BRS 230, and BSE 254 that measure various optical parameters of the sample. Computer subsystem 252 then calibrates the processing variables used to analyze the results from these optical measurement devices so that the processing variables produce accurate results. In the calibration technique described above, all system variables that affect phase and intensity are determined and compensated using the phase shift and reflectivity normalization factors discussed in U.S. Pat. No. 6,515,746, so that the optical measurements made by these calibrated optical measurement devices become absolute.

[0071] The calibration techniques described above are largely based on calibration using the derived thickness d of the thin film. However, calibration using the ellipsometer 304 can be based on any optical property of a reference sample that can be measured or determined by the ellipsometer 304 and / or is otherwise known, whether the sample has a single layer of film on it, multiple layers of films on it, or even no film on it (bare sample).

[0072] In some embodiments, the output acquisition subsystem may have at least one common optical component. For example, lens 232 is common to BPE 274, BPR 292, BRS 230, and BSE 254. In a similar manner, mirrors 243, 266, 286, and 248 are common to BPE 274, BPR 292, BRS 230, and BSE 254. Figure 3 , the ellipsometer 304 does not have any optical components common to the other output acquisition subsystems. This separation from the other output acquisition subsystems may be appropriate because the ellipsometer is used to calibrate the other output acquisition subsystems.

[0073] As further described above, the output acquisition subsystem can be configured to have multiple modes. In general, a "mode" is defined by the parameter values ​​of the output acquisition subsystem used to generate the sample output. Therefore, different modes can be different in the value of at least one of the output generation parameters of the output acquisition subsystem (in addition to the positioning of the output on the sample). For example, for a light-based output acquisition subsystem, different modes can use different wavelengths of light. As further described herein, the modes can be different in the wavelength of light directed to the sample (e.g., for different modes, by using different light sources, different spectral filters, etc.). In another embodiment, different modes can use different illumination channels. For example, as described above, the output acquisition subsystem can include more than one illumination channel. In this way, different illumination channels can be used for different modes.

[0074] The multiple modes may also differ in illumination and / or collection / detection. For example, as further described above, the output acquisition subsystem may include multiple detectors. Thus, one of the detectors may be used for one mode and another of the detectors may be used for another mode. Furthermore, the modes may differ from each other in more than one manner described herein (e.g., different modes may have one or more different illumination parameters and one or more different detection parameters). Additionally, the multiple modes may differ in viewing angle, meaning either or both of different angles of incidence and collection angles, which may be achieved as further described above. The output acquisition subsystem may be configured to scan a sample using different modes in the same scan or in different scans, for example, depending on the ability to scan a sample simultaneously using multiple modes.

[0075] In some examples, the systems described herein may be configured as inspection systems. However, the systems described herein may be configured as another type of semiconductor-related quality control type system, such as a defect re-inspection system and a metrology system. For example, the systems described herein and in Figures 1 to 3 The embodiment of the output acquisition subsystem shown in the figure can be modified in one or more parameters to provide different output generation capabilities depending on the application in which the embodiment will be used. In one embodiment, the output acquisition subsystem is configured as an electron beam defect re-inspection subsystem. For example, if Figure 2 If the output acquisition subsystem shown in is to be used for defect re-inspection or metrology rather than for inspection, then the output acquisition subsystem can be configured to have a higher resolution. In other words, Figures 1 to 3 The embodiments of the output acquisition subsystem shown in the drawings describe some common and various configurations of the output acquisition subsystem, which can be adapted in a number of ways that are obvious to technicians in the field to produce output acquisition subsystems with different output generation capabilities that are more or less suitable for different applications.

[0076] As described above, the output acquisition subsystem may be configured to direct energy (e.g., light, electrons) to and / or scan energy across a physical version of the sample, thereby generating an actual output for the physical version of the sample. In this manner, the output acquisition subsystem may be configured as a "real" output acquisition system, rather than a "virtual" system. However, Figure 1 The storage medium (not shown) and computer system 102 shown in and / or other computer subsystems shown and described herein may be configured as a "virtual" system. Specifically, the storage medium and computer system 102 are not part of the output acquisition subsystem 100 and do not have any capabilities for handling physical versions of the sample, but may be configured as a virtual inspector that performs inspection-like functions, a virtual metrology system that performs metrology-like functions, a virtual defect re-inspection tool that uses stored detector outputs to perform defect re-inspection-like functions, etc. Systems and methods configured as "virtual" systems are described in commonly assigned U.S. Patent Nos. 8,126,255 to Bhaskar et al., issued February 28, 2012, 9,222,895 to Duffy et al., issued December 29, 2015, and 9,816,939 to Duffy et al., issued November 14, 2017, which are incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in these patents. For example, the computer subsystems described herein may be further configured as described in these patents.

[0077] One embodiment of a system configured to determine information of a sample includes an output acquisition subsystem configured to generate an output of the sample at one or more target positions on the sample. The one or more target positions may be predetermined target positions, which are stored in a recipe for a process performed on the sample using the output acquisition subsystem. The one or more target positions may also be discrete positions, such as discrete positions used in a move-acquire-measure type measurement process or discrete positions defined by an area on the sample to be scanned in an inspection or scanning type process. In this way, the target position indicates the position at which the process is to be performed on the sample, but as further described herein, even with the sample positioning methods and systems of the prior art, the process may not always be performed at the predetermined target position. As further described herein, the embodiments described herein intend to minimize the difference between the predetermined target position and the actual target position and / or correct the target position of the execution process. By predicting sample positioning errors due to temperature, the embodiments described herein can improve the performance and usefulness of the quality control type processes and tools described herein.

[0078] In one embodiment, the output acquisition subsystem is configured as a measurement subsystem. As described above, Figures 1 to 3 The output acquisition subsystem shown in the embodiment can be configured as a metrology subsystem. In this way, the embodiments described herein can be configured as a metrology tool. In the field of semiconductor metrology, a metrology tool can include an illumination subsystem that illuminates a target, a collection subsystem that captures relevant information provided by the interaction (or lack of interaction) of the illumination subsystem with the target, device or feature, and a computer subsystem that analyzes the collected information using one or more algorithms. The metrology tool can be used to measure structural and material properties associated with various semiconductor manufacturing processes (e.g., material composition, dimensional properties of structures and films, such as film thickness and / or critical dimension (CD) of the structure, overlap, etc.). These measurements are used to facilitate process control and / or yield efficiency in semiconductor die manufacturing.

[0079] The metrology tool may include one or more hardware configurations that may be used in conjunction with some of the embodiments described herein to, for example, measure the various aforementioned semiconductor structures and material properties. Examples of such hardware configurations include, but are not limited to, the following.

[0080] 1. Spectroscopic Ellipsometer (SE)

[0081] 2. SE with multiple lighting angles

[0082] 3. Measure the SE of the Mueller matrix elements (e.g., using a rotation compensator)

[0083] 4. Single wavelength ellipsometer

[0084] 5. Beam Profile Ellipsometer (Angle Resolved Ellipsometer)

[0085] 6. Beam profile reflectometer (angle-resolved reflectometer)

[0086] 7. Broadband reflectance spectrometer (spectral reflectometer)

[0087] 8. Single wavelength reflectometer

[0088] 9. Angle-resolved reflectometer

[0089] 10. Imaging system

[0090] 11. Scatterometer (e.g., spot analyzer)

[0091] A hardware configuration may be separated into discrete operating systems. Alternatively, one or more hardware configurations may be combined into a single tool. Figure 3One example of combining multiple hardware configurations into a single tool is shown in , which may be further configured as described in U.S. Patent No. 7,933,026 to Opsahl et al., which is incorporated by reference as if fully set forth herein. The system described herein may be further configured as described in such reference.

[0092] For example, Figure 3 A schematic diagram showing an exemplary metrology tool comprising: a) a broadband SE (i.e., 254); b) a SE (i.e., 304) with a rotational compensator (i.e., 314); c) a beam profile ellipsometer (i.e., 274); d) a beam profile reflectometer (i.e., 292); e) a broadband reflectance spectrometer (i.e., 230); and f) a deep ultraviolet reflectance spectrometer (i.e., 230). Additionally, there are typically numerous optical elements in such systems, including specific lenses, collimators, mirrors, quarter wave plates, polarizers, detectors, cameras, apertures, and / or light sources. The wavelength of the optical system can vary from about 120 nm to 3 microns. For non-ellipsometer systems, the collected signal can be polarization-resolved or non-polarized. Figure 3 An illustration of multiple metrology heads integrated on the same tool is provided. However, in many cases, multiple metrology tools are used for measurements on single or multiple metrology targets, as described, for example, in U.S. Pat. No. 7,478,019 to Zangooie et al., which is incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in such reference.

[0093] The illumination subsystem of a particular hardware configuration includes one or more light sources. The light source may generate light having only one wavelength (i.e., monochromatic light), light having several discrete wavelengths (i.e., polychromatic light), light having multiple wavelengths (i.e., broadband light), and / or light sweeping several wavelengths continuously or jumping between wavelengths (i.e., a tunable source or swept source). Examples of suitable light sources include, but are not limited to, a white light source, an ultraviolet (UV) laser, an arc lamp or an electrodeless lamp, a laser sustained plasma (LSP) source (such as those commercially available from Energetiq Technology, Inc., Woburn, Massachusetts), a supercontinuum source (such as a broadband laser source) (such as those commercially available from NKT Photonics Inc., Morganville, New Jersey), or a shorter wavelength source, such as an x-ray source, an extreme UV source, or some combination thereof. The light source may also be configured to provide light having sufficient brightness, which in some cases may be greater than about 1 W / (nm cm 2The measurement system may also include fast feedback to the light source for stabilizing the power and wavelength of the light source. The output of the light source may be delivered via free space propagation, or in some cases via optical fiber or any type of light guide.

[0094] Metrology tools can be designed to make many different types of measurements related to semiconductor manufacturing. Specific embodiments described herein may be applicable to such measurements. For example, in specific embodiments, the tool may measure characteristics of one or more targets, such as critical dimensions, overlay, sidewall angle, film thickness, process-related parameters (e.g., focus and / or dose). The target may include a specific region of interest that is periodic in nature, such as a grating in a memory die. The target may include multiple layers (or films) whose thicknesses may be measured by the metrology tool. The target may include a target design placed (or already present) on the sample for use with, for example, alignment and / or overlay registration operations. Specific targets may be located at different locations on the sample. For example, the target may be located within a scribe line (e.g., between dies) and / or within the die itself. In specific embodiments, multiple targets are measured (simultaneously or at different times) by the same one or more metrology tools, as described in U.S. Pat. No. 7,478,019 to Zangui et al. Data from such measurements may be combined. Data from metrology tools are used in semiconductor manufacturing processes, for example, to feed forward, feedback and / or feed side corrections to the process (eg, lithography, etching) and thus can produce a complete process control solution.

[0095] As semiconductor device pattern sizes continue to shrink, smaller metrology targets are generally required. In addition, measurement accuracy and matching with actual device characteristics increase the need for device-like targets and measurements within the die and even on the device. Various metrology implementations have been proposed to achieve the purpose. For example, a focused beam ellipsometer based on a primary reflective optical device is one of them and is described in U.S. Pat. No. 5,608,526 to Piwonka-Corle et al., which is generally incorporated by reference as if fully stated herein. The embodiments described herein may be further configured as described in this patent. Apodizers may be used to mitigate the effects of optical diffraction, which causes the illumination spot to expand beyond the size defined by the geometrical optics. The use of apodizers is described in U.S. Pat. No. 5,859,424 to Norton, which is generally incorporated by reference as if fully stated herein. The embodiments described herein may be further configured as described in this patent. Simultaneous multi-incident angle illumination using a high numerical aperture tool is another way to achieve small target capability. This technique is described, for example, in U.S. Patent No. 6,429,943 to Opsahl et al., which is incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in this patent.

[0096] Other measurement examples may include measuring the composition of one or more layers of a semiconductor stack or sample, measuring specific defects on (or within) a sample, and measuring the amount of lithographic radiation exposed to a sample. In some cases, metrology tools and algorithms may be configured for measuring non-periodic targets, see, for example, U.S. Pat. No. 9,291,554 to Kuznetsov et al., issued on Mar. 22, 2016, and U.S. Pat. No. 9,915,522 to Jiang et al., issued on Mar. 13, 2018, which are incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in these patents.

[0097] The measurement of the parameters of interest typically involves several algorithms. For example, the optical interaction of the incident beam with the sample is modeled using an EM (electromagnetic) solver and using algorithms such as RCWA, FEM, moment method, surface integration method, volume integration method, FDTD, and others. The target of interest is typically modeled (parameterized) using a geometry engine or, in some cases, a process modeling engine or a combination of the two. The use of process modeling is described in U.S. Pat. No. 10,769,320 to Kuznetsov et al., issued on September 8, 2020, which is incorporated by reference as if fully stated herein. The embodiments described herein may be further configured as described in this patent. For example, a geometry engine is implemented in KLA's AcuShape software product.

[0098] The collected data can be analyzed by several data fitting and optimization techniques and technologies including: libraries, fast reduced-order models; regression; machine learning algorithms, such as neural networks, support vector machines (SVM); dimensionality reduction algorithms, such as PCA (Principal Component Analysis), ICA (Independent Component Analysis), LLE (Local Linear Embedding); sparse representations, such as Fourier or wavelet transforms; Kalman filters; algorithms to facilitate matching from the same or different tool types, and others.

[0099] The collected data may also be analyzed by algorithms that do not include modeling, optimization, and / or fitting, such as described in U.S. Pat. No. 10,591,406 to Bringoltz et al., issued on Mar. 17, 2020, which is incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in this patent.

[0100] Computational algorithms are typically optimized for metrology applications using one or more methods such as design and implementation of computational hardware, parallelization, computation distribution, load balancing, multi-service support, dynamic load optimization, etc. Different implementations of the algorithms can be done in firmware, software, FPGA, programmable optical components, etc.

[0101] The data analysis and fitting steps typically pursue one or more of the following objectives:

[0102] 1. Measuring critical dimension (CD), sidewall angle (SWA), shape, stress, composition, film, bandgap, electrical properties, focus / dose, overlay, generation process parameters (e.g., resist state, partial pressure, temperature, focus model), and / or any combination thereof;

[0103] 2. Modeling and / or design of measurement systems; and

[0104] 3. Modeling, design and / or optimization of measurement objectives.

[0105] The embodiments described herein configured for use in the field of semiconductor metrology are not limited to the hardware, algorithm / software implementations and architectures and use cases outlined above.

[0106] In another embodiment, the output acquisition subsystem is configured as an inspection subsystem. The inspection subsystem can be configured to perform inspection using light, electrons, or another energy type (such as ions). For example, this output acquisition subsystem can be configured as Figure 1 and 2 . In a system where the output acquisition subsystem is configured as an inspection subsystem, the computer subsystem may be configured to detect defects on a sample based on the output generated by the output acquisition subsystem. For example, in the simplest possible scenario, the computer subsystem may subtract a reference from the output, thereby generating a difference signal or image and then applying a threshold to the difference signal or image. The computer subsystem may determine that any difference signal or image having a value above the threshold is a defect or potential defect and any difference signal or image having a value below the threshold is not a defect or potential defect. Of course, many defect detection methods and algorithms used on commercially available inspection tools are much more complex than this example, and any such method or algorithm may be applied to the output generated by the output acquisition subsystem configured as an inspection subsystem.

[0107] In a similar manner, the process may be a defect re-detection process. Unlike an inspection process, a defect re-detection process typically revisits discrete locations on a sample where a defect has been detected. An output acquisition subsystem configured for defect re-detection may generate a sample image as described herein, which may be used to determine one or more attributes of a defect (e.g., defect shape, size, roughness, background pattern information, etc.) and / or defect classification (e.g., bridge type defect, missing feature defect, etc.). For defect re-detection applications, the computer subsystem may be configured to use any suitable defect re-detection method or algorithm used on any suitable defect re-detection tool to determine information about a defect or sample from the output generated by the output acquisition subsystem.

[0108] In some embodiments, the output acquisition subsystem is a light-based subsystem. Figure 1 and 3 The output acquisition subsystem is further configured as shown in FIG. In other embodiments, the output acquisition subsystem is an electronic-based subsystem. Figure 2. Thus, from the above description, it can be seen that the embodiments described herein can be used with output acquisition subsystems having many different characteristics. In general, the embodiments described herein can be used in any tool where temperature can be measured near or in the hardware elements responsible for positioning the sample within the tool during run time and positioning errors can be measured while navigating across the sample during calibration.

[0109] The system also includes one or more temperature sensors configured to measure one or more temperatures within the system. In general, the temperature sensors included in the system and whose measured temperatures are used for positioning correction as further described herein may include any or all temperature sensors already installed on the system. In one such example, the temperatures may include any or all of the following temperatures: the Y front temperature recorded in the tool (also referred to herein as "Y_TEMP_FRONT"), the Y back temperature recorded in the tool (also referred to herein as "Y_TEMP_BACK"), the X temperature recorded in the tool (also referred to herein as "X_TEMP"), the carriage temperature recorded in the tool (also referred to herein as "CARRIAGE_TEMP"), the theta temperature recorded in the tool (also referred to herein as "T_TEMP"), and the Y plate temperature recorded in the tool (also referred to herein as "Y_PLATE_TEMP").

[0110] Y_TEMP_FRONT is the temperature of the first of the two linear motors working together in the Y direction of the sample handling stage. Y_TEMP_BACK is the temperature of the second of the two linear motors. X_TEMP is the temperature of the X-axis linear motor working in the X direction of the sample handling stage, and if the X-axis motor is a single linear motor (unlike the Y-axis motor, which can be two linear motors working together), there may be only one temperature in the X direction. CARRIAGE_TEMP can be measured by a temperature sensor mounted near or slightly below the sample chuck and can be located on the sample handling stage. T_TEMP can be equipped with a resistance temperature detector (RTD) near the θ axis winding. Y_PLATE_TEMP is the temperature of certain components of some sample handling stages.

[0111] Although these temperature measurements and temperature sensors may be useful for many tools, they are included herein only as illustrative examples and due to all possible variations of engines and other components that may be included in the sample handling subsystem, the inventors have intentionally chosen to illustrate the scanning subsystem and temperature sensors in a relatively abstract manner in the figures further described herein. In particular, the exact locations of the various temperature sensors for the various temperatures described above may not have a significant impact on the implementation of the embodiments described herein. As long as the temperature is measured at or near at least some (i.e., one or more) of the hardware elements of the scanning subsystem that are primarily responsible for sample positioning, it should be possible to train the DL model to predict the target position error based on the temperature measurement. If, through training and testing, the DL model does not exhibit satisfactory performance, the temperature sensor may be changed or supplemented with additional temperature sensors until a sensor configuration is found that provides sufficient data for the DL model to achieve sufficiently accurate predictions. This experimentation and testing is entirely within the skill of a technician in the field and will vary depending on the configuration of the tool.

[0112] In one embodiment, one or more temperature sensors are configured to measure one or more temperatures during a process performed on a sample using the output acquisition subsystem. For example, after the DL model is trained as described herein and is ready for use during runtime, one or more temperature sensors may collect temperature measurements while the sample is scanned or measured using the output acquisition subsystem. The sampling rate of the temperature sensor may be set based on how much the temperature change may cause the positioning accuracy to change and any possible information about how fast the temperature near the scanning subsystem may change. In one such non-limiting example, the sampling rate of the temperature sensor may be set so that a temperature change of 0.1°C in the tool will be detected. In this case, this would mean a sampling rate of approximately 1 sample every 30 seconds.

[0113] Depending on the configuration and capabilities of the tool, measuring temperature during a process can mean several different things. For example, once a sample is loaded into the tool and is ready to be measured or scanned, the temperature can be measured continuously. However, considering that the temperature can almost always be expected to change relatively slowly, continuous temperature measurement may not be required. As described above, a sampling rate of about 1 sample every 30 seconds may be sufficient. Whether those temperature measurements and error predictions are performed while the system performs another task such as measurement, output acquisition (e.g., imaging), sample positioning, etc. may vary depending on the data collection and processing capabilities of the system. For example, temperature measurements and error predictions can be interspersed with sample positioning and measurement in metrology, because the frequency of execution of temperature measurements and error predictions will be much lower than other functions. In other applications (such as inspections in which relatively large areas of samples can be scanned while continuously (or almost continuously) generating outputs), it may be most practical to obtain temperature measurements during scanning (meaning while performing output acquisition). In this way, measuring temperature during a process can mean measuring the temperature while the sample is measured, inspected, imaged, etc., or in the time between the start and end of the process.

[0114] In another embodiment, one or more temperature sensors are configured to measure one or more temperatures at or within one or more hardware elements of a system configured to alter the positioning of a sample within the system. In this manner, the temperature sensor may be configured to measure the temperature at one or more locations near or within the scanning subsystem described herein. In one such example, the system may include Figure 1 1 . The temperature sensors 40, 42, 44, and 46 shown in . The temperature sensors 40 and 42 are disposed below the stage 22 (i.e., on the side of the stage 22 opposite the sample 14). These temperature sensors may be directly coupled to the stage or they may be located within one or more additional elements (not shown) coupled to the stage, which may include mechanical and / or robotic elements configured to move the stage. These temperature sensors may be located at locations opposite to each other along a diameter of the stage, which defines, for example, the X-axis or Y-axis of the stage. These temperature sensors may also be located at other locations, such as along different diameters and therefore along different axes of the stage. The temperature sensors may also be located at different distances from the center of the stage. In some aspects, the exact positioning of the sensors may be determined based on the exact hardware configuration of the scanning subsystem (e.g., where within the hardware there is space that can accommodate the temperature sensors).

[0115] The temperature sensor 44 is shown as being located within the stage 22. This temperature sensor positioning is intended to show how the temperature sensor can be located within a hardware element of the scanning subsystem. In some instances, it may be practically useless to have the temperature sensor located within the stage. If the temperature sensor is located within the stage, it will obviously be preferable to configure it so that any contact between the temperature sensor and the sample is minimized to avoid the temperature sensor altering the sample in any way. A temperature sensor "within a hardware element" of the scanning subsystem may also not be completely contained within a single hardware element of the scanning subsystem. For example, such a temperature sensor may only be partially enclosed by a hardware element of the scanning subsystem. In addition, such a temperature sensor may be completely enclosed within a space formed between multiple hardware elements of the system. In this way, the temperature sensor may be within the scanning subsystem but not necessarily within any single hardware element of the scanning subsystem (although this is also possible).

[0116] exist Figure 1 2 shows a temperature sensor 46 coupled to an outer edge of the stage 22. If the configuration of the tool permits, this temperature sensor may be fixedly attached to the outer edge of the stage. Having the temperature sensor positioned at the outer edge of the stage (or at the outer edge of the scanning subsystem by virtue of this being attached to a different hardware element of the scanning subsystem) may be advantageous because it may be appropriate and even advantageous to measure one or more temperatures of the surrounding environment that are slightly outside the scanning subsystem and / or exposed to the tool housing (not just the scanning subsystem). Although this temperature sensor is shown attached to the stage, the temperature sensor may alternatively be positioned near the stage or another scanning subsystem hardware element without necessarily being attached thereto.

[0117] Figure 1 The temperature sensors shown in are intended to illustrate some possible advantageous locations for the temperature sensors but are by no means intended to limit the embodiments described herein to the temperature sensor configurations that may be used with the embodiments. Figure 1 In addition to the temperature sensor shown in , the system may also include one or more temperature sensors (not shown) that are spaced apart from the scanning subsystem, attached to the housing of the system, attached to one or more imaging elements of the system, configured to measure the temperature of the sample in a non-contact or contact manner, etc. In addition, although Figure 1Although four temperature sensors are shown in the embodiment and specific number of measured temperature examples are further described herein, it should be clear that the system may include any suitable number of temperature sensors arranged in any suitable configuration. In addition, the temperature sensors described herein may be any suitable commercially available sensors, which may be selected based on various information about the tool (e.g., expected temperature range, size considerations, expected measurement frequency, environmental conditions within the tool, etc.). In some examples, temperature sensors mounted on the tool for other purposes may be used to measure and record the various temperatures described herein and / or other temperatures that may be suitable for use by the embodiments described herein. For tools with pre-existing temperature sensors, one may attempt to train a DL model using the temperatures measured by those pre-existing sensors, and if the trained DL model does not pass testing and validation, then one may explore adding more temperature sensors as a way to improve the performance of the DL model.

[0118] The system further includes a computer subsystem configured to acquire one or more measured temperatures from one or more temperature sensors. The computer subsystem may include and have any configuration of any of the computer subsystems or systems described herein. Acquiring the one or more measured temperatures may be performed in the same manner as described above with respect to acquiring outputs generated by the output acquisition subsystem. For example, Figure 1 The computer subsystem 36 shown in can be coupled to the temperature sensor in any suitable manner (e.g., via one or more transmission media, which may include "wired" and / or "wireless" transmission media) so that the computer subsystem can receive temperature measurements generated by the temperature sensor. The computer subsystem 36 can be configured to perform a number of functions with or without the output of the temperature sensor, including the steps and functions further described herein. As further described herein, the one or more measured temperatures obtained by the computer subsystem may include a variety of information in addition to the actual temperature measurement, such as time and space information associated with the temperature measurement. As further described herein, the additional information can be generated and obtained.

[0119] A system consists of one or more components that are executed by a computer subsystem. For example, Figure 1 , the system includes one or more components 104 executed by the computer subsystem 36 and / or the computer system 102. The systems shown in other figures described herein may be configured to include similar elements. One or more components may be executed by the computer subsystem as further described herein or in any other suitable manner known in the art. Executing at least a portion of the one or more components may include inputting one or more inputs (e.g., acquired temperatures, data, etc.) into the one or more components. The computer subsystem may be configured to input any measured temperature data, etc., into the one or more components in any suitable manner.

[0120] The one or more components include a deep learning (DL) model configured to predict an error of at least one of the one or more target positions based on at least one of the one or more measured temperatures input to the DL model by the computer subsystem. Figure 1 , one or more components 104 include a DL model 106. The embodiments described herein utilize an understanding of domain knowledge, including the correlation between the temperature in the tool and the positioning error in X and / or Y. The one or more measured temperatures input to the DL model may include any information generated by or associated with the temperature measurement, such as time, space, and sensor ID information associated with the measured temperature or only the measured temperature itself. In addition, embodiments provide methods and systems for estimating positioning corrections without external tools and by using domain-specific knowledge of the process. The embodiments described herein enable positioning corrections using trained machine learning (ML) prescriptions based on temperature domain knowledge. By using site (i.e., target location) information as input to the DL model, a trained DL prescription can provide site-specific positioning corrections. In addition, the DL model can estimate a positioning correction map at run time to improve sample placement accuracy. Any or all of the temperature domain knowledge described herein can be used to train a DL prescription.

[0121] "Prediction error in at least one of the one or more target positions" can be defined in one of two ways, depending on which target position information is input to the DL model. If the target position information is a predetermined target position, then the prediction error is defined as the difference between the predetermined target position and the predicted position that would produce the output without any temperature-based correction. If the target position is a reported target position, then the prediction error is defined as the difference between the reported target position and the actual sample position predicted to have produced the output. "Prediction error" is also referred to herein as "positioning error", and these terms are used interchangeably herein. Corrections based on prediction errors as further described herein are referred to herein as "positioning corrections" and produce corrected positioning or "corrected target positions".

[0122] Obviously, since the DL model predicts the error of at least one target position, the information of at least one target position will be input to the DL model together with the temperature information. The number of target positions for which errors are predicted can vary greatly between samples and samples and / or between processes and processes. In general, errors can be predicted for a relatively sparse number of target positions across samples, such as 100 target positions versus hundreds or even a thousand target positions. The error predicted for any one target position can be used as further described herein to determine a corrected target position for the one target position and possible other target positions near the one target position.

[0123] Therefore, the target position information input to the DL model may vary depending on the target position itself and sometimes also depending on whether the DL model is being trained or being used at run time. For example, in an instance where an error is predicted for a target position specified in a prescription for a process in which the target position is scanned, imaged, measured, etc., the target position information input to the DL model during training and run time may be a subset of all target positions in the process prescription. Although the target position information input to the DL model during training and run time may be the same for any one sample and process, this is not necessary. For example, more target position information may be input for training purposes compared to run time. The situation may also be the opposite, meaning that the target position information used for training purposes can be used for fewer target positions than the number of target positions for which errors will be predicted at run time. The target position information input for training can also be arbitrary, because it can be used for a set of arbitrarily selected sample positions. This target location information can be used for training in instances such as when the exact target location is not known prior to performing the process, which can be the case for processes such as inspection where a relatively large area on a sample is scanned and errors are then predicted in the reported scanned locations or simply where defects are detected.

[0124] The target position information input to the DL model during training and run time can be in any suitable file or format. For example, the target position information can be coordinates defined relative to a sample, an output acquisition subsystem, an alignment target on a sample, an alignment target on an element of the output acquisition subsystem, a design or patterned feature formed on the sample, and the like. In general, it will be useful to train the DL model to predict the error of the target position coordinates to be used in the process performed on the sample. However, as in any other target positioning system or method, the coordinates of the target position in one space can be converted to the coordinates of the target position in another space using any suitable coordinate transformation relationship in any suitable manner.

[0125] The inventors have numerically demonstrated that the positioning errors recorded during movement along the X and Y axes of the stage are correlated with temperature changes within the tool. Specifically, the inventors have established correlations between the various temperatures listed herein and the errors in the X and Y directions. For different sites on the sample ("Site A" and "Site B"), the correlations between the measured temperatures and the measured positioning errors in X are shown in Table 1:

[0126]

[0127] For different sites on the sample ("Site A" and "Site B"), the correlation between the measured temperature and the measured positioning error on Y is shown in Table 2:

[0128]

[0129] As can be seen from Tables 1 and 2, the inventors have found that the correlation assumption holds across several sites and that there is intra-sample variation in the correlation. The temperature in the Y direction is generally more correlated with the error than the X direction. There are two relatively highly correlated temperatures for the error in the X direction: the carriage temperature and the Y plate temperature. For the Y direction, all temperatures show a relatively strong relationship with the error.

[0130] In one embodiment, the DL model is configured to predict errors of a first target position and a second target position in one or more target positions based on at least one of the one or more measured temperatures input to the DL model, and the computer subsystem is configured to determine a first corrected target position and a second corrected target position of a first target position and a second target position in one or more target positions by applying the predicted errors of the first target position and the second target position in the one or more target positions to the first target position and the second target position in the one or more target positions, respectively. For example, the errors can be predicted in several different ways and then used to determine the corrected target positions. In one example, an overall or global error can be predicted for more than one (or even all) target positions on a sample, and then the same predicted error can be applied to each target position to determine the corrected target position of each target position. However, different errors can be predicted for different target positions or different subsets of target positions, and then the errors predicted only for the target position or only for a subset of target positions including the target position can be used to determine the corrected target position of any one target position. How the error is predicted and used to determine the corrected target position may depend on the configuration of the temperature sensor and how the temperature sensor measures temperature during scanning or processing.

[0131] In some examples, different measured temperatures can be used to predict different errors for different targets or different target subsets. For example, one temperature type can be used to predict the error of a target position on one part of a sample, and another temperature type can be used to predict the error of a target position on another part of a sample. However, in other examples, the same temperature type can be used to predict the error of multiple target positions or even all target positions, and some of those prediction errors can be determined using the same temperature type measured at different positions. Different positions can be defined relative to a sample, one or more hardware elements of an output acquisition subsystem, or some other reference coordinate system. For example, when a sample is scanned by an output acquisition subsystem, the temperature can be measured by a temperature sensor, and the position information of the temperature measurement can be defined based on the positioning of the temperature sensor relative to the sample during the temperature measurement, the positioning of the sample in the output acquisition subsystem during the temperature measurement, etc. In this way, the temperature can be measured as a function of time and position and used to predict the error of the target position. In such examples, the DL model can be trained using position information and the measurement information generated during the training scan, as further described herein.

[0132] In order to provide an accurate estimate of the positioning error at the site level, it is necessary to feed the positioning data and the associated temperature data to the DL estimator. In another embodiment, the computer subsystem is configured to obtain the position information of the one or more measured temperatures in response to measuring one or more locations of the one or more measured temperatures, and the position information of at least one of the one or more measured temperatures is input to the DL model by the computer subsystem together with at least one of the one or more measured temperatures. The computer subsystem can obtain the position information in several different ways. One way is that if the temperature sensor moves in some way, it can generate position information, such as measuring the temperature within the system based on the positioning of the temperature sensor. In this way, the computer subsystem can simply obtain the position information by receiving the position information from the temperature sensor. In other examples, the temperature sensor may have a fixed positioning relative to the output acquisition hardware. Then, during the scanning of the sample, the temperature measurement generated by the temperature sensor can be recorded as a function of time, which can be used to relate the temperature measurement to the positioning of the sample relative to the output acquisition subsystem hardware (or vice versa) and therefore to the temperature sensor. The computer subsystem may input the position information with the measured temperature into the DL model, as further described herein. The DL model may be trained to predict the error of the target position using the position information with the measured temperature, as further described herein.

[0133] In one such embodiment, the DL model is configured to predict errors of a first target position and a second target position in one or more target positions based on at least one of the one or more measured temperatures and position information input to the DL model, and the computer subsystem is configured to determine a first corrected target position and a second corrected target position of the first target position and the second target position in the one or more target positions, respectively, by applying the predicted errors of the first target position and the second target position in the one or more target positions, respectively, to the first target position and the second target position in the one or more target positions. The DL model described herein has significant flexibility in predicting errors based on measured temperature alone or using position information. For example, if the input to the DL model includes only temperature information, the correction at inference may not be site-based. However, if the input to the DL model during training includes positioning data (e.g., in X and Y) and temperature, the DL model can provide site-based correction at inference. The position information input to the DL model can vary greatly, but in general, the position data does not need to be very dense (e.g., thousands or even hundreds) as long as the input is reasonably sparse data (such as 100 temperatures measured at different locations).

[0134] Machine learning (ML) can be generally defined as a type of artificial intelligence (AI) that provides computers with the ability to learn without being explicitly programmed. ML focuses on the development of computer programs that can teach themselves to grow and change when exposed to new data. In other words, ML can be defined as a subfield of computer science that "gives computers the ability to learn without being explicitly programmed." ML explores the study and construction of algorithms that can learn from and make predictions on data - such algorithms overcome following completely static program instructions by making data-driven predictions or decisions by building models from sample inputs.

[0135] The ML described herein may be further performed as described in Morgan Kaufmann, "Introduction to Statistical Machine Learning" by Sugiyama, 2016, p. 534; "Discriminative, Generative, and Imitative Learning" by Jebara, MIT Thesis, 2002, p. 212; and "Principles of Data Mining (Adaptive Computation and Machine Learning)" by Hand et al., MIT Press, 2001, p. 578; the documents are incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in these references.

[0136] In general, "deep learning" (DL) (also known as deep structured learning, hierarchical learning, or deep ML) is a branch of ML based on a set of algorithms that attempt to model high-level abstract concepts in data. In a simple case, there may be two groups of neurons: one group that receives input signals and one group that sends output signals. When an input layer receives input, it passes a modified version of the input to the next layer. In DL-based models, there are typically many layers between input and output (and the layers are not made up of neurons but they can help to think like neurons), allowing the algorithm to use multiple processing layers consisting of multiple linear and nonlinear transformations.

[0137] In one embodiment, the DL model includes a linear model. For example, the DL model can be a multi-output linear regressor. The linear relationship between temperature and positioning error can be described by the following function:

[0138]

[0139] where δx is the estimated positioning error in the x direction, ΔT i is the change in temperature relative to some baseline steady-state value (recorded during calibration), x is the position of the stage in the X direction, y is the position of the stage in the Y direction, and the coefficient c i are the coefficients fitted to the calibration data.

[0140] In another embodiment, the DL model comprises a neural network. The neural network can be configured as a generative model. A "generative" model can generally be defined as a model that is probabilistic in nature. In other words, a "generative" model is not a model that performs forward simulation or rule-based methods and as such, a model of the physics of the process involved is not necessary. Instead, as further described herein, a generative model can be learned based on a suitable training data set (because its parameters can be learned).

[0141] In one embodiment, the neural network is configured as a deep generation model. For example, the model may be configured to have a DL architecture because the model may include multiple layers that perform several algorithms or transformations. In another embodiment, the neural network may be a deep neural network with a set of weights that models the world based on data that has been fed to train the deep neural network. Neural networks can generally be defined as computational methods based on a relatively large set of neural units that loosely model the way biological brains solve problems using relatively large biological neuron clusters connected by axons. Each neural unit is connected to many other neural units, and the connection can strengthen or inhibit its effect on the activation state of the connected neural unit. These systems are self-learning and training rather than explicitly programmed and perform well in areas where solutions or feature detection are difficult to express with traditional computer programs.

[0142] Neural networks are typically composed of multiple layers, with signal paths traversing from front to back. The goal of a neural network is to solve problems in the same way as the human brain, but several neural networks are more abstract. Modern neural network engineering typically works with thousands or even millions of neural units and millions of connections. A neural network can have any suitable architecture and / or configuration known in the art.

[0143] In another embodiment, the DL model includes a convolutional neural network (CNN). The CNN may include any suitable type of layer, such as convolution, pooling, fully connected, soft maximum, etc., with any suitable configuration known in the art. The CNN may be trained as described herein or in any other suitable manner known in the art. The DL model may have any suitable CNN configuration or architecture known in the art.

[0144] In another embodiment, the computer subsystem is configured to train the DL model using a training set comprising training inputs and training outputs. The training inputs comprise one or more temperatures measured by one or more temperature sensors during a training scan, and the training outputs comprise measured errors in at least a portion of one or more target locations during the training scan. For example, the computer subsystem may construct a DL model recipe to estimate positioning errors, as described herein.

[0145] Figure 4 The diagram illustrates a flow chart for training a DL model. These steps can be performed during the setup phase and / or the calibration phase. Figure 4 As shown in , in step 400, the computer subsystem may generate training input. Generating training input may include two steps, the first step is to change the temperature in the system, as shown in step 402. For example, this step may include heating and / or cooling a robot that is at least partially responsible for changing and controlling the positioning of the sample in the tool. Changing the temperature in the system can be performed in any suitable manner known in the art. Generating training input may also include measuring the temperature in the system, as shown in step 404. For example, when the robot is heated and / or cooled to intentionally change the temperature in the system, the temperature may be continuously measured, which may include any one or more of the temperatures described herein. In this way, the computer subsystem can obtain temperature information from the temperature sensor and measure various temperature data in the tool at different time frames. Ideally, at this stage, the measured and recorded temperatures may include the temperatures measured by all temperature sensors included in the system.

[0146] In some embodiments, the training input includes location information of one or more temperatures measured during the training scan in response to one or more locations where the one or more temperatures are measured during the training scan. For example, measuring the temperature within the system as shown in step 404 may include measuring the temperature as a function of positioning within the system or as a function of positioning relative to the sample. Therefore, the result of measuring the temperature within the system during the training scan may include the measured temperature and the location information associated with the measured temperature and any other data generated during the temperature measurement. In this way, generating the training input may include generating any and all information associated with the temperature measurement, including the measured temperature itself and other data such as location, time, or information. Therefore, the training input may include positioning data on X and Y as well as the measured temperature. Any and all of this information may be included in the training input, which is added to the training set used to train the DL model.

[0147] As shown in step 406, the computer subsystem may generate a training output. This step may also include multiple steps, including moving over a known position on the sample as shown in step 408, and determining an offset (error) from a measured position to a target position as shown in step 410. These steps are intended to measure the robot movement error by moving over a known position (site) on the sample and finding the offset (or error in positioning) from the measured position to the target position. This step may also include collecting stage sample map data for different time frames. For example, the error may be measured using repeated stage mapping, which may be performed using a stage mapping application further described herein. When the tool changes temperature, any differences between consecutive stage maps may then be attributed to thermal differences and are an indication of thermal errors in the positioning of the target position. The position information used for training may be input to the DL model as described above and may include any of the position information described above.

[0148] Generating training inputs and training outputs may be repeated n times, depending on how much data is expected to be needed to train the DL model. The amount of data required for training may vary from tool to tool and from DL model architecture to DL model architecture. For example, if the DL model uses the linear relationship described above, then in the simplest case, training may consist of finding the coefficients c i , the coefficients minimize the sum of squared errors when predicting δx from the calibration data. To perform this least squares fit, at least n calibration data points may be needed to fit. In general, DL model training may be performed, at least initially, with as much training data as can be reasonably or practically collected with the existing temperature sensor or initial temperature sensor configuration. The testing and validation phases of training will provide an indication of whether the amount of training data is sufficient. If not, then additional training data may be collected as described above, and the additional training data may be used for retraining or added to the initial training data and then used for retraining.

[0149] The result of generating the training input and training output is a training set 412, which can then be used in step 414 to train the DL model. Training can be performed in this step as further described herein, but typically includes training the DL model to estimate positioning errors based on inputs including a given target position and temperature. In addition, training can include training the DL model to output positioning corrections at the site level. Depending on how the trained DL model performs in testing and validation or use, the computer subsystem can measure positioning correction maps at unseen time frames to improve sample placement accuracy using the errors predicted by the DL model. This step can be performed before the trained DL model is released for use and / or after a certain period of time that the DL model has been used for monitoring and / or recalibration purposes. The training step can produce a trained DL model 416, which can be used as further described herein.

[0150] In one embodiment, at least one of the one or more measured temperatures input to the DL model by the computer subsystem includes all one or more temperatures measured by all one or more temperature sensors during a process performed on the sample using the output acquisition subsystem. For example, to demonstrate the effect of temperature data on the accuracy of the positioning corrections Dx, Dy, two different ML models were tested: Model A, all temperatures and functions of positioning X and Y; and Model B, only two highly correlated temperatures and functions of positioning X and Y.

[0151]

[0152] Model B:(D x ,D y )

[0153] =f(X,Y,CARRIAGE_TEMP,Y_PLATE_TEMP)

[0154] Error plots (contour plots of error in X and Y) for positioning corrections for two different models were generated but are not included herein for purposes of reproducibility and clarity only. The error plots show that the accuracy of estimating positioning corrections for unseen time period frame data (i.e., blind test data not used for DL ​​model training but used to demonstrate that the trained model does meet the performance criteria for new data) degrades compared to the case where all temperature information is used.

[0155] Training a DL model with all temperatures measured by all temperature sensors can be advantageous for several reasons. One reason is that for the tools described herein, the mechanisms by which temperature affects positioning accuracy are not necessarily well understood. This is one of the reasons why DL models are such an advantageous way to predict positioning errors caused by temperature. More specifically, a DL model such as that described herein can learn the relationship between temperature and positioning error that is currently not understood. In other words, a DL model does not require a physics and / or thermodynamics-based model to learn how to predict target position errors based on temperature measurements.

[0156] Additionally, if all available temperature measurements from all available temperature sensors are used for training, the DL model can learn which temperature measurements from which sensors are useful for predicting target position error. In this way, the user does not need to determine which temperatures are useful or useless, and as long as the data collected from the temperature sensors does not inhibit tool performance, it should not matter whether useless temperature measurements are input to the DL model. Thus, by leveraging existing temperature sensors and / or by adding some additional temperature sensors and doing some judicious placement of them (e.g., near or within the hardware elements responsible for controlling sample positioning, rather than everywhere possible within the tool), the DL model itself can learn which temperature measurements are useful and which are useless.

[0157] If there is a problem with temperature measurement data collection, the parameters of the trained DL model can indicate which temperature measurements are useful and which are not. For example, by examining the weights assigned to the various inputs corresponding to the various temperature sensors in the DL model, the user can identify any temperature measurements that are not used or are minimally useful for predicting target position error and then optionally choose to eliminate them from the DL model inputs. Thus, the embodiments described herein advantageously eliminate any need for the user to individually establish many possible paths from temperature changes to sample positioning errors. Instead, all temperatures that may have an impact on positioning error can be comprehensively measured and compensated at once.

[0158] Despite the fact that the DL model eliminates the need for a scientifically based model to predict positioning error from temperature measurements, the inventors have established the following examples of how temperature changes can affect positioning error for some commercially available tools used for their experiments and include them here to facilitate an understanding of how temperature can affect positioning accuracy and why the embodiments described herein are important for tools such as those described herein. The two most important identified connections between temperature changes of components of a microscope and the positioning accuracy of the microscope appear to be: 1) deflection of the Y-plate (part of the sample positioning robot) relative to the Y encoder, and 2) displacement of the support structure of the optics relative to the sample handling stage. In the first example, the Y-plate (the component that supports sample movement aligned with the tool's Y-axis) expands as it heats, but is constrained at several locations where it is fixed to other components. Because the Y-plate's expansion is limited, it may bend away from its center and misalign with the encoder (the component used to measure the positioning of the motor along this axis). In the second example, the support structure of the camera that measures the position on the sample may shift / deform slightly as its temperature changes. The positioning of the sample is measured using a camera (e.g., by imaging defined reference marks on the sample and using image processing to calculate where the marks are relative to where they should be), so any relative shift between the camera and the sample handling stage is converted into a positioning error.

[0159] The inventors also found that the quality of the training data can affect the performance of the DL model. For example, the DL model may be affected by different methods or different tools (i.e., data quality issues) for collecting phase map data. By improving the data quality, the large errors in the DL error prediction that may occur, for example, at sites near the edge of the sample, can be brought into the specified target range of the prediction error. Therefore, in some examples, the training of the DL model may include initial training with suitable training data, testing the DL model of the initial training, and determining whether the error range on the sample map is all within the specification or whether all sites on the sample are within the specification. If the error range exceeds the specification, additional training data may be generated, and the additional training data may include training data for the entire sample or only training data for the portion of the sample that is found to exceed the specification error range. Then, the additional training data can be used to retrain the DL model, and the DL model can be tested again to see if the prediction error range is all within the specification. In this way, the training of the DL model may include one or more training, testing, and validation cycles, wherein each cycle is performed using different or supplementary training data.

[0160] The DL model may or may not be trained by the computer subsystem and / or one of the components executed by the computer subsystem. For example, another method or system may train a DL model, which may then be stored for use as one of the components executed by the computer subsystem. In this way, the DL model may be trained on the tool or outside the tool. In either case, the training may include inputting the training input into the DL model and changing one or more parameters of the DL model until the output produced by the DL model matches (or substantially matches) the training output. The training may include changing any one or more trainable parameters of the DL model. The one or more parameters of the DL model being trained may include one or more weights of any layer of the DL model having trainable weights. In one such example, the weights may include weights for convolutional layers rather than pooling layers.

[0161] Figure 5 The diagram illustrates steps that may be performed by the embodiments described herein during run time. Temperature measurement, error prediction, and corrected target position determination may all be performed on the tool to monitor offset changes and sample placement improvements. Figure 5 , the computer subsystem may calculate the target position, as shown in step 500. The target position may or may not need to be calculated in this step, depending on how it is input to the computer subsystem. For example, in some instances, the computer subsystem actually calculates the target position at run time. In such instances, the recipe for the process contains the target position expressed in sample coordinates, and the computer subsystem converts this into sample handling coordinates. For example, this step may include a 30 degree axis rotation and one or more of the other alignment techniques described further herein.

[0162] As shown in step 502, a temperature sensor measures the temperature within the system. The temperature sensor may measure the temperature during run time, as further described herein. The calculated target location and the measured temperature are input to a trained DL model 504, which outputs a predicted error for the target location, as shown in step 506. In this way, the trained DL model may be applied to estimate the expected error when moving to a nearby location.

[0163] The computer subsystem is configured to determine a corrected target position of at least one of the one or more target positions by applying the prediction error to at least one of the one or more target positions. Figure 5As shown in step 508 of , the computer subsystem may determine a corrected target position by applying the prediction error from step 506 to the target position calculated in step 500. For example, applying the prediction error to at least one target position may include subtracting the prediction error from the target position, but may include a more complex function if a simple subtraction does not produce a sufficiently accurate result. The determining a corrected target position step may be performed by the computer subsystem using an algorithm, which may be part of one or more components executed by the computer subsystem or may be separate from those components.

[0164] In some embodiments, the computer subsystem is configured to determine a corrected target position for a second target position in one or more target positions by interpolating the prediction error determined for at least one of the one or more target positions to a second target position in the one or more target positions and applying the interpolated prediction error to the second target position in the one or more target positions. For example, as shown in step 510, the computer subsystem can interpolate the prediction error of a target position to another target position. As shown in step 512, the computer subsystem can also determine the corrected target position by applying the interpolated prediction error to another target position. In this way, the computer subsystem can be configured to interpolate the error estimate from the nearby site to the target position. Interpolating the prediction error from one site to another site can be performed in any suitable manner known in the art. Determining the corrected target position from the interpolated prediction error can be performed in the same manner as described herein for the prediction error.

[0165] As shown in step 514, the corrected target positions determined with or without interpolation may be combined into a set of corrected target positions and then used for positioning correction in one or more of the various ways described herein.

[0166] In one embodiment, the computer subsystem is configured to determine a corrected target position before the output acquisition subsystem generates an output at at least one of the one or more target positions. Figure 5 The system may move to a corrected target position and generate an output, as shown in step 516 of . In this way, the system may move to a corrected target position (e.g., target position-prediction error (or interpolated prediction error)), and then the output acquisition subsystem may image, scan, measure, etc. at the corrected target position. This sequence of steps may be most useful for a move-acquire-measure type system, in which measurements are intended to be performed at specific discrete locations on the sample and the measurements do not involve scanning a relatively large area on the sample.

[0167] In one such example, assume that a process involves measuring the space between two specific features in a die formed on a wafer and then measuring the size of other features spaced apart from the two specific features. In this example, the scanning subsystem will ideally be driven to the location of the space between the two specific features, the output acquisition subsystem will generate an output at the location, and then the scanning subsystem will move the sample so that the next one or several features to be measured are positioned in the field of view of the output acquisition subsystem. If the target position is not temperature corrected before measuring the space between the two features, the measurement may be performed at a different location. Then, the result of the process will lose the measurement or will include measurements of different locations that may be incorrectly interpreted as measurements of the space between the two features. Then, the result of the process will be incomplete or erroneous, which is disadvantageous for several obvious reasons. Therefore, correcting the target position for positioning errors caused by temperature changes as described herein can be advantageously used to prevent problems such as incomplete or erroneous sample measurements and other quality control related process results.

[0168] In another embodiment, the computer subsystem is configured to determine the corrected target position after the output acquisition subsystem generates output at at least one of the one or more target positions. Figure 5 As shown in step 518 of , the computer subsystem may modify the reported target location for a previously generated output. In this way, the output acquisition subsystem may generate an output of target location 1 on the sample, the DL model may predict the error of the target location based on one or more input temperature measurements (or may determine (e.g., interpolate) the error based on the predicted errors of nearby target locations), and the predicted error may then be used to correct the target location 1 coordinates.

[0169] Utilizing an output acquisition subsystem to generate output for a sample and then correcting the coordinates of the target location where the output is generated may be more suitable for applications such as inspection, where relatively large areas on the sample are scanned. For example, during inspection, the coordinates of any defects detected during the process may be recorded, the DL model may predict the error of the recorded defect location based on the temperature measured at or near the location where the defect was detected, and the predicted error may be used to correct the reported defect coordinates. This positioning correction may be less useful in applications where it is important to generate output at specific discrete locations on the sample, because it would be more advantageous to correct the target location before the output is generated to ensure that the specific location is imaged, measured, inspected, defect re-detected, etc.

[0170] In one embodiment, the DL model is configured to predict errors at one or more additional target locations on the sample or additional sample based on one or more additional temperatures measured by one or more additional temperature sensors of the additional system that are input to the DL model by the computer subsystem, at which additional outputs are generated by an additional output acquisition subsystem of the additional system. For example, if the DL model is trained for multiple tools, then the DL model can be used to estimate sample level positioning errors and corrections for each tool. In this case, each of the output acquisition subsystems and other components of the system (such as temperature sensors) can be configured as described herein. Each of the systems can have the same or at least similar configuration. For example, each of the systems can be as described herein. Figure 3 Training the DL model for multiple tools may additionally be performed as described herein.

[0171] In another embodiment, the training input includes one or more additional temperatures measured by one or more additional temperature sensors included in the additional system during additional training scans of the sample or additional samples performed with the additional system, and the training output includes additional measured errors of one or more additional target locations on the sample or additional sample at which the output was generated by the additional output acquisition subsystem of the additional system during the additional training scan. For example, in a more general case, the DL model may be trained based on phase map and temperature data from multiple tools simultaneously. Training the DL model for multiple tools may include preparing sample phase map data from multiple tools and training the DL model across the tools. In this way, if the model (single model) is trained with multiple tools to generate training data, the same model may be used to estimate positioning errors across the tools.

[0172] In some embodiments, the computer subsystem is configured to monitor drift in the system and the additional system relative to each other based on the predicted error and the additional predicted error. In this way, the embodiments described herein can be configured to estimate tool drift in positioning error based on the error predicted by the DL model from the input measured temperature. For example, if the DL model is trained on phase diagrams and temperature data from multiple tools at the same time, the user can estimate the amount of drift for a particular tool by estimating the amount of positioning correction for each tool.

[0173] In another embodiment, the computer subsystem is configured to calibrate the system to an additional system based on the prediction error and the additional prediction error. In this way, the embodiments described herein can be configured for performing tool-to-tool calibration. For example, if the DL model is trained on phase maps and temperature data from multiple tools simultaneously, the user can estimate the tool-to-tool calibration by estimating the positioning correction for each tool.

[0174] The computer subsystem may also be configured to produce a result including the determined information, which may include any of the results or information described herein. The result of determining the information may be produced by the computer subsystem in any suitable manner. All embodiments described herein may be configured to store the result of one or more steps of the embodiment in a computer-readable storage medium. The result may include any of the results described herein and may be stored in any manner known in the art. The result including the determined information may have any suitable form or format, such as a standard file type. The storage medium may include any storage medium described herein or any other suitable storage medium known in the art.

[0175] After the results have been stored, the results may be accessed in a storage medium and used by any of the method or system embodiments described herein, formatted for display to a user, used by another software module, method or system, etc., to perform one or more functions on the sample or another sample of the same type. For example, the measured temperature, predicted error, and corrected target position may be stored and used as described herein or in any other suitable manner. In addition, depending on the process performed by the system on the sample, the results may include any information of the sample determined from the output generated by the output acquisition subsystem for the sample. Such results generated by the computer subsystem may include information of any defects detected on the sample, such as the location of the bounding box of the detected defect, etc., a detection score, information about the defect classification (such as a class label or ID), any defect attributes determined from any of the images, etc., sample structure measurements, size, shape, etc., or any such suitable information known in the art. The information may be used by the computer subsystem or another system or method for performing additional functions on the sample and / or the detected defects, such as sampling the defects for defect re-detection or other analysis, determining the root cause of the defects, etc.

[0176] Such functions also include, but are not limited to, changing processes such as production processes or steps that have been or will be performed on a sample in a feedback or feedforward manner. For example, the computer subsystem may be configured to determine one or more changes to the process performed on the sample and / or the process to be performed on the sample based on the determined information. The changes to the process may include any suitable changes to one or more parameters of the process. In one such example, the computer subsystem preferably determines those changes so that defects can be reduced or prevented on other samples on which the correction process is performed, defects on the sample can be corrected or eliminated in another process performed on the sample, defects can be compensated in another process performed on the sample, etc. The computer subsystem can determine such changes in any suitable manner known in the art.

[0177] Those changes may then be sent to a semiconductor fabrication system (not shown) or a storage medium (not shown) accessible to both the computer subsystem and the semiconductor fabrication system. The semiconductor fabrication system may or may not be part of the system embodiments described herein. For example, the output acquisition subsystem and / or computer subsystem described herein may be coupled to the semiconductor fabrication system, e.g., via one or more common elements (e.g., housing, power supply, sample handling device or mechanism, etc.). The semiconductor fabrication system may include any semiconductor fabrication system known in the art, such as a lithography tool, an etching tool, a chemical mechanical polishing (CMP) tool, a deposition tool, and the like.

[0178] In addition to the advantages described, the embodiments described herein have several advantages. For example, the embodiments described herein do not require external metrology tools to correct sample positioning. In particular, once the DL model is trained, the only data required for positioning correction is the information of the sample and the measured temperature information. No other external tools or methods are required to predict positioning errors, which is a significant benefit of the DL model described herein. In addition, once the DL model is trained, inferences for different time frames are essentially instant. In addition, the DL model and embodiments described herein can be generalized across multiple tools.

[0179] While the embodiments described herein do not require external metrology tools for sample positioning correction, the embodiments described herein may be used in conjunction with any other positioning control / correction methods or subsystems available on the tool. One such method may be a stage mapping application. In stage mapping, during tool calibration, positioning errors may be recorded as the sample is moved in an open loop to several positions across the sample (this positioning to error mapping is referred to as a "stage mapping"), and these (unchanged) error measurements may then be compensated during run time. Another such method is an alignment technique, in which each time a sample is loaded into the tool, a scanning subsystem navigates to a pair of reference marks near the center of the sample and across the Y-axis of the sample to measure the precise angle of the sample relative to the chuck, and then takes that angle into account when moving across the sample. An additional such method is a navigation correction technique, in which the positioning of a pair of reference marks around the edge of the sample is recorded and used to calculate a Euclidean transformation (e.g., translation, stretching, rotation, tilt), which is then used in a later move.

[0180] Each of the embodiments of each of the above described systems may be combined together into a single embodiment.

[0181] Another embodiment relates to a method for determining information of a sample. The method includes measuring one or more temperatures within a system using one or more temperature sensors. The system includes an output acquisition subsystem configured to generate an output of the sample at one or more target locations on the sample. The method also includes predicting an error of at least one of the one or more target locations by inputting at least one of the one or more measured temperatures into a DL model included in one or more components executed by a computer system. In addition, the method includes determining a corrected target location of at least one of the one or more target locations by applying the prediction error to at least one of the one or more target locations. The input and determination are performed by a computer system.

[0182] Each of the steps of the method may be performed as further described herein. The method may also include any other steps that may be performed by the system, temperature sensor, output acquisition subsystem, DL model, and computer system described herein. The system, temperature sensor, output acquisition subsystem, DL model, and computer system may be configured according to any of the embodiments described herein. The method may be performed by any of the systems described herein.

[0183] Additional embodiments relate to a non-transitory computer-readable medium storing program instructions executable on a computer system to perform a computer-implemented method for determining information of a sample. Figure 6 In particular, as shown in Figure 6 , a non-transitory computer readable medium 600 includes program instructions 602 executable on a computer system 604. The computer-implemented method includes obtaining one or more temperatures measured within the system using one or more temperature sensors. The system is configured as further described herein. The computer-implemented method also includes the prediction and determination steps described above. The computer-implemented method may further include any steps of any method described herein.

[0184] Program instructions 602 implementing methods such as those described herein may be stored on computer readable medium 600. The computer readable medium may be a storage medium such as a magnetic or optical disk, tape, or any other suitable non-transitory computer readable medium known in the art.

[0185] The program instructions may be implemented in any of a variety of ways, including procedural, component-based, and / or object-oriented techniques, among others. For example, the program instructions may be implemented using ActiveX controls, C++ objects, JavaBeans, Microsoft Foundation Classes ("MFC"), SSE (Streaming SIMD Extensions), or other techniques or methods, as desired.

[0186] Computer system 604 may be configured according to any of the embodiments described herein.

[0187] In view of this description, other modifications and alternative embodiments of various aspects of the present invention will be apparent to those skilled in the art. For example, methods and systems for determining the information of a sample are provided. Therefore, this description should be considered as merely illustrative and for the purpose of teaching those skilled in the art to implement the general mode of the present invention. It should be understood that the form of the present invention shown and described herein should be considered as a currently preferred embodiment. As will be fully understood by those skilled in the art after benefiting from this description of the present invention, elements and materials can replace those elements and materials illustrated and described herein, parts and processes can be reversed, and the specific features of the present invention can be utilized independently. Changes can be made to the elements described herein without departing from the spirit and scope of the present invention as described in the appended claims.

Claims

1. A system configured to determine information of a sample, comprising: an output acquisition subsystem configured to generate an output of the sample at one or more target locations on the sample; one or more temperature sensors configured to measure one or more temperatures within the system; a computer subsystem configured to obtain the one or more measured temperatures from the one or more temperature sensors; and one or more components executed by the computer subsystem; wherein the one or more components include a deep learning model configured to predict an error of at least one of the one or more target positions based on at least one of the one or more measured temperatures input to the deep learning model by the computer subsystem; and Wherein the computer subsystem is further configured to determine a corrected target position for the at least one of the one or more target positions by applying the prediction error to the at least one of the one or more target positions. 2 . The system of claim 1 , wherein the one or more temperature sensors are further configured to measure the one or more temperatures during a process performed on the sample utilizing the output acquisition subsystem.

3. The system of claim 1 , wherein the one or more temperature sensors are further configured to measure the one or more temperatures at or within one or more hardware elements of the system, the one or more hardware elements being configured to alter the positioning of the sample within the system.

4. A system according to claim 1, wherein at least one of the one or more measured temperatures input to the deep learning model by the computer subsystem includes all of the one or more temperatures measured by all of the one or more temperature sensors during a process performed on the sample using the output acquisition subsystem.

5. The system of claim 1, wherein the computer subsystem is further configured to determine the corrected target position before the output acquisition subsystem generates the output at the at least one of the one or more target positions.

6. The system of claim 1, wherein the computer subsystem is further configured to determine the corrected target position after the output acquisition subsystem generates the output at the at least one of the one or more target positions.

7. A system according to claim 1, wherein the deep learning model is further configured to predict the errors of a first target position and a second target position among the one or more target positions based on the at least one of the one or more measured temperatures input to the deep learning model, and wherein the computer subsystem is further configured to determine a first corrected target position and a second corrected target position of the first target position and the second target position among the one or more target positions, respectively, by applying the predicted errors of the first target position and the second target position among the one or more target positions, respectively, to the first target position and the second target position among the one or more target positions.

8. A system according to claim 1, wherein the computer subsystem is further configured to obtain position information of the one or more measured temperatures in response to measuring one or more locations of the one or more measured temperatures, and wherein the position information of at least one of the one or more measured temperatures is input by the computer subsystem to the deep learning model together with the at least one of the one or more measured temperatures.

9. A system according to claim 8, wherein the deep learning model is further configured to predict the errors of the first target position and the second target position among the one or more target positions based on the at least one of the one or more measured temperatures and the position information input into the deep learning model, and wherein the computer subsystem is further configured to determine a first corrected target position and a second corrected target position of the first target position and the second target position among the one or more target positions, respectively, by applying the predicted errors of the first target position and the second target position among the one or more target positions to the first target position and the second target position among the one or more target positions, respectively.

10. A system according to claim 1, wherein the computer subsystem is further configured to determine a corrected target position of a second target position among the one or more target positions by interpolating the prediction error determined for at least one of the one or more target positions to a second target position among the one or more target positions and applying the interpolated prediction error to the second target position among the one or more target positions.

11. The system of claim 1 , wherein the computer subsystem is further configured to train the deep learning model using a training set comprising training inputs and training outputs, wherein the training inputs comprise the one or more temperatures measured by the one or more temperature sensors during a training scan, and wherein the training outputs comprise errors measured in at least a portion of the one or more target locations during the training scan.

12. The system of claim 11, wherein the training input further comprises position information of the one or more temperatures measured during the training scan in response to one or more locations at which the one or more temperatures were measured during the training scan.

13. A system according to claim 11, wherein the training input further includes one or more additional temperatures measured by one or more additional temperature sensors included in the additional system during additional training scans of the sample or additional samples performed using the additional system, and wherein the training output further includes additional measured errors at one or more additional target locations on the sample or the additional sample, at which outputs are generated by an additional output acquisition subsystem of the additional system during the additional training scans.

14. A system according to claim 1, wherein the deep learning model is further configured to predict additional errors at one or more additional target locations on the sample or additional samples based on one or more additional temperatures measured by one or more additional temperature sensors of the additional system input to the deep learning model by the computer subsystem, at which one or more additional target locations, additional outputs are generated by an additional output acquisition subsystem of the additional system, and wherein the computer subsystem is further configured to monitor the drift of the system and the additional system relative to each other based on the prediction error and the additional prediction error.

15. A system according to claim 1, wherein the deep learning model is further configured to predict additional errors at one or more additional target locations on the sample or additional samples based on one or more additional temperatures measured by one or more additional temperature sensors of the additional system and input to the deep learning model by the computer subsystem, at which one or more additional target locations, additional outputs are generated by an additional output acquisition subsystem of the additional system, and wherein the computer subsystem is further configured to calibrate the system with respect to the additional system based on the prediction error and the additional prediction error.

16. The system of claim 1, wherein the deep learning model comprises a linear model.

17. The system of claim 1, wherein the deep learning model comprises a neural network.

18. The system of claim 1, wherein the deep learning model comprises a convolutional neural network.

19. The system of claim 1, wherein the output acquisition subsystem is further configured as a measurement subsystem.

20. The system of claim 1, wherein the output acquisition subsystem is further configured as a verification subsystem.

21. The system of claim 1, wherein the output acquisition subsystem is a light-based subsystem.

22. The system of claim 1, wherein the output acquisition subsystem is an electronic based subsystem.

23. A non-transitory computer readable medium storing program instructions executable on a computer system to perform a computer implemented method for determining information of a sample, wherein the computer implemented method comprises: acquiring one or more temperatures measured within a system using one or more temperature sensors, wherein the system includes an output acquisition subsystem configured to generate an output of the sample at one or more target locations on the sample; predicting an error of at least one of the one or more target positions by inputting at least one of the one or more measured temperatures into a deep learning model included in one or more components executed by the computer system; and A corrected target position for the at least one of the one or more target positions is determined by applying the prediction error to the at least one of the one or more target positions.

24. A method for determining information about a sample, comprising: measuring one or more temperatures within a system using one or more temperature sensors, wherein the system includes an output acquisition subsystem configured to generate an output for the sample at one or more target locations on the sample; predicting an error of at least one of the one or more target positions by inputting at least one of the one or more measured temperatures into a deep learning model included in one or more components executed by a computer system; and A corrected target position for the at least one of the one or more target positions is determined by applying the prediction error to the at least one of the one or more target positions, wherein the inputting and the determining are performed by the computer system.

Images