Processing tool with hyperspectral camera for metrology-based analysis

Abstract

Examples are disclosed relating to a processing tool including a hyperspectral camera configured to take a hyperspectral image of a processing chamber of the processing tool and / or a substrate in the processing tool. Metrology data derived from the hyperspectral image is used to control the operation of the processing tool.

Description

Background Art

[0001] Semiconductor device manufacturing involves many steps of material deposition, patterning, and removal to form devices on a substrate. Metrology-based analysis can be performed on the substrate throughout the production process for quality control inspection. Exemplary metrology analysis that can be performed on the substrate includes film thickness, non-uniformity, refractive index (RI), stress, particles, and Fourier transform infrared (FTIR) spectroscopy. Summary of the Invention

[0002] The present Summary of the Invention is provided to introduce a selection of concepts in a simplified form that will be further described in the following Detailed Description. The present Summary of the Invention is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. Additionally, the claimed subject matter is not limited to implementations that solve any or all of the disadvantages noted in any part of the present disclosure.

[0003] Examples related to a processing tool are disclosed, the processing tool including a hyperspectral camera configured to acquire hyperspectral images of a processing chamber of the processing tool and / or a substrate in the processing tool. Metrology data obtained from the hyperspectral images is used to control the operation of the processing tool.

[0004] In one example, a processing tool includes: a processing chamber that includes an optical interface; and a hyperspectral camera arranged to capture a hyperspectral image of the interior of the processing chamber through the optical interface of the processing chamber.

[0005] In some such examples, the processing chamber alternatively or additionally includes a susceptor, and the hyperspectral camera is arranged to capture a hyperspectral image of a substrate located on the susceptor through the optical interface.

[0006] In some such examples, the processing chamber alternatively or additionally includes a showerhead located on an opposite side of the susceptor, and the optical interface is provided on the showerhead.

[0007] In some such examples, the optical interface is alternatively or additionally provided on the susceptor.

[0008] In some such examples, the optical interface is alternatively or additionally provided on a sidewall of the processing chamber.

[0009] In some such examples, the processing tool also alternatively or additionally includes: one or more optical elements arranged between the optical interface and the hyperspectral camera, and the one or more optical elements are configured to direct electromagnetic radiation through the optical interface to the hyperspectral camera.

[0010] In some such examples, the processing chamber is alternatively or additionally a plasma reactor chamber.

[0011] In some such examples, the processing tool also alternatively or additionally includes a computing system configured to execute a trained machine learning model configured to receive one or more hyperspectral images from the hyperspectral camera and output metrology data for the processing chamber based at least on the one or more hyperspectral images.

[0012] In some such examples, the computing system is alternatively or additionally configured to adjust control parameters of a cleaning process to clean the processing chamber based at least on the metrology data for the processing chamber.

[0013] In some such examples, the trained machine learning model is alternatively or additionally configured to receive a series of hyperspectral images of a substrate in the processing chamber during a substrate processing cycle and output time-based metrology data for the substrate based at least on the series of hyperspectral images of the substrate. The computing system is configured to adjust one or more control parameters of the process of the substrate processing cycle based at least on the time-based metrology data for the substrate during the substrate processing cycle.

[0014] In some such examples, the trained machine learning model is alternatively or additionally configured to receive one or more hyperspectral images of a first substrate in the processing chamber during or after a first substrate processing cycle and output metrology data for the first substrate based at least on the one or more hyperspectral images of the first substrate. The computing system is configured to adjust one or more control parameters of the process of a second substrate processing cycle for a second substrate based at least on the metrology data for the first substrate.

[0015] In another example, a computer-implemented method for controlling a processing tool includes: receiving one or more hyperspectral images of a processing chamber of the processing tool from a hyperspectral camera; sending the one or more hyperspectral images to a trained machine learning model configured to output metrology data for the processing chamber based at least on the one or more hyperspectral images; and adjusting one or more control parameters of a process performed by the processing tool based at least on the metrology data for the processing chamber.

[0016] In some such examples, the process is alternatively or additionally a cleaning process for cleaning the processing chamber, and the one or more control parameters include control parameters of the cleaning process.

[0017] In some such examples, the one or more hyperspectral images alternatively or additionally include a series of hyperspectral images of a substrate in the processing chamber. The series of hyperspectral images of the substrate are received from the hyperspectral camera during a substrate processing cycle of the substrate. The trained machine learning model is configured to output time-based metrology data of the substrate, and the one or more control parameters are adjusted during the substrate processing cycle of the substrate based at least on the time-based metrology data of the substrate.

[0018] In some such examples, the one or more hyperspectral images alternatively or additionally include one or more hyperspectral images of a first substrate in the processing chamber during or after a first substrate processing cycle, and the one or more control parameters are adjusted for a second substrate processing cycle of a second substrate based at least on the metrology data of the first substrate.

[0019] In some such examples, the processing chamber alternatively or additionally is a plasma reactor chamber, and the one or more hyperspectral images of the plasma reactor chamber are captured by the hyperspectral camera when plasma is present in the plasma reactor chamber, and the plasma in the plasma reactor chamber is an illumination source for the hyperspectral camera.

[0020] In another example, a processing tool includes: a hyperspectral camera arranged to capture hyperspectral images of a substrate in the processing tool; and a computing system configured to execute a trained machine learning model, the trained machine learning model being configured to receive one or more hyperspectral images from the hyperspectral camera and output metrology data of the substrate based at least on the one or more hyperspectral images.

[0021] In some such examples, the metrology data includes the thickness of one or more layers of the substrate.

[0022] In some such examples, the metrology data alternatively or additionally includes the state of a gap in the substrate.

[0023] In some such examples, the hyperspectral camera alternatively or additionally has a dynamically adjustable position.

[0024] In some such examples, the hyperspectral camera alternatively or additionally has a dynamically adjustable angle.

[0025] In some such examples, the metrology data alternatively or additionally includes the magnitude of determined stress and / or warpage in the substrate.

[0026] In some such examples, the metrology data alternatively or additionally includes the magnitude of determined haze in the substrate. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A block diagram showing an exemplary processing tool is presented.

[0028] Figures 2 - 7 Different exemplary arrangements of a hyperspectral camera in a processing chamber are schematically shown.

[0029] Figure 8 An example hyperspectral image captured by the hyperspectral camera is schematically shown.

[0030] Figure 9 A flowchart is shown that depicts an exemplary method of training and executing a machine learning model to perform metrology-based analysis on a processing chamber and / or a substrate in the processing chamber.

[0031] Figure 10 A flowchart is shown that depicts an exemplary method that performs metrology-based analysis based on a hyperspectral image to control a cleaning process of a processing chamber.

[0032] Figure 11 A flowchart is shown that depicts an exemplary method that performs metrology-based analysis based on a hyperspectral image during a substrate processing cycle for in-situ control of a processing tool.

[0033] Figure 12 A flowchart is shown that depicts an exemplary method that performs metrology-based analysis based on a hyperspectral image for ex-situ online control of a processing tool between substrate processing cycles.

[0034] Figure 13 A block diagram of an exemplary computing system is shown.

[0035] Figure 14 Different exemplary states of a substrate during a process of filling gaps in the substrate are schematically shown.

[0036] Figure 15 Different exemplary states of a substrate resulting from a filling process of filling gaps in the substrate are schematically shown.

[0037] Figure 16 An exemplary scenario is schematically shown in which the hyperspectral camera is configured to be dynamically adjustable to adjust the distance of the hyperspectral camera relative to the substrate being imaged.

[0038] Figure 17 An exemplary scenario is schematically shown in which the hyperspectral camera is configured to be dynamically adjustable to adjust the angle of the hyperspectral camera relative to the substrate being imaged.

[0039] Figure 18An exemplary graph showing multiple curves of the spectral reflectance of a material on a substrate varying with wavelength and the angle of incident light on the material.

[0040] Figure 19 An exemplary graph showing two curves of the spectral reflectance of two different substrates varying with wavelength and the number of layers of the two different substrates.

[0041] Figures 20 - 22 Schematically shows an exemplary configuration where a hyperspectral camera can be used to measure the stress and / or warping of a substrate.

[0042] Figures 23 - 24 An exemplary graph showing spectral reflectance measured at different wavelengths at different points on a substrate to measure the stress and / or warping at different points on the substrate.

[0043] Figure 25 Shows an exemplary configuration where a hyperspectral camera is configured to measure the haze on a substrate.

[0044] Figures 26 - 27 Schematically shows an exemplary arrangement of spatially separated light emitters in the light source of a hyperspectral camera for measuring the haze of a substrate.

[0045] Figure 28 Shows an exemplary graph of the electromagnetic spectrum, which includes different wavelength ranges that can be used for different operations including measuring the haze of a substrate.

[0046] Figure 29 Shows an exemplary graph including haze measurement values represented by the number of reflected light pixels and their corresponding intensities.

[0047] Figure 30 Schematically shows an exemplary processing tool that includes multiple hyperspectral cameras.

[0048] Figure 31 Schematically shows an exemplary processing module that includes multiple hyperspectral cameras.

[0049] Figure 32 Shows a flowchart depicting an exemplary method that dynamically controls the position of a hyperspectral camera in a processing tool to change the distance between the hyperspectral camera and a substrate for hyperspectral imaging and analysis.

[0050] Figure 33 Shows a flowchart depicting an exemplary method that dynamically controls the position of a hyperspectral camera to capture hyperspectral images of different substrates from different angles.

[0051] Figure 34Shows a flowchart depicting an exemplary method that performs metrology-based analysis using hyperspectral imagery to control a processing tool. DETAILED DESCRIPTION

[0052] The term "atomic layer deposition" (ALD) generally refers to a process in which a film in the form of one or more individual layers is formed on a substrate by sequentially and conformally adsorbing precursors to the substrate and reacting the adsorbed precursors to form the film layer. Examples of ALD processes include plasma-enhanced ALD (PEALD) and thermal ALD (TALD). PEALD and TALD utilize a plasma of reactive gas and heat, respectively, to promote the chemical conversion of the precursors adsorbed to the substrate into a film on the substrate.

[0053] The term "chemical vapor deposition" (CVD) generally refers to a process in which a solid-phase film is formed on a substrate by directing the flow of one or more precursor gases over the surface of the substrate under conditions configured to cause the chemical conversion of the one or more precursor gases into the solid-phase film. The term "plasma-enhanced chemical vapor deposition" (PECVD) generally refers to a CVD process in which a plasma is used to facilitate the chemical conversion of one or more precursor gases into a solid-phase film on the substrate.

[0054] The term "cleaning process" generally refers to a process for cleaning deposited materials from the inner surface of a processing chamber. The deposited materials can include materials deposited on the substrate during a deposition process, by-products of the deposition process, residues from an etching process, and / or coatings of one or more materials coated onto the processing chamber prior to performing a deposition or etching process.

[0055] The term "control parameter" generally refers to a controllable variable in a process performed in a processing chamber. Exemplary control parameters include the temperature of a heater, the pressure inside the chamber, the flow rate of each of one or more process gases, and the frequency and power level of the radio frequency power used to form a plasma in the processing chamber.

[0056] The term "configured on" generally refers to a structural relationship in which one component is supported by another component. The term "configured on" by itself does not represent a specific relative positioning of the component with respect to the other component. For example, an optical interface configured on a portion of a processing chamber or a component of a processing chamber can be flush with the surface of the portion or component, can be inserted into the surface of the portion or component, or can extend beyond the surface of the portion or component.

[0057] The term "etch" and its variants generally refer to the removal of material from a structure. A substrate can be etched by a plasma in a plasma processing tool.

[0058] The term "hyperspectral camera" generally refers to an optical device configured to acquire hyperspectral imagery.

[0059] The term "hyperspectral image" generally refers to a data structure having multiple sub-images. Each different sub-image maps to a different wavelength or band of electromagnetic radiation. Each sub-image is a two-dimensional array of pixels. Each pixel of each sub-image stores an intensity value. The intensity value is the intensity of the electromagnetic radiation at the corresponding wavelength or band for the sub-image received from the corresponding spatial location in the processing chamber. In some examples, a hyperspectral image can include more than one hundred sub-images corresponding to different wavelengths or bands. In other examples, a hyperspectral image can take the form of a multispectral image that includes multiple sub-images, with each sub-image mapping to a selected band associated with a different descriptive channel name. Examples of bands and descriptive channel names include: blue in band 2 (0.45 - 0.51 micrometers (um)), green in band 3 (0.53 - 0.59 um), red in band 4 (0.64 - 0.67 um), near infrared (NIR) in band 5 (0.85 - 0.88 um), shortwave infrared (SWIR 1) in band 6 (1.57 - 1.65 um), shortwave infrared (SWIR 2) in band 7 (2.11 - 2.29 um), panchromatic in band 8 (0.50 - 0.68 um), cirrus in band 9 (1.36 - 1.38 um), thermal infrared (TIRS1) in band 10 (10.60 - 11.19 um), and thermal infrared (TIRS2) in band 11 (11.50 - 12.51 um). In some examples, a multispectral camera can image restricted bands of interest.

[0060] The term "illumination source" generally refers to a source that provides illumination light for a hyperspectral camera to capture an image.

[0061] The term "inhibitor" generally refers to a compound that can be introduced into a processing chamber, which can deposit non-conformally on a substrate surface and inhibit the ALD growth of an oxide film.

[0062] The term "metrology data" generally refers to data obtained by measuring one or more observable properties. For example, a hyperspectral camera can be used to obtain metrology data that includes the intensity of electromagnetic energy from different spatial locations in a processing chamber. Exemplary observable properties include film thickness, non-uniformity, refractive index (RI), stress, particle detection, and Fourier transform infrared (FTIR) spectroscopy. More than one of such observable properties can be used for calibration and validation of a hyperspectral metrology model.

[0063] The term "optical interface" generally refers to an optically transparent structure that is located between the interior and the exterior of a processing chamber to perform hyperspectral imaging of the processing chamber through the optical interface. The term "interior of the processing chamber" refers to the volumetric space where the substrate is located during processing. The optical interface can be located on the wall of the processing chamber or on a structure within the processing chamber, such as a pedestal or a showerhead. The optical interface transmits electromagnetic radiation for hyperspectral imaging to a hyperspectral camera while preventing gases from passing through.

[0064] The term "optical element" generally refers to a structure that is configured to direct and / or alter electromagnetic radiation along an optical path. Exemplary optical elements include optical fibers and other waveguides, diffractive and refractive lenses and mirrors, and polarizers and other filters.

[0065] The term "optically transparent" with respect to a material generally means that the material is suitably transparent to the electromagnetic energy bands imaged by a hyperspectral camera to obtain useful hyperspectral data.

[0066] The term "pedestal" generally refers to a structure that supports a substrate in a processing chamber.

[0067] The term "plasma" generally refers to an ionized gas that contains gaseous cations and free electrons.

[0068] The term "plasma reactor chamber" generally refers to a processing chamber in which a plasma can be generated to perform a chemical process on a substrate.

[0069] The term "processing chamber" generally refers to an enclosure in which chemical and / or physical processes are performed on a substrate. The pressure, temperature, gas flow rate, and ambient (atmospheric) composition within the processing chamber can be controlled to perform chemical and / or physical processes. The controllable aspects of the ambient composition include one or more of gas mixtures or plasma conditions.

[0070] The term "processing tool" generally refers to a machine that includes a processing chamber and other hardware configured to perform a substrate processing cycle.

[0071] The term "showerhead" generally refers to a structure for distributing gases over the surface of a substrate in a processing chamber.

[0072] The term "substrate" generally refers to any object that can be placed on a pedestal in a processing tool for processing.

[0073] The term "substrate processing cycle" generally refers to a set of one or more processes used to induce physical and / or chemical changes on a substrate. For example, a substrate processing cycle can include a deposition cycle, in which a thin film is formed on the substrate. As an example, the deposition cycle can be performed by a chemical vapor deposition (CVD) process or an atomic layer deposition (ALD) process. The substrate processing cycle can also include an etch cycle, in which material is removed from the substrate. As an example, the etch cycle can be performed by plasma etching.

[0074] The term "time-based metrology data" generally refers to data corresponding to measurements of different characteristics of an object measured over a period of time.

[0075] The term "trained machine learning model" generally refers to a computer program that has been trained on a dataset to find certain patterns or outputs based on certain inputs. For example, training can involve adjusting the weights between nodes in a neural network using an algorithm such as backpropagation.

[0076] The term "viewing port" generally refers to an optically semi-transparent or transparent window through which the interior of a processing chamber can be observed.

[0077] As described above, semiconductor device manufacturing includes many independent steps such as material deposition, patterning, and removal. During process development and during control checks in production, metrology data can be collected and analyzed between process steps to monitor the process. This metrology data is often obtained using offline techniques. Examples include scanning electron microscope (SEM) imaging of the substrate cross-section, ellipsometry, and Fourier transform infrared spectroscopy (FTIR).

[0078] In some cases, the process of obtaining metrology data for a substrate may take at least 2-3 hours per substrate. During this time, production may be stopped to ensure that the production process is operating within specifications. Such production interruptions reduce the overall output of the production process. In addition, some measurement processes are destructive, thus reducing the overall production yield. In addition, when an out-of-specification substrate is found, a significant amount of effort may be required to find the root cause of the deviation. In addition, multiple substrates may have been processed before the problem was discovered. This may require discarding the substrates.

[0079] Compared to offline ("non-in-situ") metrology, in-situ metrology can be used to efficiently measure a larger number of substrates, and may even be used for each substrate being processed. In-situ metrology refers to metrology performed on a substrate while the substrate is located in a processing tool.

[0080] In the case of metrology for a quality control process, one challenge is to make sufficient measurements to ensure that a processing tool operates within specification limits. For example, it may be difficult to use the data available in current in-situ metrology methods to monitor multiple substrate parameters. Performing metrology in a tool that utilizes a plasma can present particular challenges because the energy of the plasma may interfere with the measurement. Exemplary tools that utilize a plasma are plasma deposition tools and plasma etching tools. Exemplary plasma deposition tools are plasma enhanced atomic layer deposition (PEALD) tools and plasma enhanced CVD (PECVD) tools.

[0081] Accordingly, examples are disclosed that relate to performing in-situ metrology in a substrate processing tool using hyperspectral imaging of a processing chamber. Briefly, a processing tool may include: a processing chamber that includes an optical interface; and a hyperspectral camera that is arranged to capture a hyperspectral image of the processing chamber through the optical interface. The hyperspectral image includes image data of the processing chamber under light at multiple different wavelengths. Light at each wavelength may potentially provide different information than light at other wavelengths. This can provide more data than other in-situ measurement methods. Additionally, a hyperspectral image can be acquired in-situ during a substrate processing cycle. This allows a computing system to characterize a substrate in real-time during or immediately after processing while the substrate is located in the processing chamber. In some such examples, in-situ process control can be performed by adjusting one or more control parameters of one or more processes during the substrate processing cycle based at least on the acquired metrology data. This in-situ process control allows for the real-time characterization of the quality of a substrate without damaging the substrate and without stopping the substrate processing cycle. In this way, in-situ process control provides the technical advantages of improving substrate quality and substrate processing productivity while reducing costs.

[0082] In some examples, time-based metrology data can be generated from a series of hyperspectral images captured during a substrate processing cycle. The time-based metrology data includes hyperspectral imaging-based metrology measurements taken multiple times throughout the substrate processing cycle. The time-based metrology data can be used to establish a time-based property model, including film growth kinetics (e.g., nucleation delay, growth based on different process steps, etc.). Additionally, process control can be performed by adjusting one or more control parameters of one or more processes during the substrate processing cycle based at least on the time-based metrology data and / or the time-based model. This helps to improve substrate yield compared to not using time-based metrology.

[0083] In addition, in some examples, metrology-based analysis can also be performed "off-line" in-line between substrate processing cycles. Using the off-line in-line metrology function based on hyperspectral imaging, it can be determined whether the process is operating within the specification requirements. Then, changes can be made to the process control between runs. This can help avoid tool downtime compared to operations that use SEM or other off-line, destructive or non-destructive techniques to obtain metrology data.

[0084] In some examples, a computing system is configured to execute a trained machine learning model to analyze hyperspectral image data. The trained machine learning model is configured to receive one or more hyperspectral images from a hyperspectral camera and output metrology data for a processing chamber based at least on the one or more hyperspectral images. The computing system can also be configured to control the operation of a processing tool based at least on the metrology data.

[0085] The machine learning model can use temporal and spectral features to predict electrical and optical properties in a film of interest. In addition, image data from different spectral bands may be particularly relevant to different properties of the film under test. For example, infrared imaging can correlate temperature responses with metrics such as film thickness, non-uniformity, refractive index, resistivity, stress, and particle / defect concentration.

[0086] Figure 1 A schematic diagram of an exemplary processing tool 100 is shown. The processing tool 100 includes a processing chamber 102 and a pedestal 104 within the processing chamber. The pedestal 104 is configured to support a substrate 106 disposed within the processing chamber 102. The pedestal 104 can include a substrate heater 108. In other examples, the heater can be omitted, or can be located elsewhere within the processing chamber 102.

[0087] The processing tool 100 also includes a showerhead 110, a gas inlet 112, and flow control hardware 114. In other examples, opposite or in addition to the showerhead, the processing tool can include nozzles or other devices for supplying gas into the processing chamber 102. The flow control hardware 114 is connected to one or more process gas sources 116. In the case where the processing tool 100 includes a deposition tool, the process gas sources 116 can include one or more precursor sources and an inert gas source to be used, for example, as a diluent and / or purge gas. By way of example, in the case where the processing tool 100 includes an etch tool, a CVD tool, the process gas sources 116 can include one or more etchant gas sources and one or more inert gas sources.

[0088] The flow control hardware 114 can be controlled to flow gas from a process gas source through the gas inlet 112 into the process chamber 102. The flow control hardware 114 can include one or more flow controllers (such as mass flow controllers), valves, conduits, and other hardware to place a selected gas source or selected gas sources in fluid connection with the gas inlet 112. In other examples, the process chamber can include one or more additional gas inlets.

[0089] The processing tool 100 further includes an exhaust system 118. The exhaust system 118 is configured to receive gas flowing out of the process chamber 102. In some examples, the exhaust system 118 is configured to actively remove gas from the process chamber 102 and / or apply a local vacuum. The exhaust system 118 can include any suitable hardware, which includes one or more pumps.

[0090] The processing tool 100 further includes an RF power source 120 electrically connected to the pedestal 104. The RF power source 120 is configured to form a plasma. The plasma can be used to form reactive species, such as radicals, in a film deposition or etching process. In this example, the showerhead 110 is configured as a grounded counter electrode. In other examples, the RF power source 120 can supply RF power to the showerhead 110 or other suitable electrode structures. The processing tool 100 includes a matching network 122 for impedance matching of the RF power source 120. The RF power source 120 can be configured for any suitable frequency and power. Examples of suitable frequencies include frequencies in the range of 300 kHz to 90 MHz. More specific examples of suitable frequencies include 400 kHz, 13.56 MHz, 27 MHz, 60 MHz, 90 MHz, and 2.45 GHz. Examples of suitable power include power between 0 and 15 kilowatts. In some examples, the RF power source 120 is configured to operate at multiple different frequencies and / or powers. In other examples, the processing tool alternatively or additionally can include a remote plasma generator (not shown). The remote plasma generator can be used to generate a plasma away from the substrate being processed.

[0091] The processing chamber 102 also includes an optical interface 126 disposed on a sidewall 124 of the processing chamber. In other examples, the optical interface can be disposed on different surfaces, such as the ceiling or floor of the processing chamber. The optical interface 126 is an interface that can transfer electromagnetic radiation in a desired wavelength band from the interior of the processing chamber to a hyperspectral camera located outside the processing chamber while preventing gas from passing through. In the depicted example, the optical interface includes an optically transparent window positioned in an aperture formed in the sidewall of the processing chamber. In other examples, the optical interface can be configured as a window in the top wall or bottom wall of the processing chamber. The window in the wall of the processing chamber can be configured as an observation port. The term "observation port" generally refers to an optically transparent window in the wall of the processing chamber that is configured to allow an operator to view the interior of the processing chamber during a process. As described below, in further examples, the optical interface can include an optically transparent surface on a component located inside the processing chamber. Exemplary components include a pedestal and a showerhead.

[0092] The processing tool 100 also includes a hyperspectral camera 128 arranged to capture hyperspectral images of the interior of the processing chamber 102 via the optical interface 126 of the processing chamber 102. The processing tool 100 can include any suitable number of hyperspectral cameras to capture hyperspectral images of the processing chamber 102 and / or the substrate 106. In some implementations, the processing tool can include multiple processing chambers / processing stations, and the processing tool can include one or more hyperspectral cameras arranged to capture hyperspectral images of some or all of the multiple processing chambers / processing stations.

[0093] The optical interface for hyperspectral imaging of a processing chamber can be located at any suitable position in the processing chamber. Figures 2 - 7 Schematically shows different exemplary arrangements of a hyperspectral camera and an optical interface for imaging a processing chamber.

[0094] First, Figure 2An exemplary processing chamber 200 is shown that includes a susceptor 202 with a substrate 204 positioned thereon. The processing chamber includes an optical interface 208 disposed on a top plate 206 of the processing chamber 200. A hyperspectral camera 210 is arranged to capture hyperspectral images of the substrate 204 through the optical interface 208. The optical interface 208 can be formed of any material that is suitably transparent to the electromagnetic energy bands being imaged and suitably impermeable to the process gases. Exemplary electromagnetic energy bands include ultraviolet, visible, and infrared spectral bands. Exemplary materials for the optical interface 208 include fused quartz, fused silica, sapphire, windows with anti-reflection coatings, windows with coatings to prevent degradation, and windows with coatings that reduce the impact of materials deposited on the window. In this example, the optical interface 208 is configured as a viewing port through which the hyperspectral camera 210 can directly image the substrate 204. In the depicted example, the hyperspectral camera 210 is positioned to image the substrate 204 through the optical interface. In other examples, the hyperspectral camera 210 can be positioned to image any other suitable structure within the processing chamber 200 through the optical interface. The hyperspectral camera 210 can include any suitable lenses and / or other optical elements to allow imaging of a desired field of view (FOV).

[0095] Figure 3 Another exemplary processing chamber 300 is shown. The processing chamber 300 includes a susceptor 302 with a substrate 304 positioned thereon. The susceptor 302 is configured for backside processing. In such processing, one or more process gas outlets (not shown) in the susceptor 302 are used to expose the side of the substrate 304 facing the susceptor to the process gases. Here, an optical interface 308 is disposed on a surface 306 of the susceptor 302. The optical interface 308 includes an optically transparent structure through which a hyperspectral camera 310 can capture hyperspectral images of the processing chamber 300 and / or the substrate 304. For example, images of the processing chamber 300 can be taken during a processing chamber cleaning process when the substrate 304 is absent. Alternatively or additionally, images of the backside of the substrate 304 can be acquired during backside processing of the substrate. The hyperspectral camera 310 can include any suitable optical elements for imaging a desired portion of the substrate 304 and / or the processing chamber 300. Additionally or alternatively, any suitable optical elements can be positioned between the hyperspectral camera 310 and the processing chamber 300 for imaging a desired portion of the substrate 304 and / or the processing chamber 300. In some examples, a susceptor not configured for backside processing can also include an optical interface for hyperspectral imaging. The susceptor optical interface can be used, for example, to perform hyperspectral imaging metrology during a processing chamber cleaning process.

[0096] Figure 4Shows another exemplary processing chamber 400. The processing chamber 400 includes a susceptor 402, and a substrate 404 is positioned on the susceptor 402. The processing chamber 400 includes an optical interface 408 disposed on a sidewall 406. The optical interface 408 includes an optically transparent structure, and a hyperspectral camera 410 is positioned to capture hyperspectral images of the processing chamber 400 and / or the substrate 404 through the optically transparent structure. The hyperspectral camera 410 is disposed at a position outside the processing chamber 400 and away from the optical interface 408. In addition, an optical element 412 is disposed between the optical interface 408 and the hyperspectral camera 410. The optical element 412 is configured to direct electromagnetic radiation through the optical interface 408 to the hyperspectral camera 410. In this way, the hyperspectral camera 410 captures hyperspectral images of the processing chamber 200 and / or the substrate 404 through the optical element 412. The optical element 412 is depicted as an optical fiber or other waveguide. However, the optical element 412 represents any one or more optical elements capable of collectively directing electromagnetic radiation through the optical interface 408 to the hyperspectral camera 410. Exemplary optical elements include optical fibers, fiber bundles, another optical waveguide, one or more refractive / diffractive lenses, refractive / diffractive mirrors, waveguides, and / or filters, such as polarizers. In some examples, one or more optical elements may have an adjustable optical magnification. This can help focus electromagnetic radiation of different wavelengths onto the image sensor of the hyperspectral camera 410.

[0097] In some examples, the hyperspectral camera 410 can be calibrated to accommodate the grazing angle of the optical interface 408 relative to the substrate 404. For example, a distortion correction transform can be applied to the hyperspectral images captured by the hyperspectral camera 410 to accommodate the grazing angle based on the calibration position of the hyperspectral camera 410.

[0098] Figure 5 Shows another exemplary processing chamber 500. The processing chamber 500 includes a susceptor 502, and a substrate 504 is positioned on the susceptor 502. The processing chamber 500 further includes a showerhead 506 located on the opposite side of the susceptor 502. The showerhead 506 includes an optical interface 508 disposed on a surface 512 of the showerhead 506. The optical interface 508 includes an optically transparent structure, and a hyperspectral camera 510 can capture hyperspectral images of the processing chamber 500 and / or the substrate 504 through the optically transparent structure.

[0099] Figure 6Shows another exemplary processing chamber 600. The processing chamber 600 includes a pedestal 602, and a substrate 604 is positioned on the pedestal 602. The processing chamber 600 further includes a showerhead 606 on an opposite side of the pedestal 602. An optical interface 608 is disposed on a surface 610 of the showerhead 606. The optical interface 608 includes an optically transparent structure through which a hyperspectral camera 612 can capture hyperspectral images of the processing chamber 600 and / or the substrate 604. An optical element 616 is arranged between the optical interface 608 and the hyperspectral camera 612. The optical element 616 is configured to direct electromagnetic radiation from the optical interface 608 through the showerhead 606 to the hyperspectral camera 612. Thus, the hyperspectral camera 612 captures hyperspectral images of the processing chamber 600 and / or the substrate 604 through the optical element 616. In some examples, the optical element 616 includes an optical fiber or an optical fiber bundle. Other optical elements may also be used. Examples include one or more refractive or diffractive lenses and / or mirrors.

[0100] Figure 7 Shows another exemplary processing chamber 700. The processing chamber 700 includes a pedestal 702, and a substrate 704 is positioned on the pedestal 702. The processing chamber 700 further includes a showerhead 706 on an opposite side of the pedestal 702. A plurality of optical interfaces 708A, 708B, 708C are disposed on a surface 710 of the showerhead 706. Each of the optical interfaces 708A, 708B, 708C includes an optically transparent structure through which a hyperspectral camera 712 can capture hyperspectral images of the processing chamber 700 and / or the substrate 704. The hyperspectral camera 712 is arranged at a position remote from the optical interfaces 708A, 708B, 708C. Here, the hyperspectral camera 712 is located on a top plate 714 of the processing chamber 700. In other examples, the hyperspectral camera may be positioned at any other suitable location.

[0101] A plurality of optical elements 716A, 716B, 716C are arranged between the corresponding optical interfaces 708A, 708B, 708C and the hyperspectral camera 712. The optical elements 716A, 716B, 716C are configured to direct electromagnetic radiation through the plurality of optical interfaces 708A, 708B, 708C through the showerhead 706 to the hyperspectral camera 712. In the depicted example, each of the optical elements 716A, 716B, 716C includes an optical fiber, an optical fiber bundle, or other optical waveguide or waveguide system. Alternatively or additionally, other optical elements may be used, such as one or more refractive or diffractive lenses and / or mirrors. The plurality of optical interfaces 708A, 708B, 708C may be arranged on the surface 710 of the showerhead 706 in any suitable arrangement to jointly capture hyperspectral images of the processing chamber 700 and / or the substrate 704. Three optical interfaces 708A, 708B, 708C are shown in Figure 7In other examples, any other suitable number of optical interfaces and associated optical elements may be used.

[0102] In some examples, the hyperspectral camera 712 is configured to capture images from the optical elements 716A, 716B, 716C at multiple spatially separated regions on the image sensor of the hyperspectral camera 712. In other examples, the hyperspectral camera 712 is configured to stitch together the images collected from the multiple optical elements 716A, 716B, 716C to reconstruct a spatially continuous hyperspectral image of the processing chamber 700 and / or the substrate 704. In other examples, the images from the optical elements 716A, 716B, 716C are directed onto the image sensor of the hyperspectral camera in a partially or fully overlapping manner. In such examples, the overlapping images from the optical elements 716A, 716B, 716C may be analyzed using a trained machine learning function. Example machine learning functions are described in more detail below.

[0103] The above arrangements are provided as non-limiting examples. The hyperspectral camera may be arranged in any suitable manner to capture images of the processing chamber and / or the substrate in the processing chamber.

[0104] Return Figure 1 The hyperspectral camera 128 is configured to capture a hyperspectral image that includes a plurality of sub-images, each sub-image corresponding to a different wavelength or band. In some examples, the hyperspectral camera 128 is configured to capture a hyperspectral image that includes a plurality of sub-images corresponding to a plurality of different bands in the range from 250 to 1000 nanometers. In other examples, wavelengths outside of this range may alternatively or additionally be imaged. Further, in some examples, the hyperspectral camera 128 is configured to capture a hyperspectral image that includes 20 or more sub-images, each sub-image being at a different wavelength or band. In other examples, the hyperspectral camera may be configured to capture fewer than 20 sub-images.

[0105] In some examples, the hyperspectral camera 128 includes a wavelength selection filter that separates different bands for hyperspectral imaging. An example of such a filter is a diffraction grating. In some examples, the filter is tunable to select different bands. In other examples, a high-resolution filter is configured to selectively filter out multiple fixed bands.

[0106] In some examples, the hyperspectral camera 128 includes an illumination source 129. The illumination source 129 may include a broad-spectrum illumination source filtered by a high-resolution filter. In other examples, the illumination source 129 may be configured to emit light at specific wavelengths of interest. In some examples where the processing chamber 102 is a plasma reactor chamber, the plasma present in the plasma reactor chamber can be used as the illumination source for the hyperspectral camera 128. In further examples, the hyperspectral camera 128 can capture hyperspectral images without an illumination source. In such examples, the hyperspectral camera 128 and alternatives can rely on the heat present in the processing chamber 102 to provide heat-based hyperspectral data.

[0107] Figure 8 Schematically shows an example hyperspectral image 800 captured by a hyperspectral camera (such as Figure 1 the hyperspectral camera 128 shown). The hyperspectral image 800 includes a plurality of sub-images 802 corresponding to different bands (λ) of the electromagnetic spectrum. Each sub-image includes a plurality of pixels 804. Each pixel of the sub-image has a position defined by the X-axis 806 and the Y-axis 808, and an intensity value at a wavelength (λ) associated with the band of the sub-image. Each pixel of the hyperspectral image 800 includes a spatially mapped set of hyperspectral data that includes the intensity data of each sub-image. When multiple hyperspectral images are captured over a period of time, an additional dimension (e.g., transpose / time step) can be added. Such a time dimension allows tracking of the time-dependent response of the processing chamber and / or substrate to processing conditions. The hyperspectral data indicates the spectral characteristics of different elements or materials imaged by the hyperspectral image 800. The hyperspectral data of the hyperspectral image 800 is processed to generate metrology data. Exemplary metrology data may include measurements of the thickness, non-uniformity, stress, particle FTIR spectrum, absorption, reflection, and / or fluorescence spectrum data of the substrate (or one or more layers of the substrate) or the processing chamber at each pixel 804 of the hyperspectral image 800.

[0108] Return Figure 1 , the controller 130 is operatively coupled to the substrate heater 108, the flow control hardware 114, the exhaust system 118, the RF power source 120, and the hyperspectral camera 128. The controller 130 may include any suitable computing system, examples of which are described below with reference to Figure 13This will be described. The controller 130 is configured to control various functions of the processing tool 100 to process a substrate. As an example, the controller 130 is configured to operate the substrate heater 108 to heat the substrate 106 to a desired temperature. As another example, the controller 130 is also configured to operate the flow control hardware 114 to cause a selected gas or gas mixture to flow into the processing chamber 102 at a selected rate. As another example, the controller 130 is further configured to operate the exhaust system 118 to remove gas from the processing chamber 102. As another example, the controller 130 is further configured to operate the flow control hardware 114 and the exhaust system 118 to control the pressure within the processing chamber 102. As another example, the controller 130 is configured to operate the RF power source 120 to form a plasma.

[0109] The controller 130 is also configured to control the hyperspectral camera 128 to capture hyperspectral images of the processing chamber 102 and / or the substrate 106. In some examples, the hyperspectral camera 128 may capture hyperspectral images in a point-by-point, line scan, or snapshot mode. In the point-by-point mode, the hyperspectral camera 128 is configured to capture hyperspectral data of multiple bands one pixel at a time. In the line scan mode, the hyperspectral camera 128 is configured to capture hyperspectral data of multiple bands in a row (e.g., one row at a time). In the snapshot mode, the hyperspectral camera 128 is configured to capture an image sub-frame of each of the multiple bands one at a time. In some examples, the controller 130 controls the hyperspectral camera 128 to capture hyperspectral images during a substrate processing cycle while processing the substrate. In some examples, the controller 130 controls the hyperspectral camera 128 to capture a series of hyperspectral images throughout the substrate processing cycle to track the progress while the substrate is being processed.

[0110] In some examples, the controller 130 may control the illumination source 129 of the hyperspectral camera 128 to output light to illuminate the substrate 106 or the processing chamber 102 during image capture. In other examples, the controller 130 may control the hyperspectral camera 128 to acquire an image while controlling the RF power source 120 to form a plasma. In such examples, the plasma may provide appropriate broad-spectrum light for hyperspectral imaging. Additionally, in some examples, once the substrate processing cycle is complete, the controller 130 controls the hyperspectral camera 128 to capture a hyperspectral image. The controller 130 may control the hyperspectral camera 128 to capture any suitable number of hyperspectral images at any suitable frame rate during and / or after the substrate processing cycle.

[0111] In some examples, the controller 130 is configured to execute a trained machine learning model 132. The trained machine learning model 132 is configured to receive one or more hyperspectral images from the hyperspectral camera 128 and output metrology data 134 for the processing chamber 102 and / or the substrate 106 based at least on one or more of the hyperspectral images. The metrology data 134 may characterize various properties of the processing chamber 102 and / or the substrate 106. In some examples, the metrology data 134 includes absorption, reflection, and / or fluorescence spectral data of the substrate 106 (and / or other materials in the processing chamber 102). Alternatively or additionally, in some examples, the metrology data 134 includes measurements of stress applied to the substrate 106. Alternatively or additionally, in some examples, the metrology data 134 includes measurements of the resistivity of the substrate 106. Alternatively or additionally, in some examples, the metrology data 134 includes measurements of the thickness of the substrate 106 and / or the thicknesses of the individual layers deposited on the substrate 106. Alternatively or additionally, in some examples, the metrology data 134 includes an assessment of the non-uniformity of the substrate 106. Alternatively or additionally, in some examples, the metrology data 134 includes an indication of particle detection in the processing chamber 102 and / or measurements of the sizes of the particles detected in the processing chamber 102. The metrology data 134 generated based at least on the hyperspectral images may, in some examples, measure properties of the substrate 106 and / or the processing chamber 102 at a relatively higher resolution compared to other non-in-situ metrology analysis methods that are not based on hyperspectral images.

[0112] In some implementations, the trained machine learning model 132 is configured to receive a series of hyperspectral images from the hyperspectral camera 128 over a period of time and output time-based metrology data 134 for the processing chamber 102 and / or the substrate 106 based at least on the series of hyperspectral images. In some examples, the series of hyperspectral images is captured during a substrate processing cycle for in-situ analysis and control of the processing tool 100. In some such examples, the series of hyperspectral images is captured during a period of time that begins before the start of the substrate processing cycle and ends after the completion of the substrate processing cycle. In other examples, the series of hyperspectral images is captured during a period of time that spans only a portion of the substrate processing cycle. In further examples, the series of hyperspectral images is captured during a longer period of time that encompasses multiple substrate processing cycles.

[0113] The trained machine learning model 132 can be a time-based model that is trained to analyze changes in metrology data to determine how the processing chamber 102 and / or the substrate 106 change over time. The time-based metrology data 134 can track the change of any suitable type of measurement result over time. As an example, the time-based metrology data 134 can track the growth of a film deposited on the substrate 106 over time. As another example, the time-based metrology data 134 can measure the nucleation delay at the start of a process. As a further example, the time-based metrology data 134 can measure the efficacy of an inhibition process on controlling conformality. As another example, the time-based metrology data 134 can monitor the progress of an etching process. As a further example, the time-based metrology data can monitor the particulate contamination of the substrate during a process. As another example, the time-based metrology data 134 can monitor the accumulation of material on the surface of the processing chamber 102. As another example, the time-based metrology data 134 can monitor the non-uniformity of a film deposited on the substrate 106 over time. Conventionally, non-uniformity metrics (such as thickness) are done offline at several points (e.g., between 10 and 50 points) using ellipsometry or XRF or other methods. These points are used as positions for mapping thickness, refractive index, sheet resistance, or other properties to determine the non-uniformity across a 300 mm wafer. However, more detailed mapping can be achieved using hyperspectral imaging. For example, depending on the resolution of the hyperspectral camera, measurement results with a resolution of less than 1 mm can be obtained on a 300 mm wafer using a time-based model. In this way, not only can the evolution of the thickness (or other properties) of each point be obtained, but also a higher-resolution non-uniformity evolution can be obtained than in-situ measurements in the final state.

[0114] The trained machine learning model 132 can employ any suitable method for processing the time-based metrology data 134. For example, the trained machine learning model 132 can use one or more convolutional neural networks (e.g., spatial and / or temporal convolutional neural networks for processing images and / or videos), recurrent neural networks (e.g., long short-term memory networks), support vector machines, associative memories (e.g., lookup tables, hash tables, Bloom filters, neural Turing machines, and / or neural random access memories), unsupervised spatial and / or clustering methods (e.g., nearest neighbor algorithms, topological data analysis, and / or k-means clustering), linear and / or Gaussian regression modeling, graphical models (e.g., Markov models, conditional random fields, and / or AI knowledge bases), and / or other dimensionality reduction and modeling methods.

[0115] Figure 9shows a flow chart depicting an exemplary method 900 of training and executing a machine learning model to perform metrology-based analysis on a processing chamber and / or a substrate within the processing chamber. For example, the method may be executed to train and execute Figure 1 the trained machine learning model 132 shown in. In some examples, Figure 1 the controller 130 shown in may execute the method. In other examples, an independent computing system may train the trained machine learning model 132, and the controller 130 may execute the trained machine learning model 132.

[0116] At 902, method 900 includes receiving raw data for training a machine learning model. In some examples, the raw data includes hyperspectral images of the processing chamber under different conditions / states. For example, when training a machine learning model to monitor a processing chamber cleaning process, such conditions / states may include a clean processing chamber and a processing chamber with different degrees of residue accumulation after undergoing various numbers of processing cycles. In other examples, the raw data includes hyperspectral images of substrates under different conditions / states to train a machine learning function to monitor substrate processing. For example, such conditions / states may include an unprocessed substrate, substrates at different points within a process, and substrates after undergoing different processes. The raw data may include any suitable type of training data to train a machine learning model to output metrology data based on one or more hyperspectral images. In some examples, the raw data includes metadata associated with hyperspectral camera characteristics. Exemplary hyperspectral camera characteristics include intrinsic and extrinsic characteristics of the hyperspectral camera. Exemplary intrinsic camera characteristics may include focal length, principal point, pixel size, and pixel resolution. Exemplary extrinsic camera characteristics may include the position and orientation of the camera in world space. In some examples, the raw data includes metadata associated with processing chamber operations. Exemplary processing chamber operation characteristics include information such as: one or more process gases in the processing chamber, the flow rate of each of the one or more process gases, the total chamber pressure, the plasma power level, the plasma frequency, and the substrate temperature. By considering this metadata, the machine learning model can be updated / retrained based on changes in the hardware and / or process to be more robust and accurate under different operating conditions compared to other machine learning models that are not updated / retrained.

[0117] At 904, method 900 includes preprocessing raw data by filtering out data that is not needed for machine learning model training. In some examples, the filtered data includes duplicate hyperspectral images. In some examples, the filtered data includes hyperspectral data in bands of no interest. For example, if a machine learning model is being trained to process a specific film that responds only to certain bands, hyperspectral data corresponding to other bands that do not respond to the film can be filtered out without processing. In other examples, the raw data is preprocessed using normalization and dimensionality reduction techniques (such as principal component analysis). The preprocessing step can be optionally performed to reduce the total time for training the machine learning model.

[0118] At 906, method 900 includes training / developing a machine learning model. The machine learning model can be trained / developed according to any suitable training procedure. Non-limiting examples of training procedures for machine learning models include supervised training (e.g., using gradient descent or any other suitable optimization method), zero-shot methods, few-shot methods, and unsupervised learning methods (e.g., classification based on categories derived from unsupervised clustering methods), reinforcement learning (e.g., deep Q-learning based on feedback). In some examples, training can be performed by backpropagation using an appropriate loss function. Exemplary loss functions that can be used for training include mean absolute error, mean squared error, cross entropy, Huber loss, or other loss functions.

[0119] In some examples, the machine learning model can be trained by supervised training on labeled training data that includes an image set having the same structure as the input image. In other words, the training data includes the same type of hyperspectral images as those captured by the hyperspectral camera provided as the input to the trained machine learning model. For example, raw data or preprocessed data of a substrate and / or a processing chamber under different processing conditions.

[0120] At 908, method 900 includes executing the machine learning model to perform a metrology-based analysis on a processing chamber of a processing tool and / or a substrate in the processing chamber. In particular, the machine learning model receives one or more hyperspectral images of the processing chamber and / or the substrate as input and outputs metrology data based on the one or more hyperspectral images.

[0121] In some examples, metrology data representing one or more observable characteristics of a processing chamber and / or a substrate can be used for calibration and validation of a hyperspectral metrology machine learning model. As an example, the film thickness on a substrate can be observed to determine whether a deposition process is operating within specifications as recommended by a trained machine learning model. If the film thickness is within specifications, the controller can verify whether the trained machine learning model is operating properly. Otherwise, if the film thickness is out of specifications, the trained machine learning model can be adjusted / recalibrated to adjust the control of the deposition process such that the film thickness is within specifications. The metrology data can be used to verify and / or calibrate the trained machine learning model in any suitable manner.

[0122] Return Figure 1 , the controller 130 is configured to adjust the control of the processing tool 100 based at least on metrology data 134 output by a trained machine learning model 132. In some examples, the trained machine learning model 132 is configured to output a recommended control adjustment based on the metrology data 134. In other examples, an independently trained machine learning model can be configured to recommend certain control adjustments based at least on the metrology data 134. In still other examples, the controller 130 can include independent logic that is configured to adjust the control of the processing tool 100 based at least on the metrology data 134. In still other examples, the controller 130 is configured to visually present the metrology data 134 to an operator via a display, and the controller 130 is configured to adjust the operation of the processing tool 100 based at least on user input received from the operator.

[0123] In some examples, the controller 130 is configured to adjust the operation of the processing tool 100 based on metrology data 134 of the processing chamber 102 itself. The controller 130 can be configured to adjust any suitable control parameter of any suitable process performed by the processing tool 100 based on the metrology data 134 for processing the substrate.

[0124] As described above, in some examples, the controller 130 can be configured to adjust control parameters of a cleaning process to clean the processing chamber 102 based at least on metrology data 134 for the processing chamber 102. In one example, the metrology data 134 for the processing chamber 102 can indicate the amount of material accumulated inside the processing chamber 102, and the controller 130 can be configured to determine whether the amount of material accumulated inside the processing chamber 102 is greater than a threshold amount. If the amount of material is greater than the threshold amount, the controller 130 initiates the cleaning process. Additionally or alternatively, the controller 130 can monitor the progress of the cleaning process for endpoint detection based on the amount of material accumulated inside the processing chamber 102. By intelligently controlling the cleaning process based on the metrology data 134 for the processing chamber 102, the cleaning of the processing chamber can be performed more efficiently and as needed. This can provide reduced tool maintenance time as compared to cleaning processes performed at a fixed frequency or for a fixed length / extent.

[0125] Alternatively or additionally, in some examples, the controller 130 can be configured to perform in-situ analysis on metrology data collected during a substrate processing cycle and adjust control of the processing tool in real-time during the substrate processing cycle.

[0126] In some such examples, the controller 130 can be configured to monitor particle contamination on the substrate surface or in the processing chamber during the process based at least on the metrology data 134. Such in-situ analysis allows for intelligent scheduling of other inspection operations. This can help limit the number of substrates scanned on an optical scatter tool for particle detection. This can also help limit the substrate areas on which analyses such as energy-dispersive X-ray (EDX) analysis are performed to determine particle composition for troubleshooting. As another example, the controller 130 can be configured to perform in-situ analysis on time-based metrology data for the substrate collected during a substrate processing cycle. The controller can also be configured to adjust control of the processing tool in real-time during the substrate processing cycle. The controller 130 can be configured to adjust any suitable control parameter of any suitable substrate process in real-time based on the in-situ analysis of the time-based metrology data.

[0127] Alternatively or additionally, in some such examples, the controller 130 can be configured to track film thickness during a film deposition process based at least on time-based metrology data. The controller 130 can also be configured to adjust the deposition process to control the deposition rate. As a more specific example, the controller 130 can be configured to allow a high deposition growth rate until a first threshold thickness is detected. The controller can also be configured to subsequently adjust the processing conditions to reduce the deposition rate until a desired final thickness is reached.

[0128] Alternatively or additionally, in some such examples, in a process utilizing an inhibitor, the controller 130 may be configured to track the efficacy of the inhibition process during the inhibition process at least based on time-based metrology data, and dynamically adjust the inhibition time and / or the number of inhibition cycles based on the efficacy derived from the time-based metrology data. This can help ensure that film growth is inhibited appropriately according to the desired process. As an example, an inhibited ALD process may be performed by first depositing an inhibitor on the feature such that a higher concentration of the inhibitor is deposited on the substrate surface and a lower concentration of the inhibitor is deposited within the substrate recesses. Subsequently, ALD may be used to deposit the film such that the final film is thicker within the substrate recesses and thinner or fully inhibited on the substrate surface. In such examples, hyperspectral imaging may be performed to monitor inhibitor adsorption on the substrate surface. This may allow the inhibitor deposition to continue until a desired degree of inhibitor adsorption is achieved. A hyperspectral camera may also be used to monitor film growth on the substrate surface. This may be used to determine whether the inhibitor is effectively inhibiting film growth or whether additional inhibitor deposition cycles are needed.

[0129] Alternatively or additionally, in some examples, the controller 130 may be configured to perform in-situ online analysis of metrology data for a substrate and adjust the control of the processing tool 100 between multiple substrate processing cycles. The controller 130 may be configured to adjust any suitable control parameter of any suitable substrate process on a run-to-run basis based on the in-situ online analysis of the metrology data.

[0130] Non-in-situ online measurements can be performed in a variety of different ways. In some examples, a processing tool can include a separate module for hyperspectral imaging of a substrate. The term "separate module" as used herein generally refers to a space within the processing tool that is separated from one or more processing chambers of the processing tool, and the substrate can be moved into this space by a substrate handling system for hyperspectral imaging. In other examples, non-in-situ online measurements can be performed when transferring the substrate into or out of a processing chamber of the processing tool. For example, a hyperspectral camera 128 can be positioned outside a slit valve through which the substrate moves when being transported into and out of a processing station, and the substrate can be imaged by the hyperspectral camera 128 as it passes through or exits the slit valve. In other examples, the substrate can be imaged for metrology analysis based on non-in-situ online hyperspectral imagery when the substrate is located in a transfer module, a load lock module, a front-opening unified pod (FOUP), or an equipment front-end module (EFEM). In still other examples, a hyperspectral camera with an illumination source (e.g., quartz tungsten, xenon, LED group 400nm - 1000nm) can be placed above a vacuum transfer arm that moves across the substrate, and the hyperspectral camera can be used as a line scan camera to image the substrate as the vacuum transfer arm moves relative to the substrate.

[0131] In some such examples, the controller 130 can be configured to compare a value of a parameter of interest of metrology data 134 from the substrate with an expected / ideal parameter value. For example, the thickness of a deposited film after a deposition process cycle can be measured by hyperspectral imagery and compared with a target thickness. If the measured thickness deviates from the target thickness by more than a threshold amount, the controller 130 can be configured to adjust control parameters (e.g., power, pressure, gas flow parameters) of subsequent substrate processing cycles for a different substrate such that the accuracy of the subsequent substrate processing cycles for that different substrate is increased relative to the previous substrate processing cycles.

[0132] Alternatively or additionally, in some such examples, if the controller 130 determines based at least on an analysis of the metrology data 134 that the thickness of the film deposited on the substrate is outside a threshold limit of uniformity, the controller 130 can be configured to perform an automatic correction in process control (e.g., a change in process gap, a change in spindex / rotation operation). In yet another example, the controller 130 is configured to trigger an alert for a human engineer to perform a manual correction based on determining that the thickness of the film deposited on the substrate is highly non-uniform. For example, the controller 130 can trigger the execution of a showerhead - pedestal leveling program.

[0133] In some examples, a trained machine learning model 132 is configured to generate suggestions for control adjustments for an operator based on the metrology data 134.

[0134] Figure 10 shows a flow chart depicting an exemplary method 1000 that performs metrology-based analysis using hyperspectral imagery to control a cleaning process of a processing chamber. For example, method 1000 may be performed by Figure 1 controller 130 as shown.

[0135] At 1002, method 1000 includes receiving one or more hyperspectral images of a processing chamber of a processing tool from a hyperspectral camera. At 1004, method 1000 includes transmitting the one or more hyperspectral images to a trained machine learning model configured to output metrology data of the processing chamber based at least on the one or more hyperspectral images. At 1006, method 1000 includes adjusting one or more control parameters of a process performed by the processing tool based at least on the metrology data of the processing chamber. In some implementations, at 1008, method 1000 optionally may include adjusting one or more control parameters of a cleaning process during cleaning of the processing chamber. In some examples, the frequency and / or length / extent of the cleaning process is adjusted based on the amount of material buildup on the processing chamber as indicated by the metrology data of the processing chamber. Additionally or alternatively, in some examples, the cleaning pressure, cleaning gas flow rate, and / or cleaning gas timing may be adjusted based on an analysis of the metrology data.

[0136] Alternatively or additionally, in some implementations, at 1010, method 1000 optionally may include adjusting one or more control parameters of a detection process to detect the processing chamber. In one example, the frequency of the detection process is adjusted based on detecting particles in the processing chamber as indicated by the metrology data of the processing chamber.

[0137] Method 1000 may be performed to control the operation of the processing tool in an intelligent manner based on feedback provided by the metrology data of the processing chamber. Such intelligent operation may include performing cleaning and / or detection operations only as determined necessary based on the feedback. The intelligent operation may increase the efficiency and productivity of the processing tool relative to a processing tool that performs such operations without feedback. Method 1000 may be repeatedly performed for any suitable number of processes and / or processing cycles.

[0138] Figure 11 shows a flow chart depicting an exemplary method 1100 that performs metrology-based analysis using hyperspectral imagery for in-situ control of a processing tool during a substrate processing cycle. By way of example, method 1100 may be performed by Figure 1 controller 130 as shown.

[0139] At 1102, method 1100 includes receiving, from a hyperspectral camera, a series of hyperspectral images of a substrate in a processing chamber of a processing tool during a substrate processing cycle. At 1104, method 1100 includes, during the substrate processing cycle, sending the series of hyperspectral images to a trained machine learning model that is configured to output time-based metrology data of the substrate based at least on the series of hyperspectral images. At 1106, method 1100 includes, during the substrate processing cycle, adjusting one or more control parameters of a process of the substrate processing cycle based at least on the time-based metrology data of the substrate. Examples of adjustable control parameters include one or more of the following: process time, substrate temperature, showerhead temperature (where the showerhead has a heater), spacing between the showerhead and the pedestal, total process pressure, partial pressure of each of one or more process gases, and radio frequency power. Method 1100 can thus provide in-situ metrology-based analysis using hyperspectral images, which allows for real-time adjustment and control of the processing tool. In some implementations, the in-situ metrology-based analysis / metrology data can optionally be tracked across different processing cycles for multiple substrates to adjust the control of a particular process. For example, for a given process step (in a given iteration), a particular statistic / characteristic of each of multiple substrates is tracked to determine if there is a drift / offset that can be corrected by an adjustment to the process. Method 1100 can be repeatedly executed for any suitable number of processes and / or processing cycles.

[0140] Figure 12 A flowchart is shown that depicts an exemplary method 1200 for performing metrology-based analysis using hyperspectral images for ex-situ online control of a processing tool between substrate processing cycles. For example, method 1200 can be performed by Figure 1The controller 130 shown executes. At 1202, method 1200 includes receiving, during or after a first substrate processing cycle, one or more hyperspectral images of a first substrate of a processing tool from a hyperspectral camera. Note that the substrate can be imaged by the hyperspectral camera in any suitable processing module of the processing tool or when being transferred between multiple different processing modules of the processing tool for in-situ online metrology-based analysis. At 1204, method 1200 includes transmitting the one or more hyperspectral images to a trained machine learning model configured to output metrology data of the first substrate based at least on the one or more hyperspectral images. At 1206, method 1200 includes, for a second substrate processing cycle of a second substrate, adjusting one or more control parameters of the process of the second substrate processing cycle based at least on the metrology data of the first substrate. Method 1200 can be executed to provide in-situ online metrology-based analysis using hyperspectral images, enabling batch-by-batch adjustment and control of the processing tool between multiple substrate processing cycles. Additionally, method 1200 can be executed to provide in-situ online metrology-based analysis occurring during multiple processing cycles of multiple different substrates, e.g., to correct for drift / offset in operation over a longer period of time.

[0141] In some examples, the methods and processes described herein can be associated with a computing system of one or more computing devices. In particular, such methods and processes can be implemented as a computer application or service, an application programming interface (API), a library, and / or other computer program products.

[0142] Figure 13 A non-limiting implementation of a computing system 1300 is schematically shown, which can implement one or more of the above methods and processes. Computing system 1300 is shown in a simplified form. Computing system 1300 can take the form of one or more personal computers, workstations, computers integrated with a wafer processing tool, and / or network-accessible server computers.

[0143] Computing system 1300 includes a logic machine 1302 and a storage machine 1304. Computing system 1300 can optionally include a display subsystem 1306, an input subsystem 1308, a communication subsystem 1310, and / or Figure 13 other components not shown. Controller 130 is an example of computing system 1300.

[0144] The logic machine 1302 includes one or more physical devices configured to execute instructions. For example, the logic machine may be configured to execute instructions that are part of one or more applications, services, programs, routines, libraries, objects, components, data structures, or other logical constructs. Such instructions may be implemented to perform tasks, implement data types, transform the state of one or more components, achieve a technical effect, or otherwise achieve a desired result.

[0145] The logic machine may include one or more processors configured to execute software instructions. Additionally or alternatively, the logic machine may include one or more hardware or firmware logic machines configured to execute hardware or firmware instructions. The processors of the logic machine may be single-core or multi-core, and the instructions executed thereon may be configured for sequential, parallel, and / or distributed processing. Optionally, the various components of the logic machine may be distributed between two or more separate devices, which may be remotely located and / or configured for coordinated processing. Aspects of the logic machine may be virtualized and executed by remotely accessible networked computing devices configured in a cloud computing configuration.

[0146] The storage machine 1304 includes one or more physical devices configured to hold instructions 1312 executable by the logic machine to implement the methods and processes described herein. When implementing such methods and processes, the state of the storage machine 1304 may be transformed—for example, to hold different data.

[0147] The storage machine 1304 may include removable and / or built-in devices. The storage machine 1304 may include optical memory (e.g., CD, DVD, HD-DVD, Blu-ray Disc, etc.), semiconductor memory (e.g., RAM, EPROM, EEPROM, etc.), and / or magnetic memory (e.g., hard disk drive, floppy disk drive, tape drive, MRAM, etc.), among others. The storage machine 1304 may include volatile, non-volatile, dynamic, static, read / write, read-only, random access, sequential access, location-addressable, file-addressable, and / or content-addressable devices.

[0148] It should be understood that the storage machine 1304 includes one or more physical devices. However, aspects of the instructions described herein may alternatively be propagated via a communication medium (e.g., electromagnetic signals, optical signals, etc.) that is not held by a physical device for a finite duration.

[0149] Aspects of the logic machine 1302 and the storage machine 1304 can be integrated together into one or more hardware logic components. Such hardware logic components can include, for example, field programmable gate arrays (FPGAs), program and application specific integrated circuits (PASIC / ASICs), program and application specific standard products (PSSP / ASSPs), systems on a chip (SOCs), and complex programmable logic devices (CPLDs).

[0150] When included, the display subsystem 1306 can be used to present a visual representation of data stored by the storage machine 1304. The visual representation can take the form of a graphical user interface (GUI). Since the methods and processes described herein change the data stored by the storage machine and thus transform the state of the storage machine, the state of the display subsystem 1306 can likewise be transformed to visually represent changes in the underlying data. The display subsystem 1306 can include one or more display devices using almost any type of technology. Such display devices can be combined with the logic machine 1302 and / or the storage machine 1304 in a shared enclosure, or such display devices can be peripheral display devices.

[0151] When included, the input subsystem 1308 can include or interface with one or more user input devices such as a keyboard, mouse, or touch screen. In some embodiments, the input subsystem can include or interface with selected natural user input (NUI) components. Such components can be integrated or peripheral, and the translation and / or processing of input actions can be handled on-board or off-board. Exemplary NUI components can include a microphone for language and / or speech recognition; and infrared, color, stereo, and / or depth cameras for machine vision and / or gesture recognition.

[0152] When included, the communication subsystem 1310 can be configured to communicatively couple the computing system 1300 with one or more other computing devices. The communication subsystem 1310 can include wired and / or wireless communication devices compatible with one or more different communication protocols. As a non-limiting example, the communication subsystem can be configured to communicate via a wireless telephone network or a wired or wireless local or wide area network. In some embodiments, the communication subsystem can allow the computing system 1300 to send and / or receive messages to and from other devices via a network such as the Internet.

[0153] As described above, with reference to Figure 1, the controller 130 is configured to execute a trained machine learning model 132. The trained machine learning model 132 is configured to receive one or more hyperspectral images from the hyperspectral camera 128 and output metrology data 134 of the processing chamber 102 and / or the substrate 106 based at least on the one or more hyperspectral images. The metrology data 134 may characterize various properties of the processing chamber 102 and / or the substrate 106.

[0154] In some implementations, the machine learning model 132 is trained to predict / identify various properties of the substrate 106 and / or other materials (e.g., gases) in the processing chamber 102 based at least on spectral features generated in the hyperspectral images output from the hyperspectral camera 128. Different spectral features are generated from different gases and substrates because different gases and substrates transmit and reflect light of different wavelengths and intensities. Additionally, changes in the spectral features captured in the hyperspectral images may be caused by changes in: the density of the gas / plasma, the composition of the gas / plasma (e.g., different types of gases transmit and reflect different wavelengths and intensities to create different spectral features captured in the hyperspectral images), the flow path of the gas / plasma during the process, the composition of the substrate 106, the thickness of the substrate 106, and / or the density of the substrate 106. To identify these properties, the hyperspectral camera 128 captures multiple hyperspectral images "in-situ" while the substrate is in the processing tool, and the machine learning model 132 is trained to predict / identify these properties based at least on analyzing the multiple hyperspectral images. The machine learning model 132 may be trained to predict / identify any or all of these properties based at least on analyzing the multiple hyperspectral images of the processing chamber 102 and / or the substrate 106. The machine learning model 132 may be trained to identify any suitable properties of the processing chamber 102, the substrate 106, and / or other materials in the processing chamber 102 based at least on analyzing the spectral features corresponding to these different elements captured in the multiple hyperspectral images.

[0155] In some implementations, the controller 130 is configured to execute multiple machine learning models, each of which is trained to predict / identify different properties of the processing chamber 102, the substrate 106, and / or other materials in the processing chamber 102 based at least on analyzing multiple hyperspectral images. The controller 130 may execute the multiple machine learning models simultaneously to analyze multiple hyperspectral images to predict / identify different properties of the processing chamber 102, the substrate 106, and / or other materials in the processing chamber 102.

[0156] In some implementations, the hyperspectral camera 128 is configured to in-situ capture a series of hyperspectral images of the substrate 106 during a process performed on the substrate 106 to determine whether the process is being correctly performed according to specifications or within specified tolerances. In one example, the machine learning model 132 is trained to analyze the differences in the reflection spectra of the entire substrate 106 during the process and to determine whether the substrate 106 is within specifications or within specified tolerances.

[0157] Figure 14 Schematically shows different exemplary states of a substrate 1400 during a process of filling a gap 1402 in a feature of the substrate 1400 by atomic layer deposition (ALD). ALD can provide conformal growth of a film such that the film has a substantially uniform thickness in all orientations. At 1404, a first state of the gap 1402 is shown, where the gap 1402 is empty. For example, the gap 1402 may be in the first state at the start of the process. At 1406, a second state of the gap 1402 is shown, where the gap 1402 is partially filled. At 1408, a third state of the gap 1402 is shown, where the gap 1402 is partially filled to a greater extent than the second state. At 1410, a fourth state of the gap 1402 is shown, where the gap 1402 is completely filled. For example, the gap 1402 may be in the fourth state at the end of the process.

[0158] In some implementations, the hyperspectral camera 128 is configured to capture hyperspectral images of the substrate 1400 in each of the different states 1404 - 1410 during the process of filling the gap 1402. The reflection spectra of the regions in the hyperspectral images corresponding to the substrate 1400 (and more particularly the gap 1402) are different in each of the different states 1404 - 1410. The differences in the reflection spectra in the hyperspectral images allow the different states of the gap 1402 / substrate 1400 to be identified through analysis of the hyperspectral images. In some implementations, the machine learning model 132 is trained with hyperspectral images including the reflection spectra corresponding to different substrates in different states during the process (e.g., including different fill levels of gaps on the substrate), such that the trained machine learning model 132 can identify the state of the substrate at any given point in the process based at least on analysis of the hyperspectral images of the substrate captured by the hyperspectral camera 128 during the process.

[0159] In some implementations, the controller 130 is configured to generate a thickness map of the substrate from a plurality of different hyperspectral images of the substrate captured at different points during the process. The thickness map provides a visual representation of film growth (or various other states of the substrate) during the process. By training the machine learning model 132 in this way, the trained machine learning model 132 can determine whether there are any problems with the substrate during the process and identify the type of problem if one does occur.

[0160] Figure 15 Schematically shows different exemplary states of the substrate 1500 during a fill process, where voids are formed in the gaps 1502 of the features of the filled substrate 1500. At 1504, the deposited film 1505 has tapered sidewalls near the bottom of the gap 1502. This may be caused, for example, by not saturating the substrate surface within the gap 1502 during an ALD process. Continuing to 1506, the tapered sidewalls remain as the film thickens. At 1510, it can be seen that after the gap fill process is complete, voids 1511 remain. The gap 1502 with voids 1511 will have a different reflection spectrum in the hyperspectral image compared to a gap filled with a void-free film.

[0161] In the various examples above, the problem (or lack thereof) with the substrate 1500 manifests as a change in the reflection spectrum in the hyperspectral image of the substrate 1500. In these examples, the machine learning model 132 can identify the problem with the substrate 1500 based at least on an analysis of the hyperspectral images of the substrate 1500 captured before, during, and / or after performing the fill process to determine the change in the reflection spectrum of the substrate 1500 before, during, and / or after performing the fill process. The machine learning model 132 can analyze these hyperspectral images to determine whether the process is being performed correctly, and the controller 130 can dynamically adjust the process to compensate for any problems identified by the machine learning model 132. Such control can be performed in-situ during the process or between different batches of the process depending on the implementation.

[0162] In some implementations, the position of the hyperspectral camera 128 is configured to be dynamically adjustable to adjust the distance of the hyperspectral camera 128 relative to the scene / object being imaged. By changing the distance of the hyperspectral camera 128 relative to the scene / object being imaged, the field of view of the hyperspectral camera 128 and correspondingly the physical size of the pixels in the hyperspectral images generated at different distances change. This allows for better control of hyperspectral measurements in the region of interest (especially in a relatively small region of interest), such as for determining the haze level in a layer of the substrate. Additionally, different hyperspectral images of the scene / object captured at different distances relative to the scene / object being imaged can be compared to each other in order to distinguish relevant metrology data from noise.

[0163] Figure 16 An exemplary scenario is schematically shown in which a hyperspectral camera 1600 is configured to be dynamically adjustable to adjust the distance of the hyperspectral camera 1600 relative to the substrate 1602 being imaged. The hyperspectral camera 1600 is located at a processing tool (e.g., Figure 1 100). In other examples, the hyperspectral camera 1600 can be located in a different portion of the process tool, such as a process chamber module, a load lock module, or an equipment front end module (EFEM).

[0164] The hyperspectral camera 1600 is located above the slit valve 1606 in the transfer module 1604. The robotic arm 1608 holds the substrate 1602 and moves the substrate 1602 within the transfer module 1604. The robotic arm 1608 passes the substrate 1602 through the slit valve 1606 when the substrate 1602 is transferred from the transfer module 1604 to the processing module. In addition, the robotic arm 1608 receives the substrate 1602 from the slit valve 1606 when the substrate 1602 is transferred from the processing module to the transfer module 1604. In the example shown, the height of the hyperspectral camera 1600 is adjustable within the transfer module to adjust the distance between the hyperspectral camera 1600 and the substrate 1602. When the substrate enters or leaves the slit valve 1606, the hyperspectral camera 1600 can capture one or more hyperspectral images of the substrate 1602 from different distances.

[0165] In one example, at a first time (T1), the hyperspectral camera 1600 is positioned at a first height (H1) in the transfer module 1604, such that the hyperspectral camera 1600 is located at a first distance (D1) from the substrate 1602. At the first distance (D1), the substrate 1602 is completely positioned within the field of view 1610 of the hyperspectral camera 1600. The hyperspectral camera 1600 captures one or more hyperspectral images of the substrate 1602 from the first position. In some examples, the hyperspectral camera 1600 is a line scan camera that scans the substrate 1602 as it passes under the hyperspectral camera 1600 and enters the slit valve 1606. In other examples, the hyperspectral camera 1600 is configured to take a snapshot of the entire substrate 1602 at a specific index position before the substrate is transferred to the slit valve 1606.

[0166] At a second time (T2), the hyperspectral camera 1600 is dynamically adjusted relative to the substrate 1602. In particular, the hyperspectral camera 1600 is lowered to a second height (H2) in the transfer module 1604 such that the hyperspectral camera 1600 is at a second distance (D2) that is closer to the substrate 1602 than the first distance (D1). At the second distance (D2), only a portion of the substrate 1600 is within the field of view 1610 of the hyperspectral camera 1600. The hyperspectral camera 1600 captures one or more hyperspectral images of the substrate 1602 from this second position. Since the hyperspectral camera 1600 is positioned at a second position closer to the substrate 1602 than at the first position, the pixels of the hyperspectral image captured by the hyperspectral camera 1600 at the second position correspond to smaller or more granular regions of the substrate 1602 relative to the pixels of the hyperspectral image captured by the hyperspectral camera 1600 when the hyperspectral camera 1600 was in the first position.

[0167] Alternatively or additionally, in some implementations, the hyperspectral camera 1600 can include one or more optical elements (e.g., a zoom lens) that are configured to optically adjust the distance between the image sensor of the hyperspectral camera 1600 and the scene / object being imaged (e.g., the substrate 1602). The one or more optical elements can be dynamically adjusted to adjust the distance of the hyperspectral camera 128 relative to the scene / object being imaged.

[0168] Alternatively or additionally, in some implementations, the substrate 1602 can be moved relative to the position of the hyperspectral camera 1600 by a robotic arm 1708 to dynamically adjust the distance between the hyperspectral camera 1600 and the substrate 1602.

[0169] In some implementations, a machine learning model 132 is trained based at least on hyperspectral images of a scene (e.g., a processing chamber) and / or an object (e.g., a substrate) captured at different distances relative to the hyperspectral camera that captures the hyperspectral images. The trained machine learning model 132 can be configured to receive one or more hyperspectral images of a substrate captured at a first distance relative to the substrate and output metrology data 134 for the substrate based at least on the one or more hyperspectral images captured at the first distance. Additionally, the trained machine learning model 132 can be configured to receive one or more hyperspectral images of a substrate captured at a second distance different from the first distance relative to the substrate and output metrology data 134 for the substrate based at least on the one or more hyperspectral images captured at the second distance. By way of example, the second distance can be less than the first distance. Depending on the implementation, the change in distance can be performed dynamically by physically moving the hyperspectral camera or optically by adjusting the optical elements of the hyperspectral camera.

[0170] In some examples, the metrology data 134 of the substrate can vary depending on the distance of the hyperspectral camera relative to the substrate. In some examples, the hyperspectral camera 1600 is moved closer to the substrate to obtain the metrology data 134 of a particular feature or region of interest, such as to inspect one or more filled gaps or other features on the substrate. In other examples, the hyperspectral camera is dynamically adjusted to capture more of the substrate (e.g., the entire substrate) within the field of view of the hyperspectral camera, and the machine learning model can output the metrology data 134 of the substrate based on the hyperspectral image captured at that distance. In some implementations, the machine learning model 132 is configured to receive hyperspectral images of the substrate captured at different distances, compare the reflection spectra of the different hyperspectral images to distinguish the actual spectral information from noise, and output the noise-filtered metrology data 134 of the substrate.

[0171] In some implementations, the hyperspectral camera 128 is configured to be dynamically adjustable to adjust the angle of the light emitted from the hyperspectral camera 128 relative to the imaged scene / object. The angle of incidence of the light emitted from the hyperspectral camera 128 on the imaged scene / object can change how the light interacts with the surface of the scene / object and affect the reflection spectrum. Depending on the material imaged by the hyperspectral camera 128, some angles of incidence may be more optimal for predicting the output accuracy than others. In some examples, a set of selected angles can be optimized for a particular material.

[0172] Figure 17 An exemplary scenario is schematically shown where the hyperspectral camera 1700 is configured to be dynamically adjustable to adjust the angle of incidence of the light emitted from the hyperspectral camera 1700 on different substrates 1702, 1702' being imaged. The hyperspectral camera 1700 is located within the transfer module 1704 of a processing tool (such as Figure 1 the processing tool 100 shown). In other examples, the hyperspectral camera 1700 can be located in different parts of the processing tool, such as a process chamber module, a load lock module, or an equipment front end module (EFEM).

[0173] The hyperspectral camera 1700 is located above the slit valve 1706 in the transfer module 1704. The robotic arm 1708 holds the substrates 1702, 1702' and moves the substrates 1702, 1702' within the transfer module 1704. When the substrates 1702, 1702' are transferred from the transfer module 1704 to the processing module, the robotic arm 1708 moves the substrates 1702, 1702' through the slit valve 1706. Additionally, when transferring the substrates 1702, 1702' from the processing module to the transfer module 1704, the robotic arm 1708 receives the substrates 1702, 1702' from the slit valve 1706. In the example shown, the angle of the hyperspectral camera 1700 is adjustable within the transfer module to adjust the incident angle of light emitted from the hyperspectral camera 1700 and reaching the substrate being imaged. When the substrates 1702, 1702' enter or exit the slit valve 1706, the hyperspectral camera 1700 can capture one or more hyperspectral images of the substrates 1702, 1702' from different incident angles.

[0174] In one example, at a first time (T1), the hyperspectral camera 1700 is positioned at a first angle (θ1) relative to a first substrate 1702 having a surface film containing a first material. For example, the first angle (θ1) can be selected to be optimized at least based on how the first material of the surface film responds to light of different wavelengths at a selected incident angle. The hyperspectral camera 1700 captures one or more hyperspectral images of the substrate 1702 from this first angle (θ1). In some examples, the hyperspectral camera 1700 is a line scan camera that scans the substrate 1702 as the substrate 1702 passes under the hyperspectral camera 1700 and into the slit valve 1706. In other examples, the hyperspectral camera 1700 is configured to take a snapshot of the entire substrate 1702 at a specific indexing position before the substrate is transferred into the slit valve 1706.

[0175] At a second time (T2), the hyperspectral camera 1700 is dynamically adjusted to a second angle (θ2) relative to a second substrate 1702' having a surface film containing a second material different from the first material of the surface film of the first substrate 1702. For example, the second angle (θ2) can be selected to be optimized at least based on how the second material of the surface film responds to light of different wavelengths at a selected incident angle. The hyperspectral camera 1700 captures one or more hyperspectral images of the second substrate 1702' from this second angle (θ2).

[0176] Alternatively or additionally, in some implementations, the substrates 1702, 1702' can be moved by the robotic arm 1708 relative to the position of the hyperspectral camera 1700 to dynamically adjust the angle between the hyperspectral camera 1700 and the substrates 1702, 1702'.

[0177] Figure 18 Exemplary chart 1800 shows multiple curves of the spectral reflectance of a material on a substrate varying with wavelength. Each of the multiple curves corresponds to a different angle of incidence of light reflected from the substrate and collected by a hyperspectral camera. The multiple curves can be generated based on a hyperspectral image of the substrate. The first curve 1802 corresponds to light having a first angle of incidence (θ1) on the material on the substrate. The second curve 1804 corresponds to light having a second angle of incidence (θ2) on the material on the surface of the substrate. In this example, the second angle of incidence (θ2) is greater than the first angle of incidence (θ1). The third curve 1806 corresponds to light having a third angle of incidence (θ3) on the material on the surface of the substrate. In this example, the third angle of incidence (θ3) is greater than the second angle of incidence (θ2). Note that the angle of incidence changes the way light interacts with the material on the substrate surface and thus has an impact on the reflection spectrum. In other words, the spectral reflectance of the material is different for different angles of incidence at different wavelengths. In addition, it should be noted that the difference in spectral reflectance between different angles of incidence at different wavelengths varies non-uniformly. The periodic nature of curves 1802, 1804, 1806 is caused by an interference pattern generated by light propagating through the material on the substrate.

[0178] In some implementations, the machine learning model 132 is trained at least based on hyperspectral images of a scene (e.g., a processing chamber) and / or an object (e.g., a substrate) captured at different angles relative to the hyperspectral camera that captures the hyperspectral image. In some examples, the angle of the hyperspectral camera for training the hyperspectral image is selected at least based on the material of the film on the substrate being imaged. The training hyperspectral images can be labeled with the angle of incidence of the hyperspectral camera 128 to train the machine learning model 132 to accurately predict the spectral reflectance of the material on the substrate or other metrology data of the object being imaged. The trained machine learning model 132 can be configured to receive one or more hyperspectral images of a substrate having a film containing a first material captured at a first angle selected at least based on the first material of the film, and output metrology data 134 of the first substrate based at least on the one or more hyperspectral images captured at the selected first angle. In addition, the trained machine learning model 132 can be configured to receive one or more hyperspectral images of a second substrate having a film containing a second material different from the first material captured at a second angle selected at least based on the second material of the film, and output metrology data 134 of the second substrate based at least on the one or more hyperspectral images captured at the selected second angle. In some examples, the metrology data 132 can include the spectral reflectance of the material varying with wavelength at different angles of incidence of the hyperspectral camera 128.

[0179] In some implementations, the machine learning model 132 can be trained to distinguish different numbers of material layers on a substrate and / or different thicknesses of one or more material layers on the substrate based at least on analyzing the hyperspectral image of the substrate. Figure 19 An exemplary graph 1900 showing two curves of spectral reflectance of two different substrates as a function of wavelength is shown. These curves can be generated based on the hyperspectral image of the substrate. The two substrates have the same total thickness and different numbers of layers. The first curve 1902 represents the spectral reflectance of a first substrate having a first number of layers (N1). The second curve 1902 represents the spectral reflectance of a second substrate having a second number of layers (N2). In this example, the second number of layers (N2) is greater than the first number of layers (N1). It should be noted that the first curve 1902 corresponding to the first substrate generally has a greater spectral reflectance than the second curve 1904 corresponding to the second substrate. The difference in spectral reflectance can be attributed to the first substrate having layers that are thicker than those of the second substrate. This information can be applied to train the machine learning model 132. In particular, the machine learning model 132 can be trained on hyperspectral images of different substrates having different numbers of layers and different layer thicknesses. The trained machine learning model 132 can be configured to identify the number of layers on the substrate and / or the thickness of the layers on the substrate based at least on analyzing the hyperspectral image of the substrate.

[0180] Stress and warpage are metrics that can be used to evaluate whether a process is being performed on a substrate as expected and for process control to monitor the tool health of the processing tool 100. In some implementations, the processing tool 100 can be configured to perform a scan of the substrate and measurements of the warpage and / or stress of the substrate using the hyperspectral camera 128. Indications of the stress and warpage of the substrate can be manifested at least as a change in curvature, causing some pixels of the hyperspectral image of the substrate to receive more light than other pixels. Additionally, indications of stress and warpage can be manifested at least as an offset of the reflection spectrum in a spectrum without stripes, which appears as a simple right shift or left shift, or as a change in a compression wave in the case of a spectrum with multiple stripes. Stress and / or warpage can be locally measured at multiple different points across the substrate by hyperspectral imaging, and a global measurement of the stress and / or warpage of the substrate can be determined based at least on multiple local measurements of stress and / or warpage.

[0181] Figures 20 - 22 Schematically shows an exemplary configuration in which a hyperspectral camera can be used to scan and measure a processing tool (such as Figure 1The stress and / or warping of the substrate in the processing tool 100 shown. In some implementations, the hyperspectral camera can be configured to perform in-situ scanning of the substrate and in-situ measurement of the stress and / or warping of the substrate during a process in the processing chamber of the processing tool. In other examples, the hyperspectral camera can be configured to perform ex-situ or on-line scanning of the substrate and ex-situ or on-line measurement of the stress and / or warping of the substrate when the substrate is in a transfer module, a load lock module, or an equipment front-end module (EFEM).

[0182] Figures 20 - 21 Schematically shows an exemplary configuration where a hyperspectral camera can be used to simulate a spectroscopic ellipsometer to scan a substrate and measure the stress and / or warping of the substrate. The measurement is based on the analysis of the change in the polarization state of light when reflected from the substrate. Ellipsometer measurements provide information about the multiple refractive indices and thicknesses of the film on the substrate. Changes in surface curvature can change the multiple refractive indices, and by changing the polarization of the received light, changes in stress / warping can be regarded as changes in ellipsometer parameters such as Psi (Ψ), Delta (Δ), incident angle (Θi), wavelength (λ), and refractive index (n and k). Psi (Ψ) is the amplitude ratio of the p-polarized component and the s-polarized component of the reflected light. It is related to the phase shift between the two polarization states. Delta (Δ) is the phase difference between the p-polarized component and the s-polarized component of the reflected light. It is related to the offset of the polarization ellipse. The incident angle (Θi) is the angle at which light impinges on the sample surface. The incident angle affects the sensitivity of ellipsometric measurements to film properties. The wavelength (λ) is the wavelength of the incident light. Ellipsometers typically operate in a specific wavelength range, and measurements at different wavelengths can be used to extract more information about the sample. The refractive index (n and k) is the multiple refractive indices (n + ik) of the thin film of the substrate. The real part (n) and the imaginary part (k) are related to the amplitude and absorption of light, respectively.

[0183] In Figure 20 it, the hyperspectral camera 2000 includes a light source 2002 and an image sensor 2004. The light source 2002 directly emits light of different selected wavelengths across the entire electromagnetic spectrum (e.g., UV, visible light, near IR, IR) onto the substrate 2006. The image sensor 2004 captures a hyperspectral image of the light of different wavelengths reflected from the substrate 2006 onto the image sensor 2004. This configuration focuses on the wavelength shift (compression shift) and intensity changes between pixels due to the anisotropic behavior caused by curvature and / or stress to determine warping and / or stress measurements. For example, a flat surface will provide equal amounts of reflected light for each pixel. However, in the case of a curved surface, there is additional interference between the incident light and the reflected light within the dielectric film. This will result in reflected light with changes in phase / k and changes in the amplitude of certain wavelengths.

[0184] InFigure 21 In this case, the hyperspectral camera 2100 includes a light source 2100, an image sensor 2104, and a polarizer 2106. The light source 2002 directly emits light of different wavelengths across the entire electromagnetic spectrum (e.g., UV, visible light, near IR, IR) onto the substrate 2108. The light reflected from the substrate 2108 passes through the polarizer 2106 and reaches the image sensor 2104. The polarizer 2106 changes the polarization of the light reflected from the substrate 2108, which enables the birefringence changes of the substrate 2108 to be observed by the image sensor 2104 to determine the stress and / or warping of the substrate 2108. In particular, the image sensor 2104 can observe the changes in the reflectance of the S and P light passing through the polarizer 2106 due to the stress and / or warping of the substrate 2108. In addition, the polarizer 2106 filters out the components of the reflected light that may interfere with the measurement of the stress and / or warping by the hyperspectral camera 2100, amplifying the change in birefringence relative to the noise in the signal, which allows for a more accurate measurement of the stress and / or warping. The hyperspectral camera 2100 can employ any suitable type of polarizer to change the polarization of the reflected light and filter out the unwanted components of the reflected light for the image sensor 2104 of the hyperspectral camera 2100. Examples include rotating polarizers, linear polarizers, elliptical polarizers, and other types of polarizers.

[0185] Figure 22Schematically shows an exemplary configuration in which a hyperspectral camera 2200 can be used to obtain local and global stress and / or warpage measurements via coherent gradient sensing. Coherent gradient sensing measures the surface slope and gradient of a substrate with high precision and involves analyzing the interference pattern of coherent light to extract information about the surface slope by comparing the phase shift between multiple points when the optical path of the light changes. In the illustrated configuration, a coherent light source 2202 emits coherent light towards a beam splitter 2204. In some implementations, the coherent light source 2202 includes one or more laser sources that produce lasers of different wavelengths. In some implementations, the coherent light source 2202 includes a broadband light source. The coherent light emitted from the coherent light source 2202 is used to generate a clear interference pattern. The beam splitter 2204 is configured to split the coherent light emitted from the coherent light source 2202 into two beams. One of the split beams serves as a reference beam, while the other split beam serves as a test beam that interacts with the substrate 2206. The test beam illuminates the surface of the substrate 2206, and the reflected light interacts with the surface features of the substrate 2206. The reflected test beam interferes with the reference beam, thereby generating an interference pattern. The interference pattern is sensitive to the phase change caused by the surface gradient of the substrate 2206, which indicates stress and / or warpage. The hyperspectral camera 2200 optionally may include a pattern filter 2208, which is configured to filter out unwanted light and adjust the characteristics of the incident light on the image sensor of the hyperspectral camera 2200 according to the requirements of the measurement system. The pattern filter 2208 improves the quality and accuracy of the acquired data by filtering out light that may add noise. The choice of filter depends on factors such as the characteristics of the material being measured, the wavelength range of interest, and the specific requirements of the measurement setup. Exemplary types of optical filters that can be employed in the hyperspectral camera 2200 may include wavelength-selective filters, spatial filters, polarization filters, and / or frequency filters. The image sensor of the hyperspectral camera 2200 captures the interference pattern. The hyperspectral camera 2200 analyzes the phase change of the interference pattern to extract information about the stress and / or warpage of the substrate 2206. The hyperspectral camera 2200 generates a hyperspectral image of the interference pattern at different wavelengths because the phase difference may vary at least based on different wavelengths. In some examples, the hyperspectral image of the substrate 2206 captured by the hyperspectral camera 2200 can be used to determine the stress modulus of the substrate. In some examples, the hyperspectral image of the substrate 2206 captured by the hyperspectral camera 2200 can be used to reconstruct the three-dimensional surface profile of the substrate 2206, which indicates stress and / or warpage on the substrate.

[0186] Figures 20 - 22The hyperspectral camera configuration shown and described above can measure the reflectivity of the substrate pixel-by-pixel across the entire substrate in a hyperspectral image to determine local stress and / or warpage at different pixels and global stress and / or warpage across the substrate surface. In some implementations, the hyperspectral camera can provide an in-line measurement of reflectivity, which can quickly provide insights into wafer stress and / or warpage, allowing for determination of recipe quality and tool health during the process. Additionally, in some implementations, the processing tool 100 can be dynamically adjusted to correct any issues related to the stress and / or warpage of the substrate. For example, the processing tool 100 can transfer the substrate to a different processing chamber to deposit a film on the backside of the substrate, thereby balancing the stress caused by front-side processing.

[0187] In some implementations, the machine learning model 132 is trained at least based on hyperspectral images of different substrates affected by different degrees of stress and / or warpage. The trained machine learning model 132 can be configured to receive one or more hyperspectral images of a substrate and output metrology data 134, which includes the amount of stress and / or warpage of the substrate determined at least based on the one or more hyperspectral images. In some examples, the determined stress and / or warpage can be localized to different points on the substrate. In other examples, the determined stress and / or warpage is globally applied across the entire substrate.

[0188] Figures 23 - 24 An exemplary graph showing spectral reflectivity measured at different wavelengths at different points on the substrate is shown. Figure 23 A graph 2300 showing spectral reflectivity measurements taken at a pixel corresponding to the center point of the substrate is shown. Graph 2300 includes a first curve 2302 and a second curve 2304. The first curve 2302 indicates the spectral reflectivity at different wavelengths at the lowest warpage point (warp A) on the substrate within the pixel, and the second curve 2304 indicates the spectral reflectivity at different wavelengths at the highest warpage point (warp B) on the substrate within the pixel. The curves 2302 and 2304 together indicate the stress and / or warpage of the substrate measured at the center point of the substrate.

[0189] Figure 24 A graph 2400 showing spectral reflectivity measurements taken at a pixel corresponding to an area near the edge of the substrate is shown. Graph 2400 includes a first curve 2402 and a second curve 2404. The first curve 2402 indicates the spectral reflectivity at different wavelengths at the lowest warpage point (warp A) on the substrate within the pixel, and the second curve 2404 indicates the spectral reflectivity at different wavelengths at the highest warpage point (warp B) on the substrate within the pixel. The curves 2402 and 2404 together indicate the stress and / or warpage of the substrate measured at the edge of the substrate. Comparing Figure 23 graph 2300 with Figure 24In the chart 2400, curve 2404 has a longitudinal compression displacement and an amplitude difference relative to curve 2304, which indicates a greater stress and / or warpage at the edge of the substrate relative to the center point of the substrate. When there is a curved surface or warpage in the substrate, there is a phase shift and change in the k vector, which is the angular wave vector of the light reflected from the substrate. There is not only a positional offset on the same substrate, but also the offset is more obvious in the case of different warpage values, such as the case of curve 2404 relative to curve 2304.

[0190] Haze is a metric that provides insight into the diffuseness of a surface and gives an indication of surface roughness. Additionally, the measured value of surface roughness can be used to inform and improve the accuracy of substrate thickness determination by providing a roughness correction factor (which is considered in thickness determination). In some implementations, the processing tool 100 can be configured to use the hyperspectral camera 128 to measure the haze of the substrate. Figure 25 An exemplary configuration is shown where the hyperspectral camera 2500 is configured to measure the haze on the substrate 2502. In some implementations, the hyperspectral camera 2500 can be configured to perform in-situ measurements of the haze of the substrate during a process in the processing chamber of the processing tool 100. In other examples, the hyperspectral camera 2500 can be configured to perform non-in-situ or on-line measurements of the haze of the substrate when the substrate is in a transfer module, a load lock module, a front-opening unified pod (FOUP), or an equipment front-end module (EFEM).

[0191] The hyperspectral camera 2500 includes a light source 2504 and an image sensor 2506. The light source 2504 includes a plurality of spatially separated light emitters. In some examples, the light emitters include broadband light sources. In some examples, the light emitters include LEDs configured to emit light at a specific wavelength. The spatially separated light emitters are configured to emit light onto different spatially separated regions of the substrate 2502. Each region corresponds to a plurality of pixels spaced far enough apart such that light from one light emitter only emits light onto a single region and does not emit light onto other regions of the substrate 2502. In the example shown, the light emitted from the light emitters of the light source 2504 illuminates the pixel 2508 on the substrate 2502. When the surface is more ideally specular (smooth), the surrounding pixels around the illuminated pixel 2508 will be dark. When the surface is not specular, the surrounding pixels will exhibit diffuse reflection 2510. The diffuse reflection 2510 can be caused by a variety of factors, including but not limited to the presence of particles, crystal structure, defects, crystal orientation (which results in anisotropic dispersion), surface roughness (topographical variations), or any combination thereof. In one example, haze is measured by observing the ratio of incident light to scattered light. The greater the angular spread at the point of incidence, the higher the degree of diffusion of the substrate surface. The state of the substrate surface will affect how the incident light is scattered, but with hyperspectral imaging and a broadband light source, the instrument can also show which wavelengths of light are more or less affected. This can serve as a potential correction factor for thickness / stress and potentially indicate the presence of larger defects / contaminants / particles.

[0192] It should be noted that the example shown depicts a single pixel 2508 on the substrate 2502 being illuminated to measure haze. In other examples, various other pixels spaced apart across the entire substrate 2502 can be simultaneously illuminated with different light emitters to measure haze in different regions of the substrate.

[0193] In different implementations, the plurality of spatially separated light emitters can be arranged differently in the light source 2504. Figures 26 - 27 An exemplary arrangement of the spatially separated light emitters in the light source is schematically shown. In Figure 26In the example, the light source 2600 includes a plurality of spectral light emitters 2602, a plurality of near-IR light emitters 2604, and a plurality of collimated red light emitters 2606. In some examples, the plurality of spectral light emitters 2602 includes multiple sets of red, green, and blue light emitters (e.g., LEDs). In this example, the spectral light emitters 2602 are evenly spaced from each other across the entire light source 2600. In this example, the near-IR light emitters 2604 are evenly spaced from each other across the entire light source 2600. The collimated red light emitters 2606 are designated for measuring haze and are spaced farther apart from each other compared to the spectral light emitters 2602 and the near-IR light emitters 2604. It should be noted that different types of light emitters can have any suitable spacing on the light source 2600. By spacing the collimated red light emitters 2606 farther apart than the other light emitters of the light source 2600, the light emitted by these light emitters can be directed to different regions of the substrate to measure haze without the light from other collimated red light emitters undesirably illuminating that region. In one example, in this arrangement, the collimated red light emitters can be used to measure haze, and the spectral light emitters 2602 and the near-IR light emitters 2604 can be used to perform thickness / stress / warpage measurements and other operations.

[0194] In Figure 27 the example, the light source 2700 includes a majority of light emitter groups 2702. Each light emitter group 2702 includes one or more spectral light emitters 2704, one or more near-IR light emitters 2706, and one or more collimated red light emitters 2708. These light emitter groups 2702 are spaced apart from other light emitter groups such that the light emitter groups 2702 illuminate different regions of the substrate for haze measurement without providing light pollution to other regions. In the example shown, in addition to other metrics / operations (e.g., thickness, stress, warpage measurements), all the light emitters in different groups can also be used to measure haze.

[0195] In some implementations, light of different wavelengths can be selected to illuminate the substrate to measure the haze of the substrate. Figure 28 An exemplary diagram 2800 of the electromagnetic spectrum is shown, which includes different wavelength ranges that can be used for different operations. In particular, light in the spectral wavelength ranges (e.g., red, green, blue) 2802 and the near-IR wavelength range 2804 can be designated for metrics such as measuring thickness, stress, warpage. Additionally, a wavelength range outside the spectral and near-IR ranges, such as the wavelength range 2806, can be designated for measuring haze. By using the wavelength range 2806 to measure haze, the spectral and near-IR wavelength ranges will not provide light pollution because these wavelength ranges are not considered when quantifying the haze of the substrate.

[0196] Figure 29Shows an exemplary graph 2900, which includes haze measurement values 2902 represented by the number of pixels that reflect light and their corresponding intensities. The height of the haze measurement values 2902 corresponds to the smoothness of the surface, and the spread of the haze measurement values 2902 corresponds to the roughness of the surface. The haze measurement values 2902 correspond to a single light emitter.

[0197] In some implementations, the processing tool includes a plurality of hyperspectral cameras located at different positions within the processing tool to provide inputs for controlling the processing tool. Figure 30 Schematically shows an exemplary processing tool 3000, which includes a plurality of modules 3002, 3004, 3006 connected to a vacuum transfer chamber 3008. The vacuum transfer chamber 3008 includes a plurality of hyperspectral cameras 3010, 3012, 3014 corresponding to the plurality of modules 3002, 3004, 3006. Substrates can be transferred between different modules 3002, 3004, 3006 to perform different processes on the substrates. When the substrate is transferred from one module to another, the substrate passes through the vacuum transfer chamber 3008 and the corresponding hyperspectral camera can capture the hyperspectral image of the substrate. For example, each time the substrate enters and exits a module, the corresponding hyperspectral camera can capture the hyperspectral image of the substrate to determine how the substrate is changed by the process performed by the module. In addition, the processing tool 3000 can be configured to control how to process the substrate based on the hyperspectral image of the substrate and the corresponding analysis performed on the hyperspectral image (e.g., by Figure 1 the machine learning model 132 shown).

[0198] In one example, a film deposition process is performed on the substrate in module 3004. The hyperspectral camera 3012 captures the hyperspectral image of the substrate before and after performing the process. The hyperspectral image is analyzed by the machine learning model 132, and the machine learning model determines that the process causes the substrate to bend at least based on the analysis of the hyperspectral image. The processing tool 3000 transfers the substrate to module 3006 to perform backside film deposition on the substrate at least based on the output of the machine learning model 132 to compensate for the warping on the opposite side of the substrate. The processing tool 3000 can be configured to dynamically adjust the control of the processing tool 3000 to perform any suitable process on the substrate at least based on the analysis of the hyperspectral image of the substrate performed by the machine learning model 132.

[0199] In some implementations, the module includes a plurality of hyperspectral cameras located at different positions within the module to provide inputs for controlling the process performed by the module. Figure 31 Schematically shows an exemplary module 3100, which contains a plurality of hyperspectral cameras. For example, module 3100 can correspond to Figure 30 the processing tool 3000 shown in Figure 1Any module of the processing tool 100 shown. Module 3100 is configured to perform processing on four different substrates 3102, 3104, 3106, 3108 at a time. The substrates can be indexed and transferred between four different positions within module 3100 to perform different processes on different substrates. Module 3100 includes four hyperspectral cameras 3110, 3112, 3114, 3116 corresponding to the four substrates 3102, 3104, 3106, 3108. The four hyperspectral cameras 3110, 3112, 3114, 3116 are configured to capture hyperspectral images of the four substrates 3102, 3104, 3106, 3108 before, during, and / or after performing a process on the four substrates 3102, 3104, 3106, 3108. The machine learning model 132 analyzes the hyperspectral images and outputs metrology data for the four substrates 3102, 3104, 3106, 3108. The processing tool dynamically controls the process performed by module 3100 on the substrate based at least on the metrology data output from the machine learning model 132.

[0200] In one example, a film deposition process is performed on substrate 3102 at the first position in module 3100. Hyperspectral camera 3110 captures a series of hyperspectral images of substrate 3102 during the process. The machine learning model 132 analyzes the series of hyperspectral images and determines that the amount of film growth on the substrate is less than expected. The processing tool dynamically adjusts the process based at least on the output of the machine learning model 132 to increase the film growth rate in order to compensate for the determined shortfall in the amount of film growth and thereby achieve the expected amount of film growth on the substrate.

[0201] In another example, a film deposition process is performed on substrate 3102 at the first position in module 3100. Hyperspectral camera 3110 captures hyperspectral images of substrate 3102 before and after performing the process on substrate 3102. The machine learning model 132 analyzes the hyperspectral images captured by hyperspectral camera 3110 and outputs metrology data indicating that the process performed on substrate 3102 is defective. The processing tool dynamically adjusts the next process to be performed on the substrate at the second position in module 3100 based at least on the output of the machine learning model 132. Substrate 3102 is moved to the second position in module 3100 and hyperspectral camera 3112 captures hyperspectral images of substrate 3102 before and after performing the next dynamically adjusted process on substrate 3102 at the second position. The machine learning model 132 analyzes the hyperspectral images captured by hyperspectral camera 3112 and outputs metrology data indicating that the process performed on substrate 3102 is proceeding as expected. Accordingly, substrate 3102 continues to be processed at the remaining positions in module 3100.

[0202] Module 3100 can be configured to dynamically adjust a process being performed on a substrate based at least on an analysis of a hyperspectral image of the substrate performed by machine learning model 132. Additionally, module 3100 can be configured to dynamically adjust any future processes performed on the substrate based at least on an analysis of a hyperspectral image of the substrate performed by machine learning model 132.

[0203] Figure 32 A flowchart depicting an exemplary method 3200 is shown, which method 3200 dynamically controls the position of a hyperspectral camera within a processing tool to vary the distance between the hyperspectral camera and a substrate, thereby performing hyperspectral imaging and analysis. For example, method 3200 can be performed by Figure 1 controller 130 of processing tool 100. At 3202, method 3200 includes receiving one or more hyperspectral images of a substrate within the processing tool from a hyperspectral camera located at a first position. At 3204, method 3200 includes sending the one or more hyperspectral images captured at the first distance to a trained machine learning model that is configured to output metrology data for the substrate based at least on the one or more hyperspectral images and the first distance between the hyperspectral camera and the substrate. At 3206, method 3200 includes dynamically adjusting the position of the hyperspectral camera to a second position at a second distance from the substrate. Alternatively or additionally, in some implementations, the optics of the hyperspectral camera (e.g., a zoom lens) can be dynamically adjusted to adjust the optical distance between the hyperspectral camera and the substrate. Alternatively or additionally, in some implementations, the substrate can be moved relative to the position of the hyperspectral camera to dynamically adjust the distance between the hyperspectral camera and the substrate. At 3208, method 3200 includes receiving one or more hyperspectral images of the substrate from a hyperspectral camera positioned at the second distance from the substrate. At 3210, method 3200 includes sending the one or more hyperspectral images captured at the second distance to a trained machine learning model that is configured to output metrology data for the substrate based at least on the one or more hyperspectral images and the second distance between the hyperspectral camera and the substrate. Method 3200 can be performed to capture hyperspectral images of the substrate at different distances, which images are analyzed differently by the trained machine learning model to generate different types of metrology data for the substrate based at least on the distance between the hyperspectral camera and the substrate when the hyperspectral images are captured. For example, the distance between the hyperspectral camera and the substrate can be set to capture a hyperspectral image of the entire substrate, thereby generating globalized metrology data for the entire substrate. Additionally, the distance between the hyperspectral camera and the substrate can be dynamically decreased such that the hyperspectral camera captures an image of a particular feature or region of interest of the substrate to generate localized metrology data for the particular feature or region of interest of the substrate.

[0204] Figure 33 shows a flow chart depicting an exemplary method 3300 that dynamically controls the position of a hyperspectral camera to capture hyperspectral images of different substrates from different angles. For example, method 3300 may be executed by Figure 1 a controller 130 of a processing tool 100. At 3302, method 3300 includes receiving one or more hyperspectral images of a first substrate in a processing tool from a hyperspectral camera positioned at a first angle relative to the first substrate. At 3304, method 3300 includes sending the one or more hyperspectral images of the first substrate captured at the first angle to a trained machine learning model configured to output metrology data for the first substrate based at least on the one or more hyperspectral images and the first angle between the hyperspectral camera and the first substrate. At 3306, method 3300 includes dynamically adjusting the position of the hyperspectral camera such that the hyperspectral camera is positioned at a second angle relative to a second substrate. Alternatively or additionally, in some implementations, the substrate may be moved relative to the position of the hyperspectral camera to dynamically adjust the angle between the hyperspectral camera and the second substrate. At 3308, method 3300 includes receiving one or more hyperspectral images of the second substrate from the hyperspectral camera positioned at the second angle relative to the second substrate. At 3310, method 3300 includes sending the one or more hyperspectral images of the second substrate captured at the second angle by the hyperspectral camera to the trained machine learning model configured to output metrology data for the second substrate based at least on the one or more hyperspectral images and the second angle between the hyperspectral camera and the second substrate. Method 3300 may be executed to capture hyperspectral images of different substrates at different angles, which are analyzed differently by a trained machine learning model to generate different types of metrology data for different substrates based at least on the angle between the hyperspectral camera and the different substrates when the hyperspectral images are captured. For example, different substrates may include films containing different materials that reflect incident light differently from different angles. In some examples, a particular angle of incident light on a particular material may produce more accurate metrology data relative to other angles. Thus, the angle between the hyperspectral camera and the substrate may be set / dynamically adjusted to capture hyperspectral images based at least on the material of the substrate.

[0205] Figure 34 shows a flow chart depicting an exemplary method 3300 that performs metrology-based analysis using hyperspectral images to control a processing tool. For example, method 3400 may be executed by Figure 1The controller 130 of the processing tool 100 executes. At 3402, method 3400 includes receiving one or more hyperspectral images of a substrate in the processing tool from a hyperspectral camera. At 3404, method 3400 includes transmitting the one or more hyperspectral images to a trained machine learning model configured to output metrology data of the substrate based at least on the one or more hyperspectral images. In some implementations, at 3406, the metrology data may include the composition of the gas / plasma in the processing chamber housing the substrate and / or the flow path of the gas / plasma. In some implementations, at 3408, the metrology data may include the thickness and / or density of the substrate. In some implementations, at 3410, the metrology data may include the number of layers in the substrate. In some implementations, at 3412, the metrology data may identify voids in the substrate that are not properly filled during the process performed on the substrate. In some implementations, at 3414, the metrology data may include the magnitude of the stress and / or warpage of the substrate. In some implementations, at 3416, the metrology data may include the magnitude of the haze of the substrate. At 3418, method 3400 includes adjusting one or more control parameters of the process performed by the processing tool based at least on the metrology data of the substrate. Method 3400 can provide metrology-based analysis using hyperspectral images, enabling in-situ or online adjustment and control of the processing tool.

[0206] It should be understood that the configurations and / or methods described herein are exemplary in nature and these specific embodiments or examples should not be considered limiting as many variations are possible. The specific routines or schemes described herein may represent one or more of any number of processing strategies. As such, the various acts shown and / or described may be performed in the order shown and / or described, in other orders, in parallel, or omitted. Similarly, the order of the above processes may be changed.

[0207] The subject matter of the present disclosure includes all novel and non-obvious combinations and sub-combinations of various processes, systems, and configurations, as well as other features, functions, acts, and / or characteristics disclosed herein, and any and all equivalents thereof.

Claims

1. A processing tool, comprising: A processing chamber that includes an optical interface; and A hyperspectral camera arranged to capture a hyperspectral image of the interior of the processing chamber through the optical interface of the processing chamber.

2. The processing tool according to claim 1, wherein the processing chamber includes a pedestal, and the hyperspectral camera is arranged to capture a hyperspectral image of a substrate located on the pedestal through the optical interface.

3. The processing tool according to claim 2, wherein the processing chamber includes a showerhead on an opposite side of the pedestal, and the optical interface is provided on the showerhead.

4. The processing tool according to claim 2, wherein the optical interface is provided on the pedestal.

5. The processing tool according to claim 1, wherein the optical interface is provided on a sidewall of the processing chamber.

6. The processing tool according to claim 1, further comprising: One or more optical elements arranged between the optical interface and the hyperspectral camera, wherein the one or more optical elements are configured to direct electromagnetic radiation through the optical interface to the hyperspectral camera.

7. The processing tool according to claim 1, wherein the processing chamber is a plasma reactor chamber.

8. The processing tool according to claim 1, further comprising a computing system configured to execute a trained machine learning model, the trained machine learning model being configured to receive one or more hyperspectral images from the hyperspectral camera and output metrology data of the processing chamber based at least on the one or more hyperspectral images.

9. The processing tool according to claim 8, wherein the computing system is configured to adjust control parameters of a cleaning process to clean the processing chamber based at least on the metrology data of the processing chamber.

10. The processing tool according to claim 8, wherein the trained machine learning model is configured to receive a series of hyperspectral images of a substrate in the processing chamber during a substrate processing cycle and output time-based metrology data of the substrate based at least on the series of hyperspectral images of the substrate, and wherein the computing system is configured to adjust one or more control parameters of the process of the substrate processing cycle based at least on the time-based metrology data of the substrate during the substrate processing cycle.

11. The processing tool according to claim 8, wherein the trained machine learning model is configured to receive one or more hyperspectral images of a first substrate in the processing chamber during or after a first substrate processing cycle and output metrology data of the first substrate based at least on the one or more hyperspectral images of the first substrate, and wherein the computing system is configured to adjust one or more control parameters of the process of a second substrate processing cycle for a second substrate based at least on the metrology data of the first substrate.

12. A computer-implemented method for controlling a processing tool, the computer-implemented method comprising: Receiving, from a hyperspectral camera, one or more hyperspectral images of a processing chamber of the processing tool; Sending the one or more hyperspectral images to a trained machine learning model configured to output metrology data of the processing chamber based at least on the one or more hyperspectral images; and Adjusting, based at least on the metrology data of the processing chamber, one or more control parameters of a process performed by the processing tool.

13. The computer-implemented method of claim 12, wherein the process is a cleaning process for cleaning the processing chamber, and the one or more control parameters include control parameters of the cleaning process.

14. The computer-implemented method of claim 12, wherein the one or more hyperspectral images include a series of hyperspectral images of a substrate in the processing chamber, wherein the series of hyperspectral images of the substrate are received from the hyperspectral camera during a substrate processing cycle of the substrate, wherein the trained machine learning model is configured to output time-based metrology data of the substrate, and wherein the one or more control parameters are adjusted during the substrate processing cycle of the substrate based at least on the time-based metrology data of the substrate.

15. The computer-implemented method of claim 12, wherein the one or more hyperspectral images include one or more hyperspectral images of a first substrate in the processing chamber during or after a first substrate processing cycle, and wherein the one or more control parameters are adjusted for a second substrate processing cycle of a second substrate based at least on the metrology data of the first substrate.

16. The computer-implemented method of claim 12, wherein the processing chamber is a plasma reactor chamber, and the one or more hyperspectral images of the plasma reactor chamber are captured by the hyperspectral camera when plasma is present in the plasma reactor chamber, and wherein the plasma in the plasma reactor chamber is an illumination source for the hyperspectral camera.

17. A processing tool, comprising: A hyperspectral camera arranged to capture hyperspectral images of a substrate in the processing tool; and A computing system configured to execute a trained machine learning model configured to receive one or more hyperspectral images from the hyperspectral camera and output metrology data of the substrate based at least on the one or more hyperspectral images.

18. The processing tool of claim 17, wherein the metrology data includes the thickness of one or more layers of the substrate.

19. The processing tool of claim 17, wherein the metrology data includes the state of gaps in features of the substrate.

20. The processing tool of claim 17, wherein the hyperspectral camera has a dynamically adjustable position.

21. The processing tool according to claim 17, wherein the hyperspectral camera has a dynamically adjustable angle.

22. The processing tool according to claim 17, wherein the metrology data includes the magnitude of the determined stress and / or warpage in the substrate.

23. The processing tool according to claim 17, wherein the metrology data includes the magnitude of the determined haze in the substrate.

Images