Estimation of chamber component conditions using substrate measurements

Abstract

A method includes processing a substrate in a process chamber according to a recipe, the substrate comprising at least one of a film or a feature after processing. The method further includes generating a profile map of the first substrate. The method further includes processing data from the profile map using a first model, the first model outputting at least one of an estimated mesa condition of a substrate support for the process chamber, an estimated lift pin location condition of the substrate support, an estimated seal band condition of the substrate support, or an estimated process kit ring condition for a process kit ring for the process chamber. The method further includes outputting a notification as a result of the processing.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to detecting conditions of chamber components within a processing chamber, and more particularly to generating a map and / or numerical profiling of substrates processed using the chamber and estimating conditions of one or more chamber components of the chamber based on the map and / or numerical profiling of the substrates.

Background Art

[0002] Substrate processing can include a series of processes for fabricating electrical circuits within a semiconductor according to a circuit design. These processes can be carried out within a series of process chambers. The successful operation of modern semiconductor manufacturing equipment can be aimed at facilitating a stable flow of wafers from one chamber to another during the process of forming electrical circuits within the wafer to form a product. In processes that execute many substrate processes, the conditions of the processing chamber may be changed, and as a result, the processed substrates may fail to meet the target conditions and results.

[0003] The process chamber includes many chamber components. Over time, the chamber components of the process chamber can wear and / or be damaged, which can result in deterioration of the quality of products manufactured using the process chamber and / or disposal of substrates processed using the process chamber.

[0004] In addition, new components may not conform to the specifications.

Summary of the Invention

[0005] The following is a simplified summary of the Disclosure to provide a basic understanding of some aspects of the Disclosure. This summary is not a comprehensive overview of the Disclosure. It is not intended to identify any major or significant elements of the Disclosure, nor to define the scope of any particular embodiment or claim of the Disclosure. Its sole purpose is to present some concepts of the Disclosure in a simplified form as an introduction to the more detailed descriptions that follow.

[0006] In an exemplary embodiment, the method comprises processing a first substrate in a process chamber of a substrate processing system according to a policy, wherein the first substrate has at least one of a film or feature after processing. The method further comprises generating a profile map of at least one of the films or features on the first substrate using a substrate measurement system of the substrate processing system. The method further comprises processing data from the profile map using a first model, the first model outputting at least one of the estimated mesa conditions of the substrate support to the process chamber, estimated lift pin location conditions of the substrate support, estimated seal band conditions of the substrate support, or estimated process kit ring conditions for the process kit ring to the process chamber. The method further comprises outputting a notification of at least one of the estimated mesa conditions of the substrate support, estimated lift pin location conditions, estimated process kit ring conditions, or estimated seal band conditions as a result of processing.

[0007] In exemplary embodiments, a computer-readable medium includes instructions that cause a processing device to perform an action when executed by the processing device. These actions include receiving a profile map of at least one of films or features on the first substrate measured using a substrate measurement system after the first substrate has been processed in a process chamber. These actions further include processing data from the profile map using a first trained machine learning model, the first trained machine learning model outputs at least one of the estimated mesa conditions of the substrate support to the process chamber, estimated lift pin location conditions of the substrate support, estimated seal band conditions of the substrate support, or estimated process kit ring conditions for the process kit ring to the process chamber. These actions further include outputting a notification of at least one of the estimated mesa conditions of the substrate support, estimated lift pin location conditions, estimated process kit ring conditions, or estimated seal band conditions as a result of the processing.

[0008] In exemplary embodiments, a computer-readable medium includes instructions that cause a processing device to perform an action when executed by the processing device. These actions include receiving a first profile map of at least one film or feature on a first substrate, the first profile map being generated by a substrate measurement system after the first substrate has been processed by a process chamber according to a policy for the first time. These actions further include receiving a second profile map of at least one film or feature on a second substrate, the second profile map being generated by a substrate measurement system after the second substrate has been processed by a process chamber according to a policy for the second time. These actions further include comparing the first profile map with the second profile map. These actions further include determining, based on the results of the comparison, at least one of the following: a) estimated mesa conditions for one or more mesas of a substrate support in the process chamber, b) estimated lift pin location conditions for one or more lift pins of a substrate support, or c) estimated seal band conditions for a seal band of a substrate support.

[0009] This disclosure is presented as an example, not an limitation, and in the drawings attached, similar references indicate similar elements. It should be noted that various references to “an embodiment” or “one embodiment” in this disclosure do not necessarily refer to the same embodiment, but rather mean at least one. [Brief explanation of the drawing]

[0010] [Figure 1] This figure shows an exemplary computer system architecture according to an aspect of the present disclosure. [Figure 2A] This is a schematic top view of an exemplary manufacturing system according to an aspect of the present disclosure. [Figure 2B] This is a schematic cross-sectional side view of a substrate measurement system included in an exemplary manufacturing system according to an aspect of the present disclosure, as shown in Figure 2A. [Figure 2C]This is a schematic cross-sectional side view of a substrate measurement subsystem according to an aspect of the present disclosure. [Figure 3] This figure shows an exemplary system architecture for predicting chamber component conditions of a processing chamber according to an aspect of the present disclosure. [Figure 4] This figure shows a model training workflow and a model application workflow for determining chamber component conditions according to one embodiment. [Figure 5] This is a flowchart of a method for generating a training dataset for training a machine learning model according to the aspects of this disclosure. [Figure 6] This is a flowchart illustrating a method for training a machine learning model to determine the conditions of a chamber component according to one embodiment. [Figure 7] This is a flowchart illustrating a method for determining the chamber conditions of one or more chamber components using one or more trained machine learning models according to an aspect of the present disclosure. [Figure 8] This is a flowchart illustrating a method for determining the chamber conditions of one or more chamber components using one or more trained machine learning models according to an aspect of the present disclosure. [Figure 9] This is a flowchart of a method for determining the chamber conditions of one or more chamber components based on measurements of a substrate processed by a process chamber over time, according to an aspect of the present disclosure. [Figure 10] This is a flowchart of a method for determining the chamber conditions of one or more chamber components based on a comparison of data and curves, according to an aspect of the present disclosure. [Figure 11A] This is a flowchart illustrating a method for performing a quality evaluation of a process chamber according to an aspect of this disclosure. [Figure 11B] This is a flowchart of a method for determining how a chamber component is installed within a process chamber, according to an aspect of the present disclosure. [Figure 12] This figure shows a profile map (e.g., a heat map) of a processed substrate according to an aspect of the present disclosure. [Figure 13]This figure shows a profile map (e.g., a heat map) of a portion of a processed substrate according to an aspect of the present disclosure. [Figure 14] This is a diagrammatic representation of a machine, an exemplary form of a computing device, capable of executing a set of instructions to cause the machine to perform one or more of the techniques discussed herein. [Modes for carrying out the invention]

[0011] Embodiments of this disclosure relate to systems and methods for evaluating the conditions of one or more chamber components within a process chamber, such as a process chamber for semiconductor processing. The process results of a manufacturing process depend on many factors, including process policies, tool parameter settings (e.g., chamber parameter settings, and settings for parameters of one or more tools outside the chamber), and chamber component conditions. For example, process results may vary across the surface of a substrate based on conditions such as showerhead conditions, substrate support conditions, chamber liner conditions, pump and / or valve conditions. Any changes over time in a substrate support, such as an electrostatic chuck, clamp, vacuum chuck, heater, support including a pocket with a lip at the edge of the support, and / or a substrate support including one or more embedded features (e.g., heater, cooling plate, electrical element), slowly affect the substrate results, such as temperature changes across the surface of the processed substrate. Changes in the substrate support (e.g., chuck) may, additionally or otherwise, negatively affect chucking and / or dechucking of the substrate, the substrate's lift pin area, the substrate's seal band area at the periphery of the substrate, etc. The embodiment provides a novel mechanism for evaluating the condition of chamber components. The embodiment can be used, for example, to show slow changes or degradation over time of substrate supports and / or other chamber components. The embodiment can be used to determine whether chamber components, such as process kit rings, are properly installed after a maintenance event or during new installation, and / or to determine how the chamber components are installed (e.g., the arrangement of the chamber components). The embodiment can be used to perform a quality assessment of a process chamber. Information from the chamber component assessment can be used to determine whether to adjust one or more settings for the process chamber, whether to adjust the process policy, whether to perform maintenance on the process chamber, and / or when to perform it (e.g., by repairing and / or replacing one or more chamber components of the process chamber), etc.

[0012] In some embodiments, a process is performed in a process chamber to deposit or etch films, layers, and / or one or more features (e.g., measurable features) onto the surface of a substrate. The substrate may be a bare substrate or a test substrate without a product. Alternatively, the substrate may be a product substrate. The film may be a blanket film deposited on the surface of the substrate. The film may be a laminate consisting of multiple layers, which may be patterned or unpatterned. In some embodiments, the film includes a laminate consisting of patterned layers having features such as trenches, structures, or devices. Such features may include or may have measurable limit dimensions (e.g., the width of a trench at one or more depths of a trench). The measured features may include features dispersed across the surface of the substrate. Features may include, for example, mesa, dots, structures, valleys, walls, lines, grooves, references, limit dimensions, etc. In one embodiment, the film is a polymer film and / or the features are polymer features.

[0013] After the film and / or features have been deposited or etched onto the substrate, the substrate can be removed from the process chamber and placed into a substrate measurement system. The substrate measurement system may be, for example, a reflected light measurement system or other measurement system that measures the film and / or feature thickness at multiple locations on the substrate. A thickness profile map can be generated for the substrate using the thickness and / or feature information. Alternatively, one or more other profile maps of the substrate (e.g., optical constants, roughness, particle count, etc.) can be generated from other measurement data. For example, if the substrate includes a laminate consisting of patterned layers, the profile map may include a limit dimension profile map. Next, using a model (e.g., a trained machine learning model, a physics-based model, a statistical model, etc.) and / or image processing, thickness information of the film and / or features (e.g., a thickness profile map), or other film and / or feature information (e.g., other profile maps such as an optical constant profile map, a particle number profile map, a critical dimension profile map, etc.), variations in one or more film properties can be identified, and based on the identified variations in properties (e.g., variations in film thickness, variations in critical dimensions, etc.), information about one or more chamber components of the process chamber can be determined. Embodiments will be discussed with reference to measuring a film on a substrate. However, it should be understood that embodiments also apply to measuring one or more features on a substrate (e.g., critical dimensions) in addition to, or instead of, measuring a film on a substrate.

[0014] When measuring a product substrate (e.g., a product wafer) to determine its profile map, the process chamber's operational time (the time the process chamber is used to process the product) is not affected by frequent measurements, allowing for frequent measurements of the product substrate. Alternatively, when processing and measuring a test substrate to determine its profile map, the product is generally not processed while the test substrate is being processed, meaning measurements may not be performed as frequently, resulting in a reduced chamber operational time.

[0015] In one example, the substrate support surface may include many features corresponding to substrate chucking and dechucking (or clamping and unclamping), temperature control, backside helium control, and lift pin control. Features of the substrate support may include, for example, a series of mesa and / or channels on the upper surface of the substrate support, such mesa and / or channels providing areas that can be filled with a thermally conductive gas, such as helium, during processing to facilitate thermal contact between the supported substrate and the substrate support while supporting the substrate. Features of the substrate support may further include sealing bands on the periphery of the upper surface of the substrate support, such sealing bands sealing a thermally conductive gas in the region between the lower surface of the substrate and the upper surface of the substrate support. Features of the substrate support may further include lift pin holes for housing lift pins for lifting the substrate onto or lowering the substrate from the substrate support, and sealing bands around the lift pin holes. Features of the substrate support may further include a dielectric layer crossing the surface of the substrate support. The features of the substrate support may further include bonds between components of the substrate support (e.g., between the ceramic pack and the cooling plate, or between the ceramic plates of the electrostatic chuck). Because the substrate support is used in a process chamber, one or more of the surface features of the substrate support may begin to wear, break, or fail. In addition, the bonds may also fail, causing delamination between layers or bodies of the substrate support. In another example, recesses or valleys in the surface of the substrate support, the flatness of the surface of the substrate support, the roughness across the surface of the substrate support, the concentricity of one or more circular features of the substrate support, etc., may affect the substrate being processed while supported by the substrate support. Embodiments provide a mechanism for detecting erosion and / or degradation of various features of the substrate support, and / or for detecting one or more properties of the substrate support (e.g., surface roughness, concentricity of circular elements, recesses or grooves, flatness, etc.) without adding any additional measuring devices to the process chamber housing the substrate support.In addition, embodiments can identify erosion of the process kit ring around the substrate support and can be used to determine whether and / or when to replace the process kit ring.

[0016] In embodiments, the systems and methods described herein can also be used to evaluate the conditions of chamber components other than the substrate support. For example, the systems and methods can be used to estimate gas delivery (e.g., flow rate and / or total delivered gas) for each of a plurality of regions of a showerhead's gas delivery holes, a gas delivery device (e.g., a gas distribution plate or showerhead), erosion of a protective layer coating on the surface of the gas delivery device, gas or plasma delivery rates from various parts of the gas delivery device, erosion of the process kit ring, etc., to evaluate the conditions of one or more characteristics of other chamber components. According to embodiments of the present disclosure, the conditions of many other types of chamber components can also be evaluated.The conditions of the chamber components can be used to perform a diagnosis of the process chamber, improve the design of one or more components of the process chamber, determine when to perform maintenance on one or more components of the process chamber, and / or when to replace those components, and so on.

[0017] In some embodiments, the methods and systems discussed herein can be used to detect the floating, wear, and / or erosion of chamber components over time. In addition, the methods and systems can be used to evaluate whether a new component meets the specifications. In one example, when a component is switched during a scheduled maintenance, the component can be approved by performing tests in the process described herein and approving that the new component is within the specified range of values.

[0018] Therefore, the embodiments add new detection capabilities to process chambers without increasing the cost of those process chambers. In the embodiments, these new detection capabilities can be used to predict things like decreased production rates, decreased product quality, increased amounts of discarded products, component failures, defects in chamber component design, etc. In the embodiments, these new detection capabilities can be further used to determine when maintenance should be performed on one or more chamber components of the process chamber, resulting in a reduction of unplanned downtime, a reduction in product discards, a reduction in chamber failures, etc. In the embodiments, these new detection capabilities can be further used to adjust one or more process policies, adjust the settings and / or positions of one or more chamber components, perform closed-loop control of the process, etc.

[0019] Figure 1 shows an exemplary computer system architecture 100 according to an aspect of the present disclosure. The computer system architecture 100 includes a client device 120, manufacturing equipment 122, a substrate measurement system 126, a prediction server 112 (for example, for generating prediction data, providing model adaptation, using a knowledge base, etc.), and a data store 150. The prediction server 112 may be part of the prediction system 110. The prediction system 110 may further include server machines 170 and 180. In some embodiments, the computer system architecture 100 may include, or be part of, a manufacturing system for processing substrates, such as the manufacturing system 200 in Figure 2A. In additional or alternative embodiments, the computer system architecture 100 may include, or be part of, a chamber component condition prediction system (for example, for evaluating the conditions of one or more chamber components in a process chamber). Further details regarding the chamber component condition prediction system are provided with respect to Figures 3 and 4.

[0020] The components of the client device 120, manufacturing equipment 122, substrate measurement system 126, prediction system 110, and / or data store 150 can be connected to each other via network 140. In some embodiments, network 140 is a public network providing client device 120 with access to prediction server 112, data store 150, and other publicly available computing devices. In some embodiments, network 140 is a private network providing client device 120 with access to manufacturing equipment 122, substrate measurement system 126, data store 150, and / or other privately available computing devices. Network 140 may include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.

[0021] The client device 120 may include computing devices such as personal computers (PCs), laptops, mobile phones, smartphones, tablet computers, netbooks, network-connected televisions ("smart TVs"), network-connected media players (e.g., Blu-ray® players), set-top boxes, over-the-top (OTT) streaming devices, and operator boxes.

[0022] The manufacturing equipment 122 can produce products according to a policy. In some embodiments, the manufacturing equipment 122 may include, or be part of, a manufacturing system that includes one or more stations (e.g., process chambers, transfer chambers, load locks, factory interfaces, etc.) configured to perform various operations on the substrate.

[0023] The substrate measurement system 126 may be a component of a manufacturing system that can be used to measure substrates before and / or after they are processed in one or more process chambers. The substrate measurement system 126 may be configured to generate emission spectroscopy data, reflected light measurement data, and / or other measurement data. The substrate measurement system 126 may include one or more components configured to collect and / or generate measurement data associated with one or more portions of the surface profile of the substrate after the substrate has been removed from the process chamber.

[0024] In some embodiments, the substrate measurement system 126 can be configured to generate measurement data associated with substrates processed by other manufacturing equipment 122. The measurement data may include one or more values ​​from among film property data (e.g., wafer space film properties such as thickness), dimensions (e.g., thickness, height), dielectric constant, dopant concentration, density, defects, etc. The measurement data may be of a finished or semi-finished product, or a test substrate such as a blanket wafer. In some embodiments, the use of reflected light measurement data and thickness profile maps to determine the conditions of a chamber component will be discussed with reference. However, it should be understood that the principles and embodiments described herein in relation to reflected light measurement data and thickness profile maps also apply to other types of measurement data. For example, measurements such as particle number, optical constants of the coating, surface roughness of the coating, and material composition of the coating can be performed. Such measurements can be performed over many areas on the substrate and can be used to generate profile maps of particle number, optical constants, surface roughness, material composition, etc., across the measured substrate.

[0025] The substrate measurement system 126 can be configured to generate measurement data associated with a substrate before or after a substrate process. The substrate measurement system 126 can be integrated with a station in a manufacturing system that includes manufacturing equipment 122. In some embodiments, the substrate measurement system 126 can be coupled to or part of a station of process tools (e.g., a process chamber, a transfer chamber, etc.) that is maintained in a vacuum environment. Such a substrate measurement system 126 can be called an integrated measuring instrument. Thus, the substrate can be measured by the substrate measurement system 126 while the substrate is in a vacuum environment. For example, the substrate measurement system can be mounted in a transfer chamber maintained under vacuum. For example, after a substrate process (e.g., an etching process, a deposition process, etc.) has been performed on the substrate, the substrate measurement system 126 can generate measurement data for the processed substrate without the processed substrate being removed from the vacuum environment. Such a substrate measurement system 126 can be called an integrated measuring instrument. In other or similar embodiments, the substrate measurement system 126 can be coupled to or part of a manufacturing system (e.g., a factory interface module) that is not maintained in a vacuum environment. In some embodiments, the substrate measurement system is mounted in a process chamber or tool cluster that is not maintained under vacuum, such as a chemical mechanical polishing (CMP) process chamber.

[0026] Alternatively, instead of the substrate measurement system 126 being integrated into the manufacturing system (e.g., mounted on a factory interface or transfer chamber), the substrate measurement system 126 may be a separate (i.e., external) device from the manufacturing equipment 122. For example, the substrate measurement system 126 may be a standalone device not coupled to any station of the manufacturing equipment 122. To obtain measurements on a substrate using a detached substrate measurement system 126, a user of the manufacturing system (e.g., a technician, operator) can have a substrate processed by the manufacturing equipment 122 removed from the manufacturing equipment 122 and transferred to the substrate measurement system 126 for measurement. In some embodiments, the substrate measurement system 126 can transfer the measurement data generated on the substrate to a client device 120 coupled to the substrate measurement system 126 via a network 140 (e.g., for presentation to a manufacturing user such as an operator or technician). In other or similar embodiments, a manufacturing system user can obtain measurement data on a substrate from the substrate measurement system 126 and provide the measurement data to a computer system architecture via a graphical user interface (GUI) of the client device 120.

[0027] The datastore 150 can be memory (e.g., random access memory), drivers (e.g., hard drivers, flash drivers), a database system, or another type of component or device capable of storing data. The datastore 150 may include multiple storage components (e.g., multiple drivers or multiple databases) that can span multiple computing devices (e.g., multiple server computers). The datastore 150 can store profile maps (e.g., generated from reflected light measurement data, spectral data, etc.), such as film thickness profile maps and / or other substrate profile maps. The film thickness profile maps and / or other substrate profile maps may include historical maps and / or current maps.

[0028] One or more portions of the data store 150 can be configured to store data that is inaccessible to users of the manufacturing system. In some embodiments, all data stored in the data store 150 can be made inaccessible to users of the manufacturing system. In other or similar embodiments, some of the data stored in the data store 150 is inaccessible to the user, while other portions of the data stored in the data store 150 are accessible to the user. In some embodiments, the inaccessible data stored in the data store 150 is encrypted using an encryption mechanism unknown to the user (for example, the data is encrypted using a secret encryption key). In other or similar embodiments, the data store 150 may include multiple data stores, where data inaccessible to the user is stored in a first data store and data accessible to the user is stored in a second data store.

[0029] In some embodiments, the prediction system 110 includes server machines 170 and 180. Server machine 170 includes a training set generator 172 capable of generating a training dataset (e.g., a set of data inputs and a set of target outputs) for training, certifying, and / or testing one machine learning model 190 or a set of machine learning models 190. Some operations of the training set generator 172 are described below in detail with reference to Figure 5. In some embodiments, the training set generator 172 can divide the training data into a training set, a certification set, and a test set.

[0030] The server machine 180 may include a training engine 182. The engine can refer to hardware (e.g., circuits, dedicated logic, programmable logic, microcode, processing devices, etc.), software (e.g., instructions executed on processing devices, general-purpose computer systems, or dedicated machines), firmware, microcode, or a combination thereof. The training engine 182 can enable the training of one machine learning model 190 or a set of machine learning models 190. The machine learning models 190 may refer to model artifacts resulting from the training engine 182 using training data containing training inputs and corresponding target outputs (the correct responses for each training input). The training engine 182 can discover patterns in the training data that map training inputs to target outputs (predicted responses) and provide machine learning models 190 that capture these patterns. The machine learning models 190 may include linear regression models, partial least squares regression models, Gaussian regression models, random forest models, support vector machine models, neural networks, ridge regression models, etc.

[0031] The training engine 182 can also enable the authentication of the trained machine learning models 190 using a corresponding set of features from the authentication set of the training set generator 172. In some embodiments, the training engine 182 can assign a performance grade to each of the trained machine learning models 190. The performance grade may correspond to the accuracy of each trained model, the speed of each model, and / or the efficiency of each model. According to embodiments described herein, the training engine 182 can select trained machine learning models 190 having a performance grade that satisfies the performance criteria to be used by the prediction engine 114. Further details regarding the training engine 182 are provided with reference to Figure 6.

[0032] The prediction server 112 includes a prediction engine 114, which provides data from the substrate measurement system 126 (e.g., film thickness profile map) as input to a trained machine learning model 190, and the trained model 190 is run on that input to obtain one or more outputs. In some embodiments, the trained model 190 run by the prediction engine 114 is selected by the training engine 182 as having a performance grade that meets performance criteria. As further illustrated with reference to Figure 7, in some embodiments, the prediction engine 114 uses the model 190 to process the input data and evaluate one or more conditions of one or more chamber components for a process chamber.

[0033] It should be noted that in some other embodiments, the functions of server machines 170 and 180 and the prediction server 112 can be provided by more or fewer machines. For example, in some embodiments, server machines 170 and 180 can be integrated into a single machine, while in some other or similar embodiments, server machines 170 and 180 and the prediction server 112 can be integrated into a single machine. In general, functions described in one embodiment as being performed by server machine 170, server machine 180, and / or prediction server 112 can also be performed on client device 120. In addition, functions attributable to a particular component can also be performed by different or multiple components working together.

[0034] In some embodiments, “User” can represent a single individual. However, other embodiments of this disclosure also include “User” which is an entity controlled by multiple users and / or automated sources. For example, a set of individual users integrated as a group of administrators can be considered a “User.”

[0035] Figure 2A is a schematic top view of an exemplary manufacturing system 200 according to an aspect of the present disclosure. The manufacturing system 200 can perform one or more processes on a substrate 202. According to an aspect of the present disclosure, the substrate 202 can be any preferably rigid, fixed-dimension planar object suitable for fabricating electronic devices or circuit components on it, such as a silicon-containing disk or wafer, a patterned wafer, or a glass plate. In some embodiments, the manufacturing system 200 may include, or be part of, a computer system architecture 110, as described in the embodiment with respect to Figure 1.

[0036] The manufacturing system 200 may include a process tool 204 and a factory interface 206 coupled to the process tool 204. The process tool 204 may include a housing 208 having a transfer chamber 210 inside. The transfer chamber 210 may include one or more processing chambers (also called process chambers) 214, 216, 218 arranged around and coupled to the transfer chamber 210. The processing chambers 214, 216, 218 can be coupled to the transfer chamber 210 through their respective ports, such as slit valves. The transfer chamber 210 may also include a transfer chamber robot 212 configured to transfer substrates 202 between the process chambers 214, 216, 218, a load lock 220, etc. The transfer chamber robot 212 may include one or more arms, each arm including one or more end effectors at the end of each arm. The end effectors can be configured to handle specific objects, such as wafers.

[0037] In some embodiments, the transfer chamber 210 may also include measuring instruments such as a substrate measuring system 126 attached to the transfer chamber 210. The substrate measuring system 126 may be configured to generate measurement data associated with the substrate 202 before or after the substrate process, while the substrate is maintained in a vacuum environment. As shown in Figure 2, the substrate measuring system 126 may be attached to or placed inside the transfer chamber 210. When the substrate measuring system 126 is placed inside or coupled to the transfer chamber 210, measurement data associated with the substrate 202 can be generated without the substrate 202 being removed from the vacuum environment (for example, being transferred to the factory interface 206).

[0038] Process chambers 214, 216, and 218 can be adapted to perform any number of processes on the substrate 202. The same or different substrate processes can be performed within each of the process chambers 214, 216, and 218. Substrate processes can include atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, curing, pre-cleaning, removal of metal or metal oxide, polishing, and wet processing. Other processes can also be performed on the substrate as specified herein.

[0039] The load lock 220 can also be coupled to the housing 208 and the transfer chamber 210. The load lock 220 can be configured to link and couple to the transfer chamber 210 on one side and to the factory interface 206 on the other side. In some embodiments, the load lock 220 may have an environmentally controlled atmosphere that can change from a vacuum environment (to which substrates can be transferred to and from the transfer chamber 210) to an inert gas environment at atmospheric pressure (or near atmospheric pressure) (to which substrates can be transferred to and from the factory interface 206).

[0040] The factory interface 206 can be any suitable housing, such as an equipment front-end module (EFEM). The factory interface 206 can be configured to receive substrates 202 from a substrate carrier 222 (e.g., a front-opening integrated pod (FOUP)) docked to various load ports 224 of the factory interface 206. A factory interface robot 226 (shown by dashed lines) can be configured to transfer the substrates 202 between the substrate carrier (also called a container) 222 and the load lock 220. In other and / or similar embodiments, the factory interface 206 can be configured to receive replacement parts from a replacement parts storage container 222.

[0041] In some embodiments, the manufacturing system 200 may include a substrate measurement system 126 attached to the factory interface 206. The substrate measurement system 126 attached to the factory interface may be configured to generate measurement data associated with the substrate 202 before the substrate 202 is placed in a vacuum environment (e.g., transferred to the load lock 220) and / or after the substrate 202 is removed from the vacuum environment (e.g., removed from the load lock 220).

[0042] The manufacturing system 200 can also be connected to a client device (e.g., client device 120 in Figure 1) configured to provide information about the manufacturing system 200 to a user (e.g., an operator). In some embodiments, the client device can provide information to the user of the manufacturing system 200 via one or more graphical user interfaces (GUIs). For example, the client device can provide information via the GUI about one or more chamber condition metrics for processing chambers 214, 216, and 218 (e.g., while performing a substrate process).

[0043] The manufacturing system 200 may also include or be coupled to a system controller 228. The system controller 228 may be a computing device such as a personal computer, server computer, programmable logic control unit (PLC), or microcontroller, and / or may include such computing devices. The system controller 228 may include one or more processing devices, such as general-purpose processing devices such as a microprocessor or central processing unit. More specifically, the processing devices may be composite instruction set computing (CISC) microprocessors, reduced instruction set computing (RISC) microprocessors, very long instruction word (VLIW) microprocessors, or processors that implement other instruction sets, or processors that implement combinations of instruction sets. The processing devices may also be one or more special-purpose processing devices such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), or network processors. The system controller 228 may include data storage devices (e.g., one or more disk drivers and / or solid-state drivers), main memory, static memory, network interfaces, and / or other components. The system controller 228 can execute instructions to implement one or more of the techniques and / or embodiments described herein. In some embodiments, the system controller 228 can execute instructions to perform one or more operations in the manufacturing system 200 in accordance with a process policy. These instructions can be stored in a computer-readable storage medium, which may include main memory, static memory, secondary storage, and / or processing devices (during instruction execution).

[0044] In some embodiments, the system controller 228 can receive data from the substrate measurement system 126 based on measurements of the substrate processed by the process chambers 214, 216, and 218. The data received by the system controller 228 may include spectral data, reflected light measurement data, and / or other data for all or part of the substrate 202. The data received from the substrate measurement system 126 can be stored in the data store 250. The data store 250 may be included as a component within the system controller 228, or it may be a separate component from the system controller 228. In some embodiments, the data store 250 may be a part of the data store 150, or may include a part of the data store 150, as described with reference to Figure 1.

[0045] Figure 2B shows one embodiment of a substrate measurement system 251 that can be used to measure a processed substrate. The substrate measurement system 251 may be an integrated measurement and / or imaging system (e.g., an integrated reflected light measurement (IR) system) configured to measure film properties (e.g., thickness) across the surface of the substrate 264 after the substrate 264 has been processed in a processing chamber. Reflected light measurement is a measurement technique that uses measured changes in light reflected from an object to determine the geometric and / or material properties of the object. A reflectance spectrometer measures the intensity of reflected light over a range of wavelengths. In the case of dielectric films, these intensity variations can be used to determine the thickness of the film.

[0046] Process results, including film thickness, can be monitored across one or more substrates with respect to the etching and deposition processes, for example, using a substrate measurement system 251. While the substrate 264 is still within the device manufacturing system, its surface can be measured using an integrated measurement and / or imaging system. In some embodiments, the substrate measurement system 251 can correspond to the substrate measurement system 126. The substrate measurement system 251 can be connected to a factory interface or a transfer chamber. Alternatively, the substrate measurement system 251 can be located inside the factory interface or a transfer chamber. The substrate measurement system 251 can also be a standalone system not connected to a manufacturing system. The substrate measurement system 251 can be mechanically isolated from the factory interface and the external environment to protect it from external vibrations. In some embodiments, the substrate measurement system 251 and its components can provide analytical measurements (e.g., thickness measurements) that can provide a uniformity profile across the surface of the substrate 264, which is referred to herein as a profile map. A computing device can process its data from the substrate measurement system 251 and provide feedback to the user. The substrate measurement system 251 can be an assembly capable of measuring film thickness and / or other film properties such as optical constants, particle number, and roughness over a portion or the entire substrate after the substrate has been processed in the chamber. The measurement results can be used to determine when maintenance should be performed on the process chamber, when further testing should be performed on the substrate, when the substrate should be flagged as out of specification, etc.

[0047] When the substrate 264 is lowered onto the substrate support 256 (for example, a chuck) and fixed to the substrate support 256, the center of the substrate 264 can be offset from the center of the chuck. The processing device of the substrate measurement system 251 can determine one or more coordinate transformations between the center of the substrate 264 and the center of the chuck 256 (the center of the chuck corresponds to the axis of rotation on which the chuck rotates), and apply one or more coordinate transformations to correct the offset, as will be described in more detail below.

[0048] The substrate measurement system 251 may include a rotary actuator 252 and a linear actuator 254. The rotary actuator 252 may be a motor, a rotary actuator (e.g., an electric rotary actuator), etc. The linear actuator 254 may be an electric linear actuator that can convert rotational motion in a motor into linear or linear motion along an axis. The substrate measurement system 251 may also include a substrate support 256, a camera 208, a sensor 260, and a processing device 262.

[0049] The substrate support 256 can be a vacuum chuck, electrostatic chuck, magnetic chuck, mechanical chuck (e.g., a four-jaw chuck, a three-jaw chuck, an edge / ring clamp chuck, etc.), or other type of chuck. The substrate support 256 can hold a substrate 264 (e.g., a wafer). A rotary actuator 252 can rotate the substrate support 256 around a first axis 253. The rotary actuator 252 can be controlled by a servo controller and / or servo motor, which can enable precise control of the rotary actuator, and therefore the rotational position, speed, and / or acceleration of the substrate support 256. A linear actuator 254 can move the substrate support 256 linearly along a second axis 255. The linear actuator 254 can be controlled by a servo controller and / or servo motor 272, which can enable precise control of the linear actuator 254, and therefore the linear position, speed, and acceleration of the substrate support 256.

[0050] The camera 258 can be placed on the substrate support 256 and can generate one or more images of the substrate 264 held by the substrate support 256. The camera 258 can be an optical camera, an infrared camera, or another preferred type of camera. The sensor 260 can also be placed on the chuck 256 and can measure at least one target position on the substrate at once (for example, it can generate reflected light measurement or other measurements of the target position). The camera 258 and sensor 260 can be fixed in stationary positions on the substrate measurement system 251, and the chuck 256 can be moved in rθ motion by a rotary actuator 252 and a linear actuator 254.

[0051] In some embodiments, the processing device 262 can determine, based on one or more images of the substrate 264 generated by the camera 258, that the substrate 264 is not centered on the chuck 256. The substrate 264 may not be centered on the substrate support 256 when it is first placed on the substrate support 256. The robot blade 270 can place the substrate 264 onto a transfer station 268 (for example, a set of lift pins). The substrate support 256 can be moved in a first direction along a second axis 255 so that the chuck 256 is placed on the transfer station 268. The transfer station 268 can be located on a lift mechanism 266 (or a set of lift pins), and the lift mechanism 266 can move the transfer station 268 up and down vertically (perpendicular to the second axis 255 and parallel to the first axis 253). While the substrate support 256 is placed on the transfer station 268, the substrate support 256 can receive the substrate 264. The substrate 264 may not be centered on the substrate support 256. The substrate support 256 can be moved in a second direction along the second axis 255 until the sensor 260 detects that the edge of the substrate 264 is at the target position.

[0052] The substrate support can be rotated 360 degrees, and images can be generated while the substrate support is rotating. The chuck can be used with various different θ values ​​to obtain one or more measurements and / or images, and the location of edge detection can vary. The detected edge changes can indicate that the center of the substrate (which can be a circular substrate) is shifted. In addition, the determined changes in the detected edges can be used to calculate the amount of offset.

[0053] In one embodiment, the parameter (r,θ) determines the offset of the substrate relative to the stage. These parameters allow the motion system to perform forward and inverse transformations that convert the (r,θ) coordinates of the stage to the (r,θ) coordinates of the substrate. The motion system can then calculate the trajectory of the substrate in space while transmitting commands to move the motors mounted on the substrate support 256. In one embodiment, the motion system can calculate the trajectory in arbitrary space by running real-time control software connected (e.g., via an EtherCAT network) to motion drivers of linear and rotary actuators. The controller 262 can calculate the corrected trajectory and transmit the indicated position to the motion driver in real time (e.g., at a speed of 1 kHz).

[0054] In some embodiments, the rotation of the substrate support 256 by the rotary actuator 252 for measuring the target position causes an offset between the field of view of the sensor 260 and the target position on the substrate 264, because the substrate 264 is not centered on the chuck 256. In this case, the linear actuator 254 can correct the offset by moving the substrate support 256 linearly along a second axis. The sensor 260 can then measure the target position on the substrate 264. After measurements of all target points on the substrate have been taken, the processing device 262 can determine a uniformity profile across the surface of the substrate 264 based on the measurements.

[0055] In some embodiments, the processing device 262 can determine one or more coordinate transformations between the center of the substrate support 256 (corresponding to the first axis 253 on which the chuck 256 rotates) and the center of the substrate 264, which is applied during the rotation of the substrate support 256 to correct the offset.

[0056] Figure 2C is a schematic cross-sectional side view of a substrate measurement subsystem 282 according to an aspect of the present disclosure. The substrate measurement subsystem 282 can be configured to acquire measurements of one or more portions of a substrate, such as the substrate 202 in Figure 2A, before or after processing the substrate 202 in the processing chamber. In embodiments, the substrate measurement subsystem 282 can correspond to the substrate measurement system 126 in Figure 2A. The substrate measurement subsystem 282 can acquire measurements of a portion of the substrate 202 by generating data associated with that portion of the substrate 202. In some embodiments, the substrate measurement subsystem 282 can be configured to generate spectral data, positional data, and / or other characteristic data associated with the substrate 202.

[0057] The substrate measurement subsystem 282 can be configured to generate one or more types of data on a substrate, including spectral data, positional data, and substrate characteristic data. The substrate measurement subsystem 282 can generate data on a substrate in response to a request to acquire one or more measurements on a substrate before or after the substrate is processed in a manufacturing system. The substrate measurement subsystem 282 can include one or more components that facilitate the generation of data on a substrate. For example, the substrate measurement subsystem can include a spectral sensing component for sensing one or more spectra from a portion of the substrate to generate spectral data on the substrate. In some embodiments, the spectral sensing component can be an interchangeable component and can be configured based on the type of process performed in the manufacturing system or the target type of measurement to be acquired by the substrate measurement subsystem. For example, one or more components of the spectral sensing component can be swapped in the substrate measurement subsystem to enable the acquisition of reflected light measurement spectral data, ellipsometry spectral data, hyperspectral imaging data, and chemical imaging data (e.g., X-ray photoelectron spectroscopy (XPS), energy-dispersive X-ray spectroscopy (EDX), X-ray fluorescence (XRF), etc.).

[0058] The substrate measurement subsystem 282 may include a controller 283 configured to execute one or more instructions for generating data associated with a portion of the substrate 202. The substrate measurement subsystem 282 may include a substrate sensing component 284 configured to detect when the substrate 202 has been transferred to the substrate measurement subsystem 282. The substrate sensing component 284 may include any component configured to detect when the substrate 202 has been transferred to the substrate measurement subsystem 282. For example, the substrate sensing component 284 may include an optical sensing component that transmits a beam of light across the inlet to the substrate measurement subsystem 282. When the substrate 202 is placed within the substrate measurement subsystem 282, the substrate sensing component 284 can detect that the substrate 202 has been transferred to the substrate measurement subsystem 282 in response to the substrate 202 blocking the beam of light transmitted across the inlet to the substrate measurement subsystem 282. In response to detecting that the substrate 202 has been transferred to the substrate measurement subsystem 282, the substrate sensing component 284 can transmit an instruction to the controller 283 indicating that the substrate 202 has been transferred to the substrate measurement subsystem 282.

[0059] In some embodiments, the substrate sensing component 284 can be further configured to detect identification information associated with the substrate 202. In some embodiments, when the substrate 202 is transported to the substrate measurement subsystem 282, the substrate 202 can be embedded in a substrate carrier (not shown). The substrate carrier may include one or more registration features that enable the identification of the substrate 202. For example, the optical sensing component of the substrate sensing component 284 can detect that the substrate 202 embedded in the substrate carrier has blocked a light beam transmitted across the entry point to the substrate measurement subsystem 282. The optical sensing component may further detect one or more registration features contained on the substrate carrier. In response to the detection of one or more registration features, the optical sensing component may generate an optical signature associated with one or more registration features. The substrate sensing component 284 may transmit the optical signature generated by the optical sensing component to the controller 283, along with an indication that the substrate has been placed in the substrate measurement subsystem 282. In response to receiving the optical signature from the sensing component 214, the controller 283 can analyze the optical signature to determine the identification information associated with the substrate 202. Identification information associated with the substrate 202 may include an identifier for the substrate 202, an identifier for the process for the substrate 202 (e.g., batch number or process execution number), an identifier for a certain type of substrate 202 (e.g., wafer), and so on.

[0060] The substrate measurement subsystem 282 may include one or more components configured to determine the position and / or orientation of the substrate 202 within the substrate measurement subsystem 282. The position and / or orientation of the substrate 202 can be determined based on the identification of a reference location of the substrate 202. The reference location may be a portion of the substrate 202 that contains an identification feature associated with a particular part of the substrate 202. For example, the substrate 202 may have a reference tag embedded in the center of the substrate 202. In another example, the substrate 202 may have one or more structural features on the surface of the substrate 202 that are contained in the center of the substrate 202. The controller 283 can determine the identification features associated with a particular part of the substrate 202 based on the determined identification information for the substrate 202. For example, in response to the determination that the substrate 202 is a wafer, the controller 283 may determine one or more identification features that are generally contained in a portion of the wafer.

[0061] The controller 283 can identify reference points on the substrate 202 using one or more camera components 285 configured to capture image data of the substrate 202. The camera components 285 can generate image data of one or more portions of the substrate 202 and transmit the image data to the controller 283. The controller 283 can analyze the image data to identify identification features associated with reference points on the substrate 202. Based on the identified identification features of the substrate 202, the controller 283 can further determine the position and / or orientation of the substrate 202 as shown in the image data. Based on the identified identification features of the substrate 202, as well as the determined position and / or orientation of the substrate 202 as shown in the image data, the controller 283 can determine the position and / or orientation of the substrate 202.

[0062] In response to determining the position and / or orientation of the substrate 202, the controller 283 can generate position data associated with one or more portions of the substrate 202. In some embodiments, the position data may include one or more coordinates (e.g., Cartesian coordinates, polar coordinates, etc.), each coordinate associated with a portion of the substrate 202, and each coordinate is determined based on the distance from a reference point to the substrate 202. For example, in response to determining the position and / or orientation of the substrate 202, the controller 283 can generate first position data associated with the portion of the substrate 202 containing the reference point, the first position data including a Cartesian coordinate of (0,0). The controller 283 can generate second position data associated with a second portion of the substrate 202 relative to the reference point. For example, a portion of the substrate 202 located approximately 2 nanometers (nm) due east from the reference point may be assigned a Cartesian coordinate of (0,1). In another example, a portion of the substrate 202 located 5 nm due north from the reference point may be assigned a Cartesian coordinate of (1,0).

[0063] The controller 283 can determine one or more portions of the substrate 202 to measure based on position data determined for the substrate 202. In some embodiments, the controller 283 can receive one or more operations of process strategies associated with the substrate 202. In such embodiments, the controller 283 can further determine one or more portions of the substrate 202 to measure based on one or more operations of process strategies. For example, the controller 283 can receive an instruction that an etching process has been performed on the substrate 202 if several structural features have been etched onto the surface of the substrate 202. As a result, the controller 283 can determine one or more structural features to measure and the expected locations of such features in various parts of the substrate 202.

[0064] The substrate measurement subsystem 282 may include one or more measuring components for measuring the substrate 202. In some embodiments, the substrate measurement subsystem 282 may include one or more spectral sensing components 220 configured to generate spectral data for one or more portions of the substrate 202. As already discussed, the spectral data may correspond to the intensity of the detected energy wave (i.e., the strength or amount of energy) for each wavelength of the detected wave.

[0065] In one embodiment, the reflected energy wave received by the substrate measurement subsystem 282 may contain multiple wavelengths. Each reflected energy wave may be associated with a different portion of the substrate 202. In some embodiments, the intensity can be measured for each reflected energy wave received by the substrate measurement subsystem 282. Each intensity can be measured for each wavelength of the reflected energy wave received by the substrate measurement subsystem 282. The relationship between each intensity and each wavelength can form the basis for forming spectral data. In some embodiments, one or more wavelengths may be associated with intensity values ​​outside the expected intensity range. In such embodiments, intensity values ​​outside the expected range may indicate the presence of a defect in a portion of the substrate 202.

[0066] The measuring component for measuring the substrate 202 may also include a non-spectral sensing component configured to collect and generate non-spectral data. For example, the measuring component may include an eddy current sensor or a capacitive sensor. While some embodiments of this description may refer to collecting and using spectral data for the substrate 202, the embodiments of this description may also be applicable to non-spectral data collected for the substrate 202.

[0067] The spectral sensing component 287 can be configured to detect energy waves reflected from a portion of the substrate 202 and generate spectral data associated with the detected waves. The spectral sensing component 287 may include a wave generator 288 and a reflected wave receiver 224. In some embodiments, the wave generator 288 may be an optical wave generator configured to generate a ray toward a portion of the substrate 202. In such embodiments, the reflected wave receiver 289 can be configured to receive the ray reflected from that portion of the substrate 202. The wave generator 288 can be configured to generate an energy stream 289 (e.g., a ray) and transmit the energy stream 289 toward the portion of the substrate 202. The reflected energy wave 290 can be reflected from that portion of the substrate 202 and can be received by the reflected wave receiver 291. Figure 2C shows a single energy wave reflected from the surface of the substrate 202, but multiple energy waves can be reflected from the surface of the substrate 202 and can be received by the reflected wave receiver 291.

[0068] In response to the reflected wave receiver 291 receiving the reflected energy wave 290 from that portion of the substrate 202, the spectrum sensing component 287 can measure the wavelength of each wave contained in the reflected energy wave 289. The spectrum sensing component 287 can further measure the intensity of each measured wavelength. In response to measuring each wavelength and each wavelength intensity, the spectrum sensing component 287 can generate spectral data for that portion of the substrate 202. The spectrum sensing component 287 can transmit the generated spectral data to the controller 283. In response to receiving the generated spectral data, the controller 283 can generate a mapping between the received spectral data and the position data for the measured portion of the substrate 202.

[0069] The substrate measurement subsystem 282 can be configured to generate specific types of spectral data based on the type of measurement to be acquired by the substrate measurement subsystem 282. In some embodiments, the spectral sensing component 287 may be a first spectral sensing component configured to generate one type of spectral data. For example, the spectral sensing component 287 may be configured to generate reflected light measurement spectral data, ellipsometry spectral data, hyperspectral imaging data, chemical imaging data, thermal spectral data, or conduction spectral data. In such embodiments, the first spectral sensing component can be removed from the substrate measurement subsystem 282 and replaced with a second spectral sensing component configured to generate a different type of spectral data (e.g., reflected light measurement spectral data, ellipsometry spectral data, hyperspectral imaging data, or chemical imaging data).

[0070] The controller 283 can determine the type of data to be generated for the substrate 202 (i.e., spectral data, non-spectral data) based on the type of measurement to be acquired for one or more portions of the substrate 202. In some embodiments, the controller 283 can determine one or more types of measurement based on a notification received from the system controller 228 in Figure 2A. In other or similar embodiments, the controller 283 can determine one or more types of measurement based on a command to generate a measurement for a portion of the substrate 202. In response to determining one or more types of measurements to be acquired, the controller 283 can determine the type of data to be generated for the substrate 202. For example, the controller 283 can determine that spectral data should be generated for the substrate 202 and that a second spectral sensing component is the optimal sensing component for acquiring the determined type of measurement for one or more portions of the substrate 202. In response to determining that the second sensing component is the optimal sensing component, the controller 283 may transmit a notification to the system controller indicating that the first spectral sensing component should be replaced with the second spectral sensing component, and that the second spectral sensing component should be used to obtain one or more types of measurements for one or more portions of the substrate 202. The system controller 128 may transmit the notification to a client device connected to the manufacturing system, and the client device may provide the notification to a user of the manufacturing system (e.g., an operator) via a GUI.

[0071] In other or similar embodiments, the spectral sensing component 287 may be configured to generate multiple types of spectral data. In such embodiments, the controller 283 may, according to the above embodiments, cause the spectral sensing component 287 to generate specific types of spectral data based on the type of measurement to be acquired for one or more portions of the substrate 202. In response to determining the type of measurement to be acquired, the controller 283 may determine that a first type of spectral data should be generated by the spectral sensing component 287. Based on the determination that a first type of spectral data should be generated by the spectral sensing component 287, the controller 283 may cause the spectral sensing component 287 to generate the first type of spectral data for one or more portions of the substrate 202.

[0072] As described above, the controller 283 can determine one or more portions of the substrate 202 to be measured by the substrate measurement subsystem 282. In some embodiments, one or more measurement components, such as a spectral sensing component 287, may be stationary components within the substrate measurement subsystem 282. In such embodiments, the substrate measurement subsystem 282 may include one or more positioning components 295 configured to correct the position and / or orientation of the substrate 202 relative to the spectral sensing component 287. In some embodiments, the positioning component 295 may be configured to translate the substrate 202 relative to the spectral sensing component 287 along a first axis and / or a second axis. In other or similar embodiments, the positioning component 295 may be configured to rotate the substrate 202 relative to the spectral sensing component 287 around a third axis.

[0073] When the spectral sensing component 287 generates spectral data for one or more portions of the substrate 202, the positioning component 295 can correct the position and / or orientation of the substrate 202 according to one or more determined portions to be measured relative to the substrate 202. For example, before the spectral sensing component 287 generates spectral data for the substrate 202, the positioning component 295 can position the substrate 202 at Cartesian coordinates (0,0), and the spectral sensing component 287 can generate first spectral data for the substrate 202 at Cartesian coordinates (0,0). In response to the spectral sensing component 287 generating first spectral data for the substrate 202 at Cartesian coordinates (0,0), the positioning component 240 can translate the substrate 202 along a first axis, and thus the spectral sensing component 287 is configured to generate second spectral data for the substrate 202 at Cartesian coordinates (0,1). In response to the spectral sensing component 287 generating second spectral data for the substrate 202 in Cartesian coordinates (0,1), the controller 283 can rotate the substrate 202 along the second axis, so that the spectral sensing component 287 is configured to generate third spectral data for the substrate 202 in Cartesian coordinates (1,1). This process can be repeated multiple times until spectral data is generated for each determined portion of the substrate 202.

[0074] In some embodiments, the surface of the substrate 202 may include layers 297 of one or more materials. These layers 297 may include etching materials, photoresist materials, mask materials, deposition materials, and the like. In some embodiments, the layers 297 may include etching materials to be etched according to an etching process performed in a processing chamber. In such embodiments, spectral data can be collected for one or more portions of the unetched etching material of the layer 212 deposited on the substrate 202 by embodiments already disclosed. In other or similar embodiments, the layers 297 may include etching materials already etched by an etching process in a processing chamber. In such embodiments, one or more structural features (e.g., lines, columns, openings, etc.) can be etched into one or more layers 297 of the substrate 202. In such embodiments, spectral data can be collected for one or more structural features etched into one or more layers 297 of the substrate 202.

[0075] In some embodiments, the substrate measurement subsystem 282 may include one or more additional sensors configured to capture additional data on the substrate 202. For example, the substrate measurement subsystem 282 may include additional sensors configured to determine the thickness of the substrate 202, the thickness of a film deposited on the surface of the substrate 202, and so on. Each sensor may be configured to transmit the captured data to the controller 283.

[0076] In response to receiving at least one of spectral data, positional data, or characteristic data for the substrate 202, the controller 283 may transmit the received data to the system controller 128 for processing and analysis, according to the embodiments described herein.

[0077] In some embodiments, the substrate measurement subsystem 282 includes one or more image acquisition devices 299 connected to a controller 283, such as a camera (including, for example, a complementary metal-oxide-semiconductor (CMOS) sensor or a charge-coupled device (CCD) sensor). The image acquisition devices 299 can generate images (e.g., two-dimensional (2D) color images, infrared (IR) images, near-infrared images, etc.). In embodiments, these images can be processed together with spectral data generated by a spectral sensing component 287 by one or more trained machine learning models to make decisions about one or more chamber components.

[0078] Figure 3 shows an exemplary system architecture 300 for predicting or evaluating chamber component conditions of a processing chamber according to an aspect of the present disclosure. In some embodiments, the system architecture 300 may include, or be part of, one or more components of the computer architecture 100 and / or the manufacturing system 200. The system architecture 300 may include one or more components of the manufacturing equipment 122 (e.g., a substrate measurement system 126), a server machine 320, and a server machine 350.

[0079] As described above, the manufacturing equipment 122 can produce products by following a policy or by performing operations over a period of time. The manufacturing equipment 122 may include a process chamber 310 configured to perform a substrate process on a substrate according to a substrate process policy. In some embodiments, the policy may be a blanket wafer policy for depositing a film on a test substrate. In some embodiments, the process chamber 310 may be any of the process chambers 214, 218, and 218 described with respect to Figure 2. The manufacturing equipment 122 may also include a substrate measurement system 126 as described herein.

[0080] The manufacturing machine 122 can be coupled to a server machine 320. The server machine 320 may include a processing device 322 and / or a data store 332. In some embodiments, the processing device 322 may be configured to execute one or more instructions for performing operations on the manufacturing machine 122. For example, the processing device 322 may include or be part of the system controller 228 described with respect to Figure 2. In some embodiments, the data store 332 may include or be part of the data store 150 and / or data store 250.

[0081] The processing device 322 can be configured to receive data from one or more components of the manufacturing equipment 122 (i.e., via a network). For example, the processing device 322 can receive film thickness profile data (e.g., wafer map) 336 collected by the substrate measurement system 126 after the substrate has been processed in the process chamber. In another example, the processing device 322 can receive measurement data collected by other measuring instruments before and / or after the substrate process on that substrate. The measurement data may include measurement values ​​generated on the substrate by integrated measuring instruments. In some embodiments, the processing device 322 can store the received spectral data, film thickness profile data, and / or received measurement data in a data store 332.

[0082] The processing device 352 may include a chamber component condition engine 330. The chamber component condition engine 330 in the processing device 322 may be configured to determine the chamber component conditions of one or more chamber components in the processing chamber 310 that processed the measured substrate. The chamber component condition engine 330 can determine one or more chamber component condition metrics for one or more chamber components in the process chamber 310 that processed the substrate from the film thickness profile data 336. In some embodiments, the chamber component condition metric may include a set of values ​​(e.g., vectors, matrices, etc.) that indicate a correlation of a particular combination, correlation, pattern, and / or relationship of data present in the sensor data. For example, the chamber component condition metric may include a feature vector containing binary values ​​that indicate the presence or absence of a particular feature in the data.

[0083] Chamber component condition metrics can be compared to known patterns and / or combinations of chamber component metrics (e.g., target chamber component metrics). Target chamber component metrics can be associated with new or non-worn chamber components. In response to determining that a chamber component condition metric satisfies one or more chamber component condition thresholds (e.g., conditions for excessive wear of electrostatic chuck mesa, or seal band, or lift pin hole; conditions for missing mesa; conditions for recesses or valleys in the surface of the substrate support; conditions for the concentricity of circular elements on the surface of the substrate support; conditions for the flatness of the surface of the substrate support), the chamber component condition engine 330 can flag or schedule the process chamber for maintenance and / or provide an alert to the user. Chamber component condition thresholds can be combinations of specified values ​​indicated by the chamber component condition metrics.

[0084] In some embodiments, a curve or other numerical or non-numerical inference is generated over time based on one or more chamber component metrics of a healthy or typical chamber component. The curve or other numerical or non-numerical inference may represent, for example, the typical wear of seal bands, mesas, and / or lift pin locations of an electrostatic chuck or other substrate support over time. The current chamber component condition metric can be compared to the curve or other numerical or non-numerical inference. The number of hours the chamber component has been used to process substrates, and / or the number of substrates the chamber component has been used to process, may be known. Using this information and the current chamber component condition metric, the current chamber component metric can be plotted against the curve or other numerical or non-numerical inference (for example, to determine where the current chamber component metric falls into the curve or other numerical or non-numerical inference). If the current chamber component metric is within a threshold distance from the curve, it can be determined that the chamber component condition is as expected. If the current chamber component metric deviates significantly beyond the threshold distance from the curve or other numerical or non-numerical inference, it can be determined that the chamber component condition is not as expected (for example, the chamber component is wearing faster than normal with respect to one or more areas or components of the chamber component).

[0085] As shown in Figure 3, the processing device 322 may, in some embodiments, include a training set generator 324 and / or a training engine 326. In some embodiments, the training set generator 324 may correspond to the training set generator 172 described with respect to Figure 1, and / or the training engine 326 may correspond to the training engine 182. The training set generator 324 may be configured to generate a training set 340 for training one machine learning model 334 or a set of machine learning models 334. For example, the training set generator 324 may generate training inputs based on historical film thickness profile data 336. Film thickness profile data 336 (e.g., film profile maps) may be associated with labels that identify the state of one or more features or regions of one or more chamber components (e.g., the amount of wear or erosion of seal bands in one or more regions, the amount of wear or erosion to one or more mesas, etc.). In some embodiments, the training set generator 324 may generate training inputs by retrieving historical film thickness profile data 336 (e.g., film thickness profile data) from a data store 332. The training set generator 324 can generate a target output that indicates chamber component conditions (e.g., chamber component condition metrics) for a training input, based on historical film profile data 336. The training set generator 324 can include the generated training input and the generated target output in a training set 340. The training set 340 may further include, for each historical film profile, an indicator of the years of use of one or more components (e.g., expressed as the number of hours the component was used to process substrates, and / or the number of substrates processed using the chamber component). Further details regarding the generation of the training set 340 are provided with reference to Figure 5.

[0086] The training engine 326 can be configured to train, authenticate, and / or test its machine learning model 334 or multiple sets of machine learning models 334. The training engine 326 can provide a training set 340 to train the machine learning model 334 and store the trained machine learning model 334 in the data store 332. In some embodiments, the training engine 326 can authenticate the trained machine learning model 334 using an authentication set 342. The authentication set 342 may include membrane profile data 336 and associated chamber component condition metrics. The training set generator 324 and / or the training engine 326 can generate the authentication set 342 based on historical membrane profile data 336. In some embodiments, the authentication set 342 may include historical membrane profile data 336 that is different from the historical membrane profile data 336 included in the training set 340.

[0087] The training engine 326 can provide historical film profile data 336 as input to a trained machine learning model 334, and can extract chamber component condition metrics for chamber components of the processing chamber from one or more outputs of the trained model 334. The input may further include the years of use of one or more components and / or one or more images of substrates used to generate the historical film profile data 236. The training engine 326 can assign a performance score to the trained model 334 based on the accuracy of the chamber component condition metrics for chamber components of the processing chamber. The training engine 326 can select a trained model 334 to be used to evaluate the conditions of one or more chamber components of the process chamber based on film profile data for substrates processed by the process chamber.

[0088] As already discussed, the training set generator 324 and / or training engine 326 may, in some embodiments, be components of a processing device 322 located in server 320. In additional or alternative embodiments, the training set generator 324 and / or training engine 326 may also be components of a processing device 352 located in server 350. Server 350 may include, or be part of, a computing system separate from the manufacturing system 200. As previously stated, server 320 may, in some embodiments, include, or be part of the system controller 228 described with respect to Figure 2A. In such embodiments, server 350 may include, or be part of, a computing system connected to (i.e., via a network) the system controller 228 but separate from the system controller 228. For example, users of the manufacturing system 200 may be provided with access to data stored in one or more parts of the data store 332, or to one or more processes running on processing device 322. However, users of the manufacturing system 200 may not be given access to any of the data stored in one or more parts of the data store 354, or to any of the processes running on the processing device 352.

[0089] The processing device 352, like the processing device 322, can be configured to run a training set generator 324 and / or a training engine 326. In some embodiments, the server 350 can be coupled to the manufacturing equipment 122 and / or inline measuring instrument 130 via a network. Thus, the processing device 352 can acquire thickness profile data 336 and chamber part condition metrics corresponding to film profile data 336 to be used by the training set generator 324 and / or training engine 326 to generate training sets 340 and certification sets 342, according to the embodiment described with respect to the processing device 322. In other or similar embodiments, the server 350 is not coupled to the manufacturing equipment 122 and / or an external measuring instrument 130. Accordingly, the processing device 352 can acquire thickness profile data 336 from the processing device 322.

[0090] A training set generator 324 in the processing device 352 can generate a training set 340 according to the embodiments described above. A training engine 326 in the processing device 352 can train and / or authenticate a machine learning model 334 according to the embodiments described above. In some embodiments, the server 350 can be coupled to other manufacturing equipment and / or other server machines different from the manufacturing equipment 122 and / or server machine 320. The processing device 352 can acquire film profile data 336 from other manufacturing equipment and / or server machines according to the embodiments described herein. In some embodiments, the training set 340 and / or authentication set 342 can be generated based on film thickness profile data 336 acquired for substrates processed in the process chamber 310, as well as other film profile data acquired for other substrates processed in process chambers in other manufacturing systems.

[0091] In response to the training engine 326 selecting the trained model 334 to be used, the processing device 352 can transmit the trained model 334 to the processing device 322. The chamber component condition engine 330 can then use the trained model 334, as described above, to provide a chamber component condition evaluation.

[0092] Figure 4 shows a model training workflow 405 and a model application workflow 417 for chamber component condition evaluation according to one embodiment. The model training workflow 405 and the model application workflow 417 can be executed by processing logic performed by the processor of a computing device. One or more of these workflows 405, 417 can be executed, for example, by one or more machine learning models implemented on the processing device, and / or by other software and / or firmware running on the processing device.

[0093] The model training workflow 405 is for training one or more machine learning models (e.g., deep learning models) to determine one or more features and / or conditions of one or more regions of one or more chamber components. The model application workflow 417 is for applying one or more trained machine learning models to perform chamber component condition evaluation. Each of the profile maps 412 can be associated with one or more labels indicating, for example, the erosion state and / or damage of the seal band, the state of one or more mesas (e.g., the erosion state and / or presence of mesas), the dielectric thickness of regions of the substrate support, the state of the seal band around lift pin holes and / or surrounding areas (e.g., to identify lift pin hotspots), damage or deterioration of other components, etc.

[0094] Various machine learning outputs are described herein. Specific numbers and configurations of machine learning models are described and illustrated. However, it should be understood that the number and types of machine learning models used, as well as their configurations, can be modified to achieve the same or similar final results. Therefore, the configurations of machine learning models described and illustrated are merely examples and should not be interpreted as limiting.

[0095] In one embodiment, one or more machine learning models are trained to perform one or more chamber component condition estimation tasks. Each task may be performed by a separate machine learning model. Alternatively, a single machine learning model may perform each or part of the tasks. For example, a first machine learning model may be trained to detect a first condition of a first chamber component, and a second machine learning model may be trained to detect a second condition of the first chamber component. In another example, a first machine learning model may be trained to detect one or more conditions of a first chamber component, and a second machine learning model may be trained to detect one or more conditions of a second chamber component. Additionally or alternatively, different machine learning models may be trained to perform different combinations of tasks. In one example, one or more machine learning models may be trained, and the trained machine learning (ML) model is a single shared neural network having multiple shared layers and multiple higher-level distinct output layers, each outputting a different prediction, classification, identification, etc. For example, a first higher-level output layer can determine the chamber component conditions for a first type of chamber component (e.g., an electrostatic chuck), and a second higher-level output layer can determine the chamber component conditions for a second type of chamber component (e.g., a shower head).

[0096] One type of machine learning model that can be used to perform some or all of the tasks described above is an artificial neural network, such as a deep neural network. An artificial neural network generally consists of feature representation components that have classifier or regression layers that map features to a target output space. A convolutional neural network (CNN), for example, hosts multiple layers of convolutional filters. Pooling is performed to handle nonlinearity in the lower layers, and typically a multilayer perceptron is attached on top of it to map the upper layer features extracted by the convolutional layers for decision (e.g., classification output). Deep learning is a type of machine learning algorithm that uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. Deep neural networks can be trained in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. A deep neural network consists of a hierarchy of multiple layers, where different layers learn different levels of representation corresponding to different levels of abstraction. In deep learning, each level learns to transform its input data into a slightly more abstract and complex representation. In particular, the deep learning process can learn which features naturally and optimally fit into which level. The "deep" in "deep learning" refers to the number of layers through which data is transformed. More precisely, a deep learning system has a substantial contribution distribution path (CAP) depth. A CAP is a chain of transformations from input to output. A CAP describes the potentially causal connections between inputs and outputs. For a forward neural network, the CAP depth may be the network depth, which can be the number of hidden layers plus one. For a regressive neural network, where a signal can propagate through a single layer more than once, the CAP depth is potentially unlimited.

[0097] Training a neural network can be achieved using supervised learning methods, which involve feeding the network a training dataset consisting of labeled inputs, observing its output, defining the error (by measuring the difference between the output and the labeled value), and adjusting the network's mass across all its layers and nodes to minimize the error using techniques such as deep gradient descent and backpropagation. In many applications, repeating this process across many labeled inputs in the training dataset results in a network that can produce the correct output when presented with inputs different from those present in the training dataset.

[0098] For the model training workflow 405, a training dataset should be used to form a training dataset containing hundreds, thousands, tens of thousands, hundreds of thousands, or more profile maps 412 of a single substrate. The data may include, for example, uniformity profiles determined using a given number of measurements, each associated with a specific target location. This data can be processed to generate one or more training datasets 436 for training one or more machine learning models. The training data items in the training dataset 436 may include profile maps 412, labels indicating the chamber component conditions of the chamber components used to process the substrates measured to generate the profile maps, one or more images of the substrate, and / or years of use information for one or more chamber components used to process the substrate.

[0099] To perform training, the processing logic inputs the training dataset 436 into one or more untrained machine learning models. Before inputting the first input into the machine learning models, the machine learning models can be initialized. The processing logic trains the untrained machine learning models based on the training dataset to generate one or more trained machine learning models that perform the various operations described above. Training can be performed by inputting input data, such as one or more profile maps 412 (e.g., thickness profile map, spectral profile map, roughness profile map, particle number profile map, optical constant profile map, etc.), component images and / or years of use information, one at a time into the machine learning model.

[0100] A machine learning model processes this input to produce an output. An artificial neural network includes an input layer consisting of values ​​within the data points. The next layer is called the hidden layer, and each node in the hidden layer receives one or more of the input values. Each node contains parameters (e.g., mass) to apply to the input values. Thus, each node essentially inputs the input values ​​into a multivariate function (e.g., a nonlinear mathematical transformation) to produce an output value. The next layer can be another hidden layer or an output layer. In either case, the nodes in the next layer receive output values ​​from the nodes in the previous layer, each node applies mass to those values, and then produces its own output value. This can be done in each layer. The final layer is the output layer, and there is one node for each class, prediction, and / or output that the machine learning model can produce.

[0101] Accordingly, the output may include one or more predictions or inferences (e.g., estimates of chamber component conditions for one or more regions or features of one or more chamber components of a process chamber on which a measured substrate was processed). The processing logic can compare the output estimated chamber component conditions with the labeled chamber component conditions. The processing logic determines an error (i.e., classification error) based on the difference between the estimated chamber component conditions and the target chamber component conditions. Based on the error, the processing logic adjusts the mass of one or more nodes in the machine learning model. An error item or delta can be determined for each node in the artificial neural network. Based on this error, the artificial neural network adjusts one or more of its parameters (masses for one or more inputs to the node) for one or more of its nodes. Parameters can be updated by backpropagation, so that the nodes in the highest layer are updated first, followed by the nodes in the next layer, and so on. The artificial neural network consists of multiple layers of "neurons," each layer receiving input values ​​from neurons in the previous layer. The parameters for each neuron include the mass associated with the values ​​received from each of the neurons in the previous layer. Accordingly, adjusting the parameters may include adjusting the mass assigned to each of the neurons in one or more layers within the artificial neural network.

[0102] After the model parameters have been optimized, model authentication can be performed to determine whether the model has improved and to determine the current accuracy of the deep learning model. After one or more training iterations, the processing logic can determine whether the stopping criteria have been met. The stopping criteria can be a target accuracy level, a target number of processed images from the training dataset, a target change in parameters with respect to one or more previous data points, a combination of these, and / or other criteria. In one embodiment, the stopping criteria are met when at least a minimum number of data points have been processed and at least a threshold accuracy has been achieved. The threshold accuracy can be, for example, 70%, 40%, or 90% accuracy. In one embodiment, the stopping criteria are met when the accuracy of the machine learning model stops improving. If the stopping criteria are not met, further training is performed. If the stopping criteria are met, training can be completed. After the machine learning model has been trained, the model can be tested using a reserved portion of the training dataset. After one or more trained machine learning models 438 have been generated, they can be stored in the model storage 445 and added to the chamber component condition engine 330.

[0103] According to one embodiment, in the model application workflow 417, the input data 462 can be input to one or more chamber component condition determiners 467, each of which may include a trained neural network, a statistical model, a physics-based model, etc. Additionally or alternatively, one or more chamber component condition determiners 467 may apply an image processing algorithm to determine the chamber component conditions. The input data may include profile maps generated based on measurements of substrates processed using the process chamber (e.g., of polymer layers on a substrate measured using an integrated reflective spectroscopy device or substrate measurement system, or of laminates of patterned layers, etc.). The process chamber may include new and / or used components (e.g., new and / or used substrate supports, process kit rings, gas delivery plates, etc.). The input data may further include one or more images of the substrate and / or usage information for one or more chamber components of the process chamber used to process the substrate (e.g., usage hours, number of substrates processed).

[0104] Based on the input data 462, the chamber component condition determination unit 467 outputs one or more estimated chamber component conditions 469. The estimated chamber component conditions may include chamber component conditions for one or more chamber components (which may be new or used), such as an electrostatic chuck or other substrate support, a gas delivery plate, a shower head, another gas delivery device, or a process kit ring. Multiple chamber component conditions can be determined for a single chamber component. For example, mesa conditions, seal band conditions, and lift pin location conditions (e.g., lift pin locations for the locations where the substrate is placed in the process chamber) can all be determined for an electrostatic chuck or other substrate support.

[0105] Estimated mesa conditions may include estimates of the amount of erosion to one or more mesas of the substrate support, estimates of missing mesas (e.g., new or used substrate support), estimates of the mesa pattern (e.g., new or used substrate support), and estimates of the degree to which the estimated mesa pattern deviates from the target mesa pattern (e.g., new or used substrate support). Estimated seal band conditions may include estimates of erosion to one or more portions of the seal band of the substrate support, and estimates of the concentricity of the seal band. Estimated lift pin location conditions may include estimates of hot spots at one or more locations on the substrate support, estimates of lift pin locations, and estimates of the deviation of estimated lift pin locations from the target lift pin locations. Process kit rings may surround and / or be placed beneath the substrate support. Process kit ring condition estimates for process kit rings may include estimates of erosion to the process kit ring, estimates of defects within the process kit ring (e.g., the amount by which the process kit ring becomes non-circular), and estimates of missing process kit rings.

[0106] In some embodiments, the chamber component condition determination unit 467 includes one or more models that receive input data 462 (e.g., profile maps of substrates processed by a process chamber) and output chamber component conditions 469 and / or suggested adjustments to one or more chamber components of a process chamber that compensate for the current chamber component conditions 469. These models may be, or include, trained machine learning models (e.g., convolutional neural networks), tables relating profile map data to policies and / or instrument settings, or physics-based models (e.g., digital twins) relating profile map data to policies and / or instrument settings. These models may output suggested adjustments to one or more parameters of the policies used to process the substrates associated with the profile maps input to their models. Additionally or alternatively, these models may output suggested settings to one or more parameters of the policies used to process the substrates associated with the profile maps input to their models. Additionally or alternatively, these models may output suggested settings (e.g., chamber parameter settings) to one or more chamber components used to process the substrates associated with the profile maps input to their models. Additionally or alternatively, these models can output recommended settings (tool parameter settings) for the parameters of one or more tools, which may include chamber parameter settings and / or settings for tools located outside the process chamber. Such tools located outside the process chamber may be connected to multiple different process chambers. Examples of external tools include mitigation, gas supply, and vacuum pumps. Such settings may be tool parameter settings (e.g., instrument settings) that are not associated with any specific policy, such as the amount of power to deliver to the heating element to achieve a target temperature or valve settings to achieve a target flow rate.Examples of policy settings that can be adjusted or controlled based on the model's output include one or more plasma output settings, one or more temperature setpoints (e.g., for different sections of a multi-section heater or chuck), one or more process step durations, one or more gas delivery flow rates, pressure, and so on.

[0107] The action determination unit 472 can determine one or more actions 470 to take based on the chamber component conditions 469. In one embodiment, the action determination unit 472 compares the chamber component condition estimations to one or more chamber component condition thresholds or other criteria. If one or more of the chamber component condition estimations meet or exceed the chamber component condition thresholds or other criteria, the action determination unit 472 may determine that preventive maintenance is recommended and may output a recommendation or notification to schedule maintenance and / or replacement of one or more chamber components of the process chamber. Alternatively, the action determination unit 472 may output one or more design change recommendations to modify the chamber component (e.g., to mitigate lift pin hotspots). Alternatively or additionally, the action determination unit 472 may output one or more suggestions for policy adjustments, setting adjustments for one or more chamber components, etc.

[0108] In some embodiments, the action determination unit 472 automatically schedules process chamber maintenance based on whether the chamber component conditions 469 meet one or more maintenance criteria. For example, the chamber component conditions may include estimations of the amount of erosion or wear on the seal band of the substrate support, on mesa over one or more regions of the substrate support, and / or around the lift pin holes of the substrate support. The chamber component conditions may further include present or missing mesa. The chamber component conditions may further include the dielectric layer thickness of the substrate support (e.g., electrostatic chuck). The chamber component conditions may further include the bond layer delamination state. The amount of erosion or wear for each region or feature can be compared to an erosion threshold. If the erosion exceeds the erosion threshold, maintenance can be recommended or initiated. In addition, if one or more mesa are missing, pack replacement for the electrostatic chuck can be recommended or initiated. In addition, if it is determined that one or more regions of the electrostatic chuck have a dielectric layer smaller than the target dielectric layer thickness, pack replacement for the electrostatic chuck can be recommended or initiated.

[0109] In some embodiments, the action determination unit 472 may output one or more policy adjustments (e.g., suggestions for one or more policy adjustments) and / or adjustments to the settings or positions of one or more chamber components to compensate for the chamber component condition estimations. Such adjustments can be performed automatically by the action determination unit 472 in some embodiments. For example, if the edge seal band of the substrate support is eroded (e.g., the estimated seal band erosion value falls within a range associated with a particular amount of erosion), the chucking voltage can be increased (e.g., to an amount associated with that range) to maintain substrate chucking. In another example, if the mesa is worn (e.g., the estimated mesa erosion value falls within a range associated with a particular amount of erosion), the backside cooling gas pressure can be reduced (e.g., to an amount associated with that range) to compensate for the worn mesa. In yet another example, if process kit erosion is detected (e.g., the estimated process kit ring erosion value falls within a range associated with a particular amount of erosion), the height of the process kit ring can be adjusted (e.g., to an amount associated with that range) to compensate. This enables closed-loop control of the process policy based on the detected chamber component conditions.

[0110] In some embodiments, the action determination unit 472 determines one or more actions to take based on proposed policy setpoints or parameters, equipment setpoints, etc., output by the chamber component condition determination unit 467. Examples of actions that can be taken include adjusting the setpoints of chamber components, adjusting parameter values ​​for one or more steps of a process policy executed on the process chamber, etc. Accordingly, in embodiments, the chamber component condition engine 330 can perform closed-loop process control, forward process control, feedback process control, etc.

[0111] In one embodiment, the action determination device 472 includes a quality evaluator capable of performing one or more tests or checks to determine the quality of one or more chamber components. The quality evaluator can determine whether the chamber component conditions satisfy one or more lift pin criteria, one or more seal band criteria, one or more mesa criteria, one or more delamination criteria, one or more process kit ring criteria, etc. For example, the seal band criterion may include concentricity with respect to a substrate support threshold. If the detected concentricity with respect to the substrate support does not meet the concentricity threshold, the seal band criterion may not be met. In another example, the mesa criterion may include a mesa presence criterion. If the absence of any mesa is detected, the mesa presence criterion may not be met. In another example, the mesa pattern can be determined and compared with a target mesa pattern. If the determined mesa pattern deviates from the target mesa pattern by a threshold amount, the mesa pattern criterion may not be met. In another example, the presence of a process kit ring can be detected. If the absence of a process kit ring is determined, the process kit ring presence criterion may not be met. In another example, process kit ring, mesa, and / or seal band erosion values ​​can be determined and compared to process kit ring, mesa, and / or seal band erosion thresholds. If any of the detected erosion amounts exceed their respective thresholds, the erosion criteria may not be met. In yet another example, the amount of delamination from the substrate support (indicating the amount of delamination between layers of the substrate support) can be determined. If the amount of delamination exceeds the delamination threshold, the delamination criteria may not be met. If any of these quality criteria are not met, the action determination device 472 can take appropriate action.

[0112] In one embodiment, the action determination unit 472 determines a quality metric value for the process chamber and / or one or more chamber components of the process chamber. The quality metric value may be based on the conditions of one or more chamber components, whether one or more chamber components passed their respective tests, and / or the test scores of the chamber components. The quality metric value may indicate the quality of the process chamber and / or individual chamber components. In one embodiment, the action determination unit 472 may determine, based on the quality metric value and / or estimated chamber component conditions (e.g., estimated mesa conditions, estimated lift pin conditions, and estimated seal band conditions), whether a chamber component (e.g., a substrate support) causes a reduction in product quality. If the chamber component causes a reduction in product quality (e.g., an increased device failure rate), the processing logic may determine that maintenance should be performed to replace or repair the chamber component.

[0113] In one embodiment, the action determination unit 472 compares the chamber component condition with one or more previous chamber component conditions for the chamber component of the chamber. Based on such a comparison, the action determination unit 472 can determine a change and / or rate of change (e.g., the rate of erosion to one or more areas of the substrate support) of one or more chamber component conditions. Based on this change or rate of change, the action determination unit 472 can determine one or more actions to be taken. In one embodiment, based on the rate of change of the chamber component and the current state of the chamber component, the processing logic can determine when to perform maintenance on the chamber component.

[0114] In one embodiment, the action determination 472 compares the chamber component conditions to a curve. The curve can be a curve of the chamber component conditions over time for one or more chamber components of the process chamber. Based on where the chamber component conditions fall off the curve, the action determination 472 can determine one or more actions to take.

[0115] Figure 5 is a flowchart of Method 500 for generating a training dataset for training a machine learning model to perform chamber component condition evaluation, according to an aspect of the present disclosure. Method 500 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as those running on a general-purpose computer system or dedicated machine), firmware, or a combination of several of these. In one embodiment, Method 500 can be performed by a computer system such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of Method 500 can be performed by one or more other machines not shown. In some embodiments, one or more operations of Method 500 can be performed by a training set generator 324 of a server machine 320 or server machine 350, as described with respect to Figure 3.

[0116] In block 510, the processing logic initializes the training set T to an empty set (e.g., {}). In block 512, the processing logic acquires substrate surface data associated with the substrate processed in the process chamber of the manufacturing system (e.g., reflected light measurement data of the surface of a film on the substrate, such as a film thickness profile map or wafer map).

[0117] In block 514, the processing logic obtains chamber component condition information for the process chamber that processed the substrate. As previously mentioned, the chamber component condition information may include erosion information relating to one or more features of the substrate support, and information relating to the conditions of one or more features of the showerhead (e.g., gas delivery rate and / or amount to one or more areas of the showerhead).

[0118] In block 516, the processing logic generates a training input based on the sensor data acquired from the substrate in block 512. In some embodiments, the training input may include a normalized set of sensor data (for example, including surface reflectometer data as described herein).

[0119] In block 518, the processing logic can generate a target output based on the chamber component condition information obtained in block 514. The target output can correspond to the chamber component condition metric (data indicating the characteristic conditions for one or more chamber components) of the process chamber that processed the substrate.

[0120] In block 520, the processing logic generates input / output mappings. The input / output mappings point to training inputs containing or based on data for the board, and target outputs for those training inputs, where the target outputs identify chamber component conditions, and the training inputs are associated with (or mapped to) the target outputs. In block 522, the processing logic adds the input / output mappings to the training set T.

[0121] In block 524, the processing logic determines whether the training set T contains a sufficient amount of training data to train a machine learning model. Note that in some embodiments, it may be determined that the training set T is sufficient based solely on the number of input / output mappings in the training set, while in some other embodiments, it may be determined that the training set T is sufficient based on the number of input / output mappings, or, instead, on one or more other criteria (e.g., the degree of diversity of the training examples). In response to determining that the training set T contains a sufficient amount of training data to train a machine learning model, the processing logic provides the training set T for training the machine learning model. In response to determining that the training set does not contain a sufficient amount of training data to train a machine learning model, method 500 returns to block 512.

[0122] In block 526, the processing logic provides a training set T for training a machine learning model. In some embodiments, the training set T is provided to the training engine 326 of server machine 320 and / or server machine 350 to perform training. For a neural network, for example, input values ​​of a given input / output mapping (e.g., spectral data and / or chamber data for a previous substrate) are input to the neural network, and output values ​​of the input / output mapping are stored in the output nodes of the neural network. The connection mass in the neural network is then adjusted according to a learning algorithm (e.g., backpropagation), and the procedure is repeated for other input / output mappings in the training set T. After block 526, the machine learning model 190 can be used to provide chamber component conditions (e.g., chamber condition metrics) for the chamber components of the process chamber.

[0123] Figure 6 is a flowchart showing one embodiment of method 600 for training a machine learning model to estimate the conditions of one or more chamber components of a process chamber. Method 600 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as that which runs on a general-purpose computer system or dedicated machine), firmware, or a combination of several of these. In one embodiment, method 600 can be performed by a computer system such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of method 600 can be performed by one or more other machines not shown. In some embodiments, one or more operations of method 600 can be performed by a training engine 326 to a server machine 320 or server machine 350 as described with respect to Figure 3.

[0124] In block 602 of Method 600, the processing logic collects a training dataset which may include data from multiple film profile maps (e.g., thickness profile maps showing the thickness of polymer films at multiple locations on a substrate). Each data item in the training dataset may include one or more labels. Data items in the training dataset may include input-level (e.g., image-level) labels indicating the presence or absence of one or more chamber component conditions without identifying the coordinates associated with those conditions. For example, some data items may include labels for worn seal bands, some for worn mesa, and some for unfavorable chamber component conditions. Data items in the training dataset may also include other labels, such as coordinate-level (e.g., pixel-level) classifications of chamber conditions. In some embodiments, each data item includes a feature vector containing thickness values ​​for each of multiple substrate coordinates. In some embodiments, each data item includes a wafer map, which may be an image of the wafer (e.g., a heatmap of the wafer). Colors in the heatmap may indicate thickness or other parameter values. Alternatively, the actual thickness value can be used for each of many coordinates on the surface of the substrate (e.g., wafer). The label may be a mask containing coordinate-level labels, or may contain them, and each coordinate may be associated with one or more labels such as erosion value, pass or fail value, or classification of one or more specific chamber component conditions.

[0125] In block 604, data items from the training dataset are input to an untrained machine learning model. In block 606, the machine learning model is trained to generate a trained machine learning model that classifies or estimates one or more chamber part conditions for one or more chamber parts of a process chamber based on the training dataset. The machine learning model can also be trained to output one or more other types of predictions, coordinate-level classifications, decisions, etc.

[0126] In one embodiment, in block 610, training data items are input to the machine learning model. This input may include data from a film profile map (e.g., a wafer map) showing one or more film properties (e.g., thickness, optical constants, particle number, roughness, material properties, etc.) across the substrate. In embodiments, the data may be input as an image or feature vector. In block 612, the machine learning model processes the input to produce an output. This output may include one or more chamber component condition classifications (e.g., worn seal band, worn mesa, worn lift pin hole, worn showerhead gas delivery hole, etc.). This output may additionally or alternatively include one or more coordinate-level chamber component condition classifications. For example, this output may show wear or erosion values ​​for each coordinate of the substrate. In some embodiments, the output may show wear or erosion values ​​for one or more areas of the substrate support that supported the substrate during processing. For example, the output may show the amount of erosion or wear for each coordinate of the substrate support (e.g., an electrostatic chuck). The output may include the probability of each chamber component condition class existing for each chamber component condition class that the machine learning model is trained to identify.

[0127] In block 614, the processing logic compares the output probability and / or value of the chamber component condition to the label of one or more chamber component conditions associated with the input. In block 616, the processing logic determines the error based on the difference between the output and the label. In block 618, the processing logic adjusts the mass of one or more nodes in the machine learning model based on this error.

[0128] In block 620, the processing logic determines whether the stopping criteria are met. If the stopping criteria are not met, the method returns to block 610, and another training data item is input to the machine learning model. If the stopping criteria are met, the method proceeds to block 625, and the training of the machine learning model is completed.

[0129] In one embodiment, multiple different ML models are trained to identify different types of chamber component conditions and / or to identify chamber component conditions for different types of chamber components.

[0130] In one embodiment, one or more ML models are trained to be applied across multiple process chambers, which may be process chambers of the same type or model. The trained ML models can then be further adjusted for use for a specific case of process chambers. Further adjustments can be made by using additional training data items, including thickness profile maps of substrates processed by the process chamber. Such adjustments can account for chamber mismatches between chambers and / or specific hardware process kits of several process chambers. In addition, in some embodiments, further training is performed to adjust the ML models to process chambers after maintenance in the process chamber and / or one or more changes to the hardware of the process chamber.

[0131] Figure 7 is a flowchart of a method 700 for determining the chamber conditions of one or more chamber components using one or more trained machine learning models, according to an aspect of the present disclosure. The method 700 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as those running on a general-purpose computer system or dedicated machine), firmware, or a combination of several of these. In one embodiment, the method 700 can be performed by a computer system, such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of the method 700 can be performed by one or more other machines not shown. In some embodiments, one or more operations of the method 700 can be performed by a chamber component condition engine 330 of a server machine 320, as described with respect to Figure 3.

[0132] In block 710 of Method 700, the process chamber of the substrate processing system (also called the manufacturing system) processes the substrate. For example, the process chamber can perform a film deposition process to deposit a film on the substrate, or an etching process to partially remove a film on the substrate. In some embodiments, the film is a polymer. In other embodiments, the film is a ceramic (e.g., a metal oxide).

[0133] In block 712, one or more robots transfer the substrate from the process chamber to the substrate measurement system. If the substrate measurement system is connected to or contained within the transfer chamber, the transfer chamber robot can remove the substrate from the process chamber and insert it into the substrate measurement system. If the substrate measurement system is connected to or contained within the factory interface, the transfer chamber robot can remove the substrate from the process chamber and place it into a load lock. The factory interface robot can then remove the substrate from the load lock and insert it into the substrate measurement system. The substrate measurement system can be any of the aforementioned substrate measurement systems, such as an integrated reflected light measurement (IR) device.

[0134] In block 714, the substrate measurement system generates measurements at many locations on the surface of the substrate. Each location may have a unique set of coordinates. In block 716, the substrate measurement system and / or computing device may generate one or more profile maps of the film on the substrate based on the measurements of the substrate measurement system. The profile map may be an image, or may contain an image, where each pixel in the image corresponds to a coordinate on the substrate. Each pixel may have an intensity value corresponding to a measurement (e.g., thickness value) at the substrate coordinate associated with the pixel. The profile may be a feature vector, or may contain a feature vector, where each entry in the feature vector is associated with a substrate coordinate, and each entry may have a value representing a value of the film at the substrate coordinate.

[0135] In block 718, the computing device processes one or more profile maps (e.g., thickness profile map, particle map, optical constant map, roughness map, etc.) using one or more trained machine learning models. Each of the trained machine learning models outputs one or more chamber component condition estimates for one or more chamber components of the process chamber. The output may include image-level or component-level chamber component estimates. For example, the output may indicate that wear has been detected on the electrostatic chuck or that wear has been detected on the showerhead. The output may include classifications of one or more chamber component conditions. For example, the output may indicate that seal band erosion, mesa erosion, dielectric layer erosion, lift pinhole erosion, etc., has been detected. The output may provide values ​​for one or more chamber component conditions, such as the amount of seal band erosion, the amount of mesa erosion, the amount of lift pinhole erosion, etc. In some embodiments, the output includes coordinate-level or region-level classifications of chamber component conditions for a chamber component. For example, the output may show, for various coordinates on the substrate support, erosion value, bond erosion value, dielectric thickness, surface planarity of the substrate support, concentricity of elements on the substrate support surface, surface roughness of the substrate support, presence and / or size of valleys or grooves within the substrate support surface, and / or chamber condition classification identified for those coordinates. This allows the user to determine which areas on the chamber component are worn or problematic.

[0136] In one embodiment, instead of using a trained ML model, or in addition to it, one or more computer vision algorithms are used to process the profile map. In one embodiment, a gold standard or reference film profile map associated with a healthy process chamber component is compared to the generated profile map. Based on this comparison, differences between the reference film profile map and the generated film map can be determined. Different types of differences can be associated with different chamber component condition classifications. For example, the reference film profile map may include representations of circles or ridges representing mesas, and the absence of similar circles or ridges in the generated film profile map may indicate that the electrostatic chuck lacks mesas. This comparison can be a full-image comparison between the reference film profile map and the generated film profile map. Alternatively, this comparison can be performed between patches or regions of the generated film profile map and patches or regions of the reference film profile map (for example, in the case of feature mapping).

[0137] In one embodiment, in block 720, the processing logic of the computing device determines, based on the output of a machine learning model, the probability that a chamber component (or a number of chamber components) will cause a reduction in product quality, product discard, processing delay, etc. In one embodiment, in block 722, the processing logic estimates the failure time of one or more chamber components of the process chamber, at least partially based on the output of a machine learning model (for example, based on chamber component condition estimation).

[0138] In block 724, the processing logic determines one or more actions to take, at least in part, based on estimated chamber component conditions and / or the probability that a chamber component causes a reduction in product quality. These actions may include generating a notification, recommending maintenance, scheduling maintenance, recommending or scheduling replacement of one or more chamber components, or recommending or scheduling repairs for one or more chamber components. In one embodiment, the processing logic determines in block 726 whether maintenance should be performed on one or more chamber components of the process chamber. This determination may then be output to a notification. In one embodiment, the processing logic determines when to perform maintenance on one or more chamber components based on estimated failure times.

[0139] In one embodiment, the processing logic generates an image of the chamber component and outputs the image to a graphical user interface. The image shows various parts of the chamber component, and one or more regions can be labeled with a chamber component condition classification. For example, the image may include one or more visualizations showing worn or chipped mesas, areas of worn seal bands, areas of the electrostatic chuck surface with dielectric thickness less than a threshold thickness, etc. The generated image may show the estimated dielectric layer thickness and / or other properties across the surface of the electrostatic chuck, such as mesa and / or seal band thickness across the surface of the electrostatic chuck. If a hot spot is identified around the lift pin hole, such a hot spot can be shown in the image. Thus, a graphical display of the electrostatic chuck showing diagnostic information about the electrostatic chuck can be provided. From the image, the user can distinguish lift pin position, mesa position and / or thickness, seal band position and / or thickness, dielectric layer thickness, bond thickness, etc. The image can also identify cracks, heat distribution, etc.

[0140] Accordingly, in embodiments, Method 700 can be implemented to perform a health check and / or anomaly detection on one or more chamber components, such as an electrostatic chuck. Embodiments make it possible to identify subtle signs of a problem in one or more chamber components and provide steps to be taken to address the problem before it affects the product.

[0141] Figure 8 is a flowchart of a method 800 according to an aspect of the present disclosure for determining conditions for one or more chamber components using one or more models (e.g., one or more trained machine learning models). Method 800 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as those running on a general-purpose computer system or dedicated machine), firmware, or some combination thereof. In one embodiment, Method 800 can be performed by a computer system such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of Method 800 can be performed by one or more other machines not shown. In some embodiments, one or more operations of Method 800 can be performed by a chamber component condition engine 330 of a server machine 320 described with respect to Figure 3.

[0142] In block 810 of Method 800, the process chamber of a substrate processing system (also called a manufacturing system) processes a substrate according to a process policy (e.g., using an etching process or a deposition process). The process can result in a film (e.g., a polymer film) having specific film properties such as thickness, dielectric constant, and density. Depending on the process performed, one or more film properties may depend heavily on specific conditions to be evaluated. For example, the thickness for some deposition or etching processes may depend on temperature. Accordingly, areas of the substrate with thicker films may be associated with higher or lower temperatures during processing, depending on the process. Thus, based on the film thickness at different locations on the substrate, it is possible to infer the temperature of the substrate at those different locations during the process. In another example, the film thickness at a certain location may depend on the amount of plasma reaching the substrate at that location or the energy of the plasma. Thus, based on the film thickness at different locations, it is possible to infer the RF field affecting the plasma at those different locations. The process policy used to process the substrate may be a policy designed to minimize changes in parameters of interest (e.g., temperature, RF field intensity, plasma energy level, etc.) and cause the maximum change in film properties such as thickness.

[0143] In block 812, one or more robots can transfer a substrate from the process chamber to the substrate measurement system. If the substrate measurement system is connected to or contained within the transfer chamber, the transfer chamber robot can remove the substrate from the process chamber and insert it into the substrate measurement system. If the substrate measurement system is connected to or contained within the factory interface, the transfer chamber robot can remove the substrate from the process chamber and place it into a load lock. The factory interface robot can then remove the substrate from the load lock and insert it into the substrate measurement system. The substrate measurement system can be any of the aforementioned substrate measurement systems, such as an integrated reflected light measurement (IR) device.

[0144] In block 814, the substrate measurement system generates measurements at many locations on the surface of the substrate. Each location may have a unique set of coordinates. In block 816, the substrate measurement system and / or computing device may generate one or more profile maps of the film on the substrate based on the measurements of the substrate measurement system. The profile map may be an image, or may contain an image, where each pixel in the image corresponds to a coordinate on the substrate. Each pixel may have an intensity value corresponding to a measurement (e.g., thickness value) at the substrate coordinate associated with the pixel. The profile may be a feature vector, or may contain a feature vector, where each entry in the feature vector is associated with a substrate coordinate, and each entry may have a value representing a value of the film at the substrate coordinate. In one embodiment, each value of the film corresponds to the conditions under which a process is being performed on the measured substrate. For example, each film thickness measurement may correspond to the temperature of the substrate during processing. In one embodiment, a new profile map is generated using a profile map and information regarding the correlation between film property values ​​and processing conditions (e.g., a function relating film thickness to temperature), where each point in this profile map contains a value representing the processing conditions at that point (e.g., each value represents the temperature at that point during processing).

[0145] In one embodiment, the substrate measurement device further generates one or more images of the substrate. These images may include, for example, 2D or 3D color images, IR images, near-IR images, ultraviolet images, and the like.

[0146] In block 818, the computing device processes one or more profile maps (e.g., thickness profile map, particle map, optical constant map, roughness map, temperature map, etc.) using one or more trained machine learning models. In some embodiments, the input to the machine learning model may include data from one or more profile maps, as well as data from one or more images of the substrate. Each of the trained machine learning models outputs one or more chamber component condition estimates for one or more chamber components of the process chamber. The outputs may include image-level or component-level chamber component estimates.

[0147] In one embodiment, the output includes a profile map of the substrate conditions during processing (e.g., a temperature map of the substrate during processing). Alternatively, such a profile map of the substrate conditions during processing may already be generated in block 816 (may be input into a trained machine learning model) without the use of a machine learning user.

[0148] In some cases, the output of a trained machine learning model includes information about the conditions of one or more chamber components. In some embodiments, the information may include information about one or more components of the chamber. In embodiments, the output may include estimated mesa conditions, lift pin location conditions, seal band conditions, and / or delamination conditions for an electrostatic chuck or other substrate support. For example, the output may include indications of detected mesa and their estimated wear height and / or amount, indications of estimated wear height and / or amount in the seal band area, indications of estimated locations of lift pin hotspots (points that had a temperature exceeding a temperature threshold during processing), indications of estimated areas where the electrostatic chuck plate is delaminating and / or indications of the estimated amount of delamination in those areas.

[0149] In one embodiment, one or more profile maps are input to a trained machine learning model along with one or more images of the substrate. In some cases, the addition of images can improve the accuracy of the output of the trained machine learning model.

[0150] In block 820, the processing logic outputs a notification of one or more conditions of a chamber component (e.g., a substrate support). This may include outputting estimated mesa conditions (e.g., missing mesa, mesa with excessive wear exceeding a wear threshold, etc.) 824, estimated lift pin conditions (e.g., estimated lift pin hotspots, etc.) 826, estimated seal band conditions (e.g., locations on the seal band with excessive wear exceeding a wear threshold, etc.) 828, estimated delamination conditions (e.g., areas of delamination and / or the degree or level of delamination in those areas, etc.). In embodiments, this output may be provided via a graphical user interface (GUI). In some embodiments, the output may be a message sent to a device (e.g., the user's mobile device) via text message, email, etc.

[0151] Figure 9 is a flowchart of a method 900 for determining the chamber conditions of one or more chamber components based on measurements of a substrate processed by a process chamber over time, according to an aspect of the present disclosure. Method 900 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as that which runs on a general-purpose computer system or a dedicated machine), firmware, or a combination of several of these. In one embodiment, Method 900 can be performed by a computer system, such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of Method 900 can be performed by one or more other machines not shown. In some embodiments, one or more operations of Method 900 can be performed by a chamber component condition engine 330 of a server machine 320, as described with respect to Figure 3.

[0152] In block 910 of method 900, the process chamber of the substrate processing system first processes the first substrate according to the process policy.

[0153] In block 916, the substrate measurement system can generate measurements at many locations on the surface of the substrate and generate one or more first profile maps of the substrate from the multiple measurements. The first profile maps may be, or may include, a first thickness map, a first particle map, a first temperature map, a first dielectric constant map, a first optical constant map, a first roughness map, and so on.

[0154] In block 918, the computing device can process one or more first profile maps using one or more trained machine learning models. Each of the trained machine learning models outputs one or more first chamber component condition estimates for one or more chamber components of a process chamber. The first outputs may include image-level or component-level chamber component estimates. In one embodiment, the first outputs include a profile map of the conditions of the substrate being processed (e.g., a temperature map of the substrate being processed). Alternatively, such a profile map of the conditions of the substrate being processed may already be generated in block 916 without using machine learning (and may be input to a trained machine learning model).

[0155] In some cases, the first output of the trained machine learning model includes information about the conditions of one or more chamber components. In some embodiments, the information may include information about one or more components of the chamber. In embodiments, the output may include estimated mesa conditions, lift pin location conditions, seal band conditions, and / or delamination conditions for an electrostatic chuck or other substrate support.

[0156] In block 920, the process chamber of the substrate processing system processes the second substrate for the second time according to the process policy.

[0157] In block 924, the substrate measurement system can generate measurements at many locations on the surface of a second substrate, and from the multiple measurements, it can generate one or more second profile maps of the second substrate. The second profile maps may be, or may include, a second thickness map, a second particle map, a second temperature map, a second dielectric constant map, a second optical constant map, a second roughness map, and so on.

[0158] In block 928, the computing device can process one or more second profile maps using one or more trained machine learning models. Each trained machine learning model outputs one or more second chamber component condition estimates for one or more chamber components of a process chamber. The second outputs may include image-level or component-level chamber component estimates.

[0159] In one embodiment, the second output includes a profile map of the substrate conditions being processed (e.g., a temperature map of the substrate being processed). Alternatively, such a profile map of the substrate conditions being processed may already be generated in block 924 without using machine learning (it may be input into a trained machine learning model).

[0160] In some cases, the second output of the trained machine learning model includes information about the conditions of one or more chamber components. In some embodiments, the information may include information about one or more components of the chamber. In embodiments, the output may include estimated mesa conditions, lift pin location conditions, seal band conditions, and / or delamination conditions for the electrostatic chuck or other substrate support.

[0161] In block 930, the processing logic performs at least one of the following: a) comparing a second profile map with the first profile map, or b) comparing a second machine learning output with the first machine learning output. Based on this comparison, the processing logic can determine the change over time of one or more chamber components and / or the rate of change of one or more chamber components. In some embodiments, before processing the second substrate, several additional substrates are processed at other times, the substrates are measured, profile maps are generated for the additional substrates, and the profile maps are processed by a trained machine learning model to generate additional outputs. The second outputs and / or profile maps can be compared with the additional outputs and / or profile maps, as well as the first outputs and / or profile maps. This provides additional data points from which changes over time of erosion rate, flotation, wear rate, and / or other chamber component conditions can be determined. For example, the processing logic can determine whether the wear rate on one or more mesas is accelerating.

[0162] In block 932, the processing logic makes one or more decisions regarding the chamber components of the process chamber based on the comparison made in block 930. The decisions may include, for example, one or more of the following: the erosion rate of one or more chamber components (block 920), the change in the erosion rate of one or more chamber components (block 922), whether it is time to replace any of the chamber components (block 924), the remaining lifespan of one or more chamber components (block 926), the expected yield of devices coming from a second substrate (block 928), and / or future substrates to be processed using the process chamber.

[0163] In one example, the substrate support can be replaced when the thickness of the seal band reaches a certain threshold. Based on the comparison performed in block 932, the processing logic can determine how quickly the seal band on the substrate support is eroding. This erosion rate can be applied to the currently determined thickness of the seal band and extrapolated to the future service life of the substrate support or the time when the seal band thickness will fall below the threshold thickness. The process chamber can then schedule maintenance so that the substrate support is replaced at the time when it is predicted that the seal band thickness will fall below the threshold thickness or after a further determined service life of the substrate support.

[0164] Figure 10 is a flowchart of a method 1000 for determining the condition of one or more chamber components based on a comparison of data and curves, according to an aspect of the present disclosure. Method 1000 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as that which runs on a general-purpose computer system or a dedicated machine), firmware, or a combination of several of these. In one embodiment, Method 1000 can be performed by a computer system, such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of Method 900 can be performed by one or more other machines not shown. In some embodiments, one or more operations of Method 1000 can be performed by a chamber component condition engine 330 of a server machine 320, as described with respect to Figure 3.

[0165] In block 1010 of method 1000, the process chamber of the substrate processing system first processes the substrate according to the process policy. The processing logic can first determine the service life of the chamber components. Service life can be measured using exact service time (e.g., 100 days of active use or 1000 hours of use of the processed substrate), the number of processed substrates or lots (multiple groups of substrates processed together), or other metrics. Service life can be determined for a single chamber component of interest (e.g., a substrate support or electrostatic chuck) or for multiple chamber components (e.g., a substrate support and a process kit ring).

[0166] In block 1016, the substrate measurement system can generate measurements at many locations on the surface of the substrate and generate one or more first profile maps of the substrate from the multiple measurements. The first profile maps may be, or may include, a first thickness map, a first particle map, a first temperature map, a first dielectric constant map, a first optical constant map, a first roughness map, and so on.

[0167] In block 1018, the computing device can process one or more first profile maps using one or more trained machine learning models. Each of the trained machine learning models outputs one or more first chamber component condition estimates for one or more chamber components of a process chamber. The outputs may include image-level or component-level chamber component estimates. In one embodiment, the outputs include a profile map of the conditions of the substrate being processed (e.g., a temperature map of the substrate being processed). Alternatively, such a profile map of the conditions of the substrate being processed may already be generated in block 1016 (and input into a trained machine learning model) without using machine learning.

[0168] In some cases, the output of a trained machine learning model includes information about the conditions of one or more chamber components. In some embodiments, the information may include information about one or more components of the chamber. In embodiments, the output may include estimated mesa conditions, lift pin location conditions, seal band conditions, and / or delamination conditions for an electrostatic chuck or other substrate support. The output may also include, additionally or alternatively, estimated process kit ring conditions, showerhead conditions, etc.

[0169] In block 1020, the processing logic compares at least one of the profile map or the output of a trained machine learning model to a curve to determine where the data falls off the curve. The curve may be generated based on measurements and / or machine learning model outputs of substrates processed by an ideal or golden process chamber over time, or based on measurements and / or machine learning model outputs of substrates processed by multiple process chambers of the same type and / or performing the same policy over time. The curve can show the range of expected chamber component conditions for a chamber component for any given number of years of use of a chamber component of a process chamber of that type and / or performing the same policy. The curve may include a single value with upper and / or lower deviation thresholds (e.g., standard deviation) for each number of years of use.

[0170] In block 1024, based on the comparison, the processing logic can determine whether the data deviates from the curve beyond a deviation threshold (for example, outside the upper or lower deviation threshold). In block 1028, the processing logic can generate an alert in response to determining that the data deviates from the curve by a significant amount beyond the deviation threshold. In the embodiment, the processing logic can plot the current chamber component condition values ​​against the curve on the display to show the user how the chamber components compare to the curve.

[0171] In the embodiment, method 1000 can be performed periodically on a substrate processed by a process chamber, and the results can be plotted against a curve. This can show the user how the chamber component conditions are changing over time compared to how the chamber component conditions are expected to change over time. For example, if the process kit ring and / or electrostatic chuck on a particular side is eroding faster than expected, this can indicate that these components are misaligned.

[0172] Figure 11A is a flowchart of a method 1100 for performing a process chamber quality assessment according to an aspect of the present disclosure. Method 1100 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as that which runs on a general-purpose computer system or a dedicated machine), firmware, or a combination of several of these. In one embodiment, Method 1100 can be performed by a computer system such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of Method 1100 can be performed by one or more other machines not shown. In some embodiments, one or more operations of Method 1100 can be performed by a chamber component condition engine 330 of a server machine 320 described with respect to Figure 3.

[0173] In block 1110 of method 1100, the process chamber of the substrate processing system processes a first substrate according to a first process policy. The first process policy can be designed to maximize variations in the film properties to be measured in block 1116 based on variations in the first chamber component conditions (e.g., RF field, temperature, etc.) during processing.

[0174] In block 1116, the substrate measurement system can generate measurements at many locations on the surface of the substrate and generate one or more first profile maps of the substrate from the multiple measurements. The first profile maps may be, or may include, a first thickness map, a first particle map, a first temperature map, a first dielectric constant map, a first optical constant map, a first roughness map, and so on.

[0175] In block 1118, the computing device can process one or more first profile maps using one or more trained machine learning models. Each of the trained machine learning models outputs one or more first chamber component condition estimates for one or more chamber components of a process chamber. The first outputs may include image-level or component-level chamber component estimates. In one embodiment, the first outputs include information about the first conditions of one or more chamber components. In some embodiments, the information may include information about one or more components of the chamber component.

[0176] In block 1120, the process chamber of the substrate processing system may process a second substrate according to a second process strategy. The second process strategy can be designed to maximize variations in the film properties to be measured in block 1122 based on variations in the second chamber component conditions (e.g., RF field, temperature, etc.) during processing.

[0177] In block 1122, the substrate measurement system can generate measurements at many locations on the surface of a second substrate, and from the multiple measurements, it can generate one or more second profile maps of the second substrate. The second profile maps may be, or may include, a second thickness map, a second particle map, a second temperature map, a second dielectric constant map, a second optical constant map, a second roughness map, and so on.

[0178] In block 1124, the computing device can process one or more second profile maps using one or more additional trained machine learning models (which may differ from the machine learning models used in block 1118). Each of the additional trained machine learning models outputs one or more second chamber component condition estimates for one or more chamber components of a process chamber. The second outputs may include image-level or component-level chamber component estimates. In one example, a first policy used to process a first substrate can generate a temperature-sensitive film which can be used to determine lift pin hotspots, electrostatic chuck delamination, mesa erosion, missing mesa, and / or seal band erosion. A second policy used to process a second substrate can generate a film which can sense RF field fluctuations and / or plasma fluctuations which can be used to determine process kit ring placement and / or process kit ring erosion.

[0179] In block 1130, the processing logic performs a quality assessment of the process chamber based on a profile map, or the output of one or more trained machine learning models, and / or at least one of one or more additional trained machine learning models. Performing the quality assessment may include performing multiple tests on the chamber component condition estimation and / or comparing the chamber component estimation to multiple criteria. Different criteria may be associated with different chamber component condition tests.

[0180] In one embodiment, in block 1132, the processing logic determines whether one or more lift pin criteria are met. The lift pin criteria may include the detected temperature at a lift pin location below a temperature threshold. In one embodiment, in block 1134, the processing logic determines whether one or more mesa criteria are met. For example, the mesa criteria may not be met if a missing mesa is identified and / or a mesa with an erosion amount greater than the erosion threshold is detected and / or a threshold number of mes with an erosion amount greater than the erosion threshold (e.g., a thickness less than the thickness threshold) is detected. In one embodiment, in block 1136, the processing logic determines whether one or more seal band criteria are met. The seal band criteria may not be met if a chip or crack is detected within a seal band and / or a location on the seal band with an erosion amount greater than the erosion threshold (e.g., a thickness less than the thickness threshold) is detected. In one embodiment, in block 1138, the processing logic determines whether one or more delamination criteria are met. If an area of ​​delamination is detected between the substrate support or the plates or layers of the electrostatic chuck, the delamination criterion may not be met. In one embodiment, in block 1140, the processing logic determines whether one or more process kit ring criteria are met. If the process kit ring is not centered around the substrate support (for example, if the process kit ring is off-center by a threshold), or if a threshold erosion amount is detected with respect to the process kit ring, the process kit ring criterion may not be met.

[0181] In block 1150, the processing logic generates a quality evaluation report based on the results of the quality evaluation performed in block 1130. The quality evaluation report may indicate whether the process chamber passed the quality evaluation and / or whether any specific chamber component within the process chamber passed the quality evaluation. If any chamber component failed to meet the quality evaluation, the processing logic may output why the quality evaluation failed for that chamber component (for example, which criteria were not met) and / or to what extent the chamber component failed to meet the criteria.

[0182] Method 1100 can, in embodiments, be performed periodically and / or as part of a re-authorization procedure after a maintenance event. Re-authorization of the process chamber can be confirmed if all quality evaluation tests and / or criteria are successfully passed.

[0183] Figure 11B is a flowchart of a method 1150 for determining how chamber components are positioned within a process chamber, according to an aspect of the present disclosure. Method 1150 is performed by processing logic that may include hardware (circuits, dedicated logic, etc.), software (such as that which runs on a general-purpose computer system or a dedicated machine), firmware, or a combination of these. In one embodiment, Method 1150 may be performed by a computer system, such as the computer system architecture 100 of Figure 1. In other or similar embodiments, one or more operations of Method 1150 may be performed by one or more other machines not shown. In some embodiments, one or more operations of Method 1150 may be performed by a chamber component condition engine 330 of a server machine 320, as described with respect to Figure 3.

[0184] In block 1152 of method 1150, the process chamber of the substrate processing system processes the substrate within the process chamber according to a first process policy.

[0185] In block 1154, the substrate measurement system can generate measurements at many locations on the surface of the substrate and generate one or more first profile maps of the substrate from the multiple measurements. The first profile maps may be, or may include, a first thickness map, a first particle map, a first temperature map, a first dielectric constant map, a first optical constant map, a first roughness map, and so on.

[0186] In block 1156, the computing device can process one or more first profile maps using one or more trained machine learning models. Each of the trained machine learning models outputs one or more chamber component condition estimates for one or more chamber components of a process chamber. The outputs may include image-level or component-level chamber component estimates. In some embodiments, the outputs may include information about one or more components of the chamber component.

[0187] In block 1160, the processing logic determines one or more installation parameters of the chamber component based on the output. In some cases, the output of the machine learning model indicates the location or position of the installation (e.g., the offset of the process kit ring from the center of the electrostatic chuck around which the process kit ring is positioned). Other installation parameters that can be detected include, for example, the degree to which one or more mounting fixtures or bolts are tightened.

[0188] In block 1162, the processing logic can determine whether the chamber components were properly installed based on the determined installation parameters for the chamber components. For example, the process kit ring criterion may include a threshold offset from the center. If the process kit ring has an offset greater than the threshold offset, it can be determined that the process kit ring was not properly installed.

[0189] In block 1164, the processing logic can adjust policies to correspond to determined installation parameters for future process executions. For example, the RF field and / or temperature can be adjusted in one or more areas of the substrate support based on the offset of the process kit ring. Additionally or alternatively, the processing logic can output a notification to reinstall chamber components. In some embodiments, the processing logic automatically schedules maintenance for the process chamber. In some cases, the process chamber may already be offline for maintenance, and method 1150 can be performed as part of the authorization procedure. If chamber components were not installed correctly, further maintenance can be performed before reauthorizing the process chamber (e.g., by reinstalling chamber components).

[0190] Figure 12 is a profile map 1200 of a processed substrate according to an aspect of the present disclosure. The profile map 1200 is a heat map showing the temperature at different locations on the substrate during processing in a process chamber. The temperature may be based on the thickness of the film deposited or etched at different locations during processing. A key 1202 is provided to show how to interpret the profile map 1200. As shown, hot spots 1205 are provided at each lift pin location, and the hot spots had high temperatures during processing. Also shown, a hot spot 1210 was present around the center of the substrate. In addition, the right-side region 1215 of the substrate was warmer than other regions around the substrate, which may indicate, for example, delamination on the right side of the substrate support.

[0191] Figure 13 is a profile map 1300 of a portion of a processed substrate according to an aspect of the present disclosure. The profile map 1300 is a heat map showing the temperature of different locations on the substrate during processing in a process chamber. The temperature may be based on the thickness of the film deposited or etched at different locations during processing. Note that other profile maps may represent non-thermal properties such as RF field intensity, plasma intensity or quantity, film thickness, etc. A key 1302 is provided to show how to interpret the profile map 1300. As shown, the area around the substrate 1310 was considerably cooler than the rest of the substrate during processing. In addition, several localized spots 1305 with reduced temperatures are shown, each of which represents a mesa of the substrate support. Points where a mesa made contact with the substrate may have improved thermal contact and therefore can have a lower temperature due to cooling by the substrate support during processing. Based on such a profile map, processing logic can determine whether any mesa is missing (e.g., based on the identification that there are no localized low-temperature spots where a mesa is expected). Such information can be extracted by image processing algorithms or trained machine learning models, for example, to determine whether any mesa is missing from the electrostatic chuck.

[0192] Figure 14 shows a graphical representation of an exemplary form of computing device 1400, a machine capable of executing a set of instructions to cause the machine to perform one or more of the techniques discussed herein. In alternative embodiments, the machine may be connected to other machines in a local area network (LAN), intranet, extranet, or the internet (e.g., a network connection). The machine may operate as a server or client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), tablet computer, set-top box (STB), personal digital assistant (PDA), cellular telephone, web device, server, network router, switch or bridge, or any machine capable of executing (sequentially or otherwise) a set of instructions specifying the actions to be taken by that machine. Furthermore, although only a single machine is shown, the term “machine” should also be interpreted to include any group of machines (e.g., computers) that individually or collectively execute a set (or more sets) of instructions for carrying out any one or more of the techniques discussed herein. In embodiments, the computing device 1400 may correspond to one or more of the server machines 170, 180, prediction server 112, system controller 228, 320, or 350, as described herein.

[0193] An exemplary computing device 1400 includes a processing device 1402, main memory 1404 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM), e.g., synchronous DRAM (SDRAM)), static memory 1406 (e.g., flash memory, static random access memory (SRAM)), and secondary memory (e.g., data storage device 1428), which communicate with each other via bus 1408.

[0194] The processing device 1402 can represent one or more general-purpose processors, such as a microprocessor or a central processing unit. More specifically, the processing device 1402 can be a composite instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor that implements other instruction sets, or a processor that implements a combination of instruction sets. The processing device 1402 can also be one or more dedicated processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), or a network processor. The processing device 1402 can also be, or include, a system-on-a-chip (SoC), a programmable logic control unit (PLC), or other types of processing devices. The processing device 1402 is configured to execute processing logic for performing the operations and steps discussed herein.

[0195] The computing device 1400 may further include a network interface device 1422 for communicating with the network 1464. The computing device 1400 may also include a video display unit 1410 (e.g., a liquid crystal display (LCD) or cathode ray tube (CRT)), an alphanumeric input device 1412 (e.g., a keyboard), a cursor control device 1414 (e.g., a mouse), and a signal generation device 1420 (e.g., a speaker).

[0196] The data storage device 1428 may include a machine-readable storage medium (or more specifically, a non-temporary computer-readable storage medium) 1424 in which one or more sets of instructions 1426 are stored, performing one or more of the techniques or functions described herein. For example, the instructions 1426 may include instructions for the chamber component condition engine 330. A non-temporary storage medium refers to a storage medium other than a carrier. The instructions 1426 may also reside entirely or at least partially in the main memory 1404 and / or the processing device 1402 while they are being executed by the computer device 1400, the main memory 1404, and the processing device 1402, which also constitute the computer-readable storage medium.

[0197] In exemplary embodiments, the computer-readable storage medium 1424 is shown as a single medium, but the term “computer-readable storage medium” should be interpreted to include a single medium or multiple mediums that store one or more sets of instructions (e.g., a centralized or distributed database, and / or associated caches and servers). The term “computer-readable storage medium” should also be interpreted to include any medium capable of storing or encoding a set of instructions for machine execution that causes a machine to execute one or more of the techniques of the present disclosure. Accordingly, the term “computer-readable storage medium” should be interpreted to include, but not limited to, solid-state memory and optical and magnetic media.

[0198] The above description includes numerous specific details, such as examples of specific systems, components, and methods, in order to provide a good understanding of some embodiments of the present disclosure. However, it will be apparent to those skilled in the art that at least some embodiments of the present disclosure can be implemented without these specific details. In other cases, well-known components or methods are not described in detail or are presented in simple block diagram form to avoid unnecessarily obscuring the present disclosure. Thus, the specific details described are merely illustrative. Certain embodiments may vary from these exemplary details, yet are still intended to remain within the scope of the present disclosure.

[0199] Throughout this specification, any reference to “one embodiment” or “an embodiment” means that the specific features, structures, or characteristics described in relation to that embodiment are included in at least one embodiment. Thus, expressions “in one embodiment” or “in an embodiment” in various arrangements throughout this specification do not necessarily all refer to the same embodiment. In addition, the term “or” is intended to mean inclusive rather than exclusive. When the terms “about” or “approximately” are used herein, this is intended to mean that the nominal values ​​presented are accurate within ±10%.

[0200] Although the operations of the methods described herein are illustrated and described in a specific order, the order of the operations of each method can be changed, and thus certain operations can be performed in reverse order, and certain operations can be performed at least partially concurrently with other operations. In another embodiment, separate operations or subordinate operations can be commanded intermittently and / or alternately.

[0201] It will be understood that the above description is intended to be descriptive, not restrictive. Those skilled in the art will see many other embodiments upon reading and understanding the above description. Therefore, the scope of this disclosure should be determined with reference to the appended claims, along with the full scope of the equivalents given to such claims.

Claims

1. Processing a substrate in a process chamber of a substrate processing system, wherein the substrate has at least one of a film or feature after the processing. The process involves measuring multiple locations on the substrate using the substrate measurement system of the substrate processing system, Based on measurements of the plurality of locations on the substrate, a profile map is generated on the substrate of at least one of the films or features. Processing data from the profile map using a first trained machine learning model, wherein the first trained machine learning model outputs a first chamber component condition estimate for the first chamber component of the process chamber. To determine whether to perform maintenance on the first chamber component of the process chamber, at least partially based on the estimation of the first chamber component conditions. A method that includes this.

2. The method according to claim 1, wherein the first chamber component comprises a substrate support, and the estimation of the conditions of the first chamber component includes at least one of the estimated mesa erosion conditions of the substrate support, the estimated seal band erosion conditions of the substrate support, or the estimated lift pin location erosion conditions of the substrate support.

3. The estimated mesa erosion conditions include the estimation of the amount of erosion of one or more mesas of the substrate support, The estimated seal band erosion conditions include estimation of erosion to one or more portions of the seal band on the substrate support, The estimated erosion conditions at the lift pin locations include the estimation of erosion at each lift pin location of the substrate support. The method according to claim 2.

4. The method according to claim 1, wherein the first chamber component comprises a substrate support, and the estimation of the first chamber component conditions includes at least one of estimating recesses in the surface of the substrate support, estimating the roughness across the surface of the substrate support, estimating the flatness of the surface of the substrate support, or estimating the concentricity of one or more circular elements of the substrate support.

5. Further comprising determining at least one defect or deterioration of the first chamber component based on the estimation of the first chamber component condition, The method according to claim 1, wherein the determination of whether to perform maintenance on the first chamber component is based on at least one of the defects or deterioration of the first chamber component.

6. The method according to claim 1, wherein the first chamber component comprises a shower head, and the first chamber component condition estimation includes estimated gas delivery to each of a plurality of regions of the shower head.

7. Based on the estimation of the first chamber component conditions, the probability that the first chamber component causes a reduction in product quality is determined, Based on the probability that the first chamber component causes the reduction in product quality, it is determined whether to perform the maintenance on the first chamber component of the process chamber. The method according to claim 1, further comprising:

8. To estimate the failure time of the first chamber component based at least partially on the estimation of the first chamber component conditions, Based on the estimated downtime, it is determined when to perform the maintenance on the first chamber component of the process chamber. The method according to claim 1, further comprising:

9. Processing the data from the profile map using a second trained machine learning model, wherein the second trained machine learning model outputs a chamber component condition estimate for the second chamber component of the process chamber. To determine whether to perform maintenance on the second chamber component of the process chamber, at least partially based on the estimation of the chamber component conditions for the second chamber component, and The method according to claim 1, further comprising:

10. Training a first machine learning model to produce the first trained machine learning model, wherein the first machine learning model is trained using data from multiple process chambers that share a common process chamber type. The method according to claim 1, further comprising:

11. Before processing the data from the profile map, the first trained machine learning model is adjusted using additional data from the process chamber. The method according to claim 10, further comprising:

12. The method according to claim 1, wherein the substrate is a blanket wafer.

13. A process chamber for processing a substrate having at least one of a film or feature after processing, One or more robots for moving the substrate from the process chamber to the substrate measurement system, A substrate measurement system for measuring at least one of several locations on the substrate, which is either the film or the feature, and generating a profile map of at least one of the film or the feature based on the measurements of the multiple locations, The computing device comprises a computing device, and the computing device is The process involves processing the data from the profile map using a first trained machine learning model, wherein the first trained machine learning model outputs a first chamber component condition estimate for the first chamber component of the process chamber. This is for determining whether to perform maintenance on the first chamber component of the process chamber, at least in part, based on the estimation of the first chamber component conditions. PCB processing system.

14. The substrate processing system according to claim 13, wherein the first chamber component comprises a chuck, the profile map includes a thickness profile map, and the first chamber component condition estimation includes at least one of the estimated mesa erosion conditions of the chuck, the estimated seal band erosion conditions of the chuck, or the estimated lift pin location erosion conditions of the chuck.

15. The estimated mesa erosion conditions include the estimation of the amount of erosion of one or more mesas by the chuck, The estimated seal band erosion conditions include estimation of erosion of one or more portions of the seal band of the chuck, The estimated erosion conditions for the lift pin locations include the estimation of erosion at each lift pin location of the chuck. The substrate processing system according to claim 14.

16. The substrate processing system according to claim 13, wherein the first chamber component comprises a shower head, and the estimation of the first chamber component conditions includes estimated gas delivery to each of a plurality of regions of the shower head.

17. The computing device further, Based on the estimation of the first chamber component conditions, the probability that the first chamber component causes a reduction in product quality is determined, A substrate processing system according to claim 13, which is for determining whether to perform the maintenance on the first chamber component of the process chamber based on the probability that the first chamber component causes the reduction in product quality.

18. The computing device further, To estimate the failure time of the first chamber component based at least partially on the estimation of the first chamber component conditions, The substrate processing system according to claim 13, which determines when to perform the maintenance on the first chamber component of the process chamber based on the estimated failure time.

19. The computing device further, Processing the data from the profile map using a second trained machine learning model, wherein the second trained machine learning model outputs a chamber component condition estimate for the second chamber component of the process chamber. The substrate processing system according to claim 13, which determines whether to perform maintenance on the second chamber component of the process chamber, at least partially based on the estimation of chamber component conditions for the second chamber component.

20. A computer-readable medium containing instructions that cause a processing device to perform an operation, when executed by the processing device, wherein the operation is The receiving of a thickness profile map of a film on a substrate, wherein the substrate is processed in a process chamber of a substrate processing system, and the thickness profile map is generated by the substrate measurement system of the substrate processing system after the substrate has been processed in the process chamber based on measurements of multiple locations on the substrate. The process involves processing data from the thickness profile map using a first trained machine learning model, wherein the first trained machine learning model outputs a first chamber component condition estimate for the first chamber component of the process chamber. A computer-readable medium that includes determining whether to perform maintenance on the first chamber component of the process chamber, at least in part, based on the estimation of the first chamber component conditions.

21. Processing a first substrate in a process chamber of a substrate processing system according to a policy, wherein the first substrate has at least one of a film or feature after the processing. Using the substrate measurement system of the substrate processing system, generate a profile map of at least one of the films or features on the first substrate, Processing data from the profile map using a first model, wherein the first model outputs at least one of the estimated mesa conditions of the substrate support relative to the process chamber, the estimated lift pin location conditions of the substrate support, the estimated seal band conditions of the substrate support, or the estimated process kit ring conditions for the process kit ring toward the process chamber. As a result of the above processing, output is provided a notification of at least one of the estimated mesa conditions, estimated lift pin location conditions, estimated process kit ring conditions, or estimated seal band conditions of the substrate support. A method that includes this.

22. To determine whether to perform maintenance on the substrate support based at least partially on at least one of the estimated mesa conditions, the estimated lift pin location conditions, or the estimated seal band conditions. The method according to claim 21, further comprising:

23. The method according to claim 21, wherein the first model outputs the estimated mesa conditions, the estimated mesa conditions include at least one of the following: estimation of the amount of erosion of one or more mesas of the substrate support, estimation of missing mesas, estimation of the mesa pattern, or estimation of the degree to which the estimated mesa pattern deviates from the target mesa pattern.

24. The method according to claim 21, wherein the first model outputs the estimated seal band conditions, the estimated seal band conditions include at least one of the following: estimation of erosion of one or more portions of the seal band of the substrate support, or estimation of the concentricity of the seal band.

25. The method according to claim 21, wherein the first model outputs the estimated lift pin location conditions, the estimated lift pin location conditions include at least one of the following: estimation of hot spots at one or more locations on the substrate support, estimation of lift pin locations, or estimation of deviation of the estimated lift pin locations from target lift pin locations.

26. The first model outputs the estimated process kit ring conditions for the process kit ring, wherein the process kit ring is at least one of a) surrounding the substrate support, or b) beneath the substrate support, and the estimated process kit ring conditions include at least one of erosion estimation for the process kit ring, defect estimation within the process kit ring, or missing process kit ring estimation, and the method is The method according to claim 21, further comprising determining whether to replace the process kit ring based on the estimated process kit ring conditions.

27. Processing a second substrate in the process chamber, wherein the second substrate, after processing, contains at least one of a second film or a second feature. Using the substrate measurement system of the substrate processing system, a second profile map is generated of at least one of the second film or the second feature on the second substrate. Processing data from the second profile map using the first model, wherein the first model outputs at least one of the following: a second estimated mesa condition of the substrate support, a second estimated lift pin location condition of the substrate support, a second estimated seal band condition of the substrate support, or a second estimated process kit ring condition of the process kit ring. a) comparing the second estimated mesa condition with the estimated mesa condition; b) comparing the second estimated lift pin location condition with the estimated lift pin location condition; c) comparing the second estimated seal band condition with the estimated seal band condition; or d) comparing the second estimated process kit ring condition with the estimated process kit ring condition, further comprising at least one of these: The method according to claim 21.

28. Based on the results of the comparison, the erosion rate is determined for at least one of the following: a) one or more mesas of the substrate support, b) one or more lift pin locations of the substrate support, c) one or more seal band portions of the substrate support, or d) the process kit ring. The method according to claim 27, further comprising:

29. Based on the results of the above comparison, a) Whether it is time to replace the chamber components of the process chamber, b) The remaining lifespan of the chamber component, or c) Expected yield performance of the device produced from the second substrate. The method according to claim 27, further comprising determining at least one of the following.

30. The quality of the substrate support is determined based on at least one of the estimated mesa conditions, the estimated lift pin location conditions for the location where the substrate is placed in the process chamber, or the estimated seal band conditions. The method according to claim 21, further comprising:

31. Based on the estimated mesa conditions, determine whether one or more mesas are damaged, missing, or improperly positioned. Considering that one or more of the mesa are damaged, missing, or improperly positioned, it is determined that the substrate support has failed the quality test. The method according to claim 30, further comprising:

32. a) determining when to replace the process kit ring based on the estimated process kit ring conditions, or b) determining at least one of the placement locations of the process kit ring around the substrate support. The method according to claim 21, further comprising:

33. The method according to claim 21, wherein the substrate support comprises a first component, a second component, and a bond between the first component and the second component, and the first model or the second model further outputs a bond condition estimation for the bond indicating the degree of failure of the bond and whether delamination of the second component from the first component has occurred.

34. The method according to claim 21, wherein the profile map of the film includes a thickness profile map of the film.

35. Based on at least one of the estimated mesa conditions, estimated lift pin location conditions, or estimated seal band conditions of the substrate support, the probability that the substrate support will cause a reduction in product quality is determined. The determination of whether to perform maintenance on the substrate support based on the probability that the substrate support causes the reduction in product quality. The method according to claim 21, further comprising:

36. At least partially, the failure time of the substrate support is estimated based on at least one of the estimated mesa conditions, the estimated lift pin location conditions, the estimated seal band conditions, or the estimated process kit ring conditions. Based on the estimated failure time, determine when to perform maintenance on at least one of the substrate support or the process kit ring. The method according to claim 21, further comprising:

37. The first model is a first trained machine learning model, and the method is Training a first machine learning model to produce the first trained machine learning model, wherein the first machine learning model is trained using data from multiple process chambers that share a common process chamber type. The method according to claim 21, further comprising:

38. The first model is a first trained machine learning model, and the method is Before processing the data from the profile map, the first trained machine learning model is adjusted using additional data from the process chamber. The method according to claim 21, further comprising:

39. Using an imaging device, generate one or more images of the first substrate, Inputting the one or more images into the first model along with the profile map. The method according to claim 21, further comprising:

40. Determine one or more adjustments to at least one of the process strategy or tool parameter settings to compensate for at least one of the estimated mesa conditions, estimated lift pin location conditions, estimated seal band conditions, or estimated process kit ring conditions. The method according to claim 21, further comprising:

41. The method according to claim 21, wherein at least one of the substrate support or the process kit ring is a novel component that has not yet been used to process a product substrate.

42. A computer-readable medium containing instructions that cause a processing device to perform an operation, when executed by the processing device, wherein the operation is After the first substrate is processed in a process chamber, a profile map of at least one of the films or features on the first substrate is received, measured using a substrate measurement system. The process involves processing data from the profile map using a first trained machine learning model, wherein the first trained machine learning model outputs at least one of the following: estimated mesa conditions for the substrate support relative to the process chamber, estimated lift pin location conditions for the substrate support, estimated seal band conditions for the substrate support, or estimated process kit ring conditions for the process kit ring for the process chamber. As a result of the above processing, output is provided a notification of at least one of the estimated mesa conditions, estimated lift pin location conditions, estimated process kit ring conditions, or estimated seal band conditions of the substrate support. Computer-readable media, including [specific text / data].

43. The aforementioned operation, The computer-readable medium according to claim 42, further comprising determining, at least in part, whether to perform maintenance on the substrate support based on at least one of the estimated mesa conditions, the estimated lift pin location conditions, the estimated seal band conditions, or the estimated process kit ring conditions.

44. The aforementioned operation, The computer-readable medium according to claim 42, further comprising determining one or more adjustments to the process policy to compensate for at least one of the estimated mesa conditions, the estimated lift pin location conditions, the estimated seal band conditions, or the estimated process kit ring conditions.

45. A computer-readable medium containing instructions that cause a processing device to perform an operation, when executed by the processing device, wherein the operation is The process involves receiving a first profile map of at least one film or feature on a first substrate, wherein the first profile map is generated by a substrate measurement system after the first substrate has been processed by a process chamber according to a policy. The means of receiving a second profile map of at least one film or feature on a second substrate, wherein the second profile map is generated by the substrate measurement system after the second substrate has been processed a second time by the process chamber according to the policy. Comparing the second profile map with the first profile map, A computer-readable medium that, based on the results of the comparison, determines at least one of the following: a) estimated mesa conditions for one or more mesas of the substrate support in the process chamber; b) estimated lift pin location conditions for one or more lift pins of the substrate support; or c) estimated seal band conditions for the seal band of the substrate support.

Images