Methods, systems, and computer-readable media for substrate placement optimization

Abstract

A computer-readable medium includes instructions that, when executed by a processing device, cause the processing device to perform operations. The operations include processing a first substrate in a processing chamber while the first substrate is supported by a substrate support in a first placement position. The first substrate includes a first surface profile after processing. The operations also include generating a first surface profile map of the first surface profile using a substrate measurement system. The operations also include determining a plurality of etch rates corresponding to a plurality of locations on the first substrate. The operations also include processing data associated with the plurality of etch rates using a model that outputs one or more estimated surface profiles associated with one or more estimated placement positions. The operations also include determining a suggested placement of the substrate on the substrate support based on the one or more estimated placement positions.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure generally relate to optimizing placement of substrates in a processing chamber, and in particular to generating a map and / or a numerical profile of a substrate processed using a chamber and optimizing placement of substrates in a processing chamber based on the map and / or the numerical profile of the substrate. BACKGROUND

[0002] Substrate processing can include a series of processes to create electronic circuits in semiconductors according to a circuit design. These processes can be performed in a series of processing chambers. Successful operation of a modern semiconductor manufacturing facility can aim to facilitate the stable movement of wafers from one chamber to another in the process of forming circuits in wafers to form products. During the execution of many substrate processes, the conditions of the processing chambers can change and can cause processed substrates to not meet target conditions and results.

[0003] Substrates are placed in a processing chamber for processing. The placement of the substrates relative to chamber elements can cause variations in product quality produced using the processing chamber and / or substrates processed using the processing chamber to be scrapped. SUMMARY

[0004] The following is a simplified summary of the disclosure to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is not intended to identify key or critical elements of the disclosure or to delineate any scope of the particular implementations of the disclosure or any scope of the claims. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.

[0005] In an exemplary embodiment, a computer-readable medium includes instructions that, when executed by a processing device, cause the processing device to perform operations. The operations include causing a first substrate to be processed in a processing chamber of a substrate processing system while the first substrate is supported on a substrate support in a first placement position by the substrate support. The first substrate includes a first surface profile after processing. The operations also include generating a first surface profile map of the first surface profile using a substrate measurement system. The operations also include determining a plurality of first etch rates corresponding to a plurality of first locations on the first substrate based on the first surface profile map. The operations also include processing data associated with the plurality of first etch rates using a model. The model outputs one or more estimated surface profiles associated with one or more estimated placement positions on the substrate support based on the plurality of first etch rates. The operations also include determining a suggested placement of the substrate on the substrate support based on the one or more estimated placement positions.

[0006] In an exemplary embodiment, a system includes a processing chamber including a substrate support. The system also includes a substrate measurement tool, a memory, and a processing device operably coupled to the memory. The processing device is to cause a first substrate to be processed in the processing chamber while the first substrate is supported by the substrate support in a first placement position on the substrate support. The first substrate includes a first surface profile after processing. The processing device is also to generate a first surface profile map of the first surface profile using the substrate measurement tool. The processing device is also to determine a plurality of first etch rates corresponding to a plurality of first locations on the first substrate based on the first surface profile map. The processing device is also to process data associated with the plurality of first etch rates using a model. The model outputs one or more estimated surface profiles associated with one or more estimated placement positions on the substrate support based on the plurality of first etch rates. The processing device is also to determine a suggested placement of the substrate on the substrate support based on the one or more estimated placement positions.

[0007] In an exemplary embodiment, a method includes processing a first substrate in a processing chamber while the first substrate is supported by a substrate support in a first placement position on the substrate support. The first substrate includes a first surface profile after processing. The method also includes generating a first surface profile map of the first surface profile using a substrate measurement system. The method also includes determining a plurality of first etch rates corresponding to a plurality of first locations on the first substrate based on the first surface profile map. The method also includes processing data associated with the plurality of first etch rates using a model. The model outputs one or more estimated surface profiles associated with one or more estimated placement positions on the substrate support based on the plurality of first etch rates. The method also includes determining a suggested placement of the substrate on the substrate support based on the one or more estimated placement positions. BRIEF DESCRIPTION OF DRAWINGS

[0008] The disclosure is illustrated by way of example and not by way of limitation in the figures of the accompanying drawings in which like references indicate similar elements. It should be noted that different "exemplary embodiments" described herein can each be implemented differently and need not have all of the features needed to implement other such embodiments. These embodiments are depicted in terms of sequences of actions or events but it should be noted that the sequences depicted are exemplary and need not be performed in the order shown and / or described. Individual sequences can also be performed concurrently. Not all process steps are required, but all illustrated features can be included in some embodiments. In addition, not all of the features of the components need to be included in some embodiments. Embodiments described herein are not mutually exclusive.

[0009] Figure 1 An illustrative computer system architecture in accordance with aspects of the present disclosure is depicted.

[0010] Figure 2A is a top view schematic of an example manufacturing system in accordance with aspects of the present disclosure.

[0011] Figure 2B is a cross-sectional schematic side view of a substrate measurement system included in Figure 2A an example manufacturing system in accordance with aspects of the present disclosure.

[0012] Figure 2Cis a cross-sectional schematic side view of a substrate measurement subsystem according to an aspect of the present disclosure.

[0013] Figure 3 depicts an illustrative system architecture for substrate placement prediction for a processing chamber according to an aspect of the present disclosure.

[0014] Figure 4 shows a model training workflow and a model application workflow for substrate placement determination according to one embodiment.

[0015] Figure 5 is a flowchart of a method for generating a training dataset for training a machine learning model according to an aspect of the present disclosure.

[0016] Figure 6 shows a flowchart of a method for training a machine learning model to determine a substrate placement according to one embodiment.

[0017] Figure 7 is a flowchart of a method of determining a suggested substrate placement according to an aspect of the present disclosure.

[0018] Figure 8 is a flowchart of a method of comparing a second estimated substrate placement to a first estimated substrate placement according to an aspect of the present disclosure.

[0019] Figure 9 is a profile plot (e.g., heat map) of a processed substrate according to an aspect of the present disclosure.

[0020] Figures 10A-10B is a flowchart of a method of determining an optimal substrate placement according to an aspect of the present disclosure.

[0021] Figure 11A is an example plot of an experimental substrate placement design according to an aspect of the present disclosure.

[0022] Figure 11B is an example radial plot of a substrate etch rate versus theta according to an aspect of the present disclosure.

[0023] Figure 12A is an example plot of a substrate etch rate versus theta according to an aspect of the present disclosure.

[0024] Figure 12B is an example plot of a normalized substrate etch rate versus theta according to an aspect of the present disclosure.

[0025] Figure 13A is an example plot of a normalized substrate etch rate of an experimental substrate placement design according to an aspect of the present disclosure.

[0026] Figure 13Bis an example plot of a substrate design for experiments according to aspects of the present disclosure.

[0027] Figure 13C is an example plot of a linear fit of substrate etch rates according to aspects of the present disclosure.

[0028] Figure 14 is a flowchart of a method of determining optimal substrate placement according to aspects of the present disclosure.

[0029] Figure 15 depicts an illustrative diagrammatic representation of the machine in the example form of a computing device in which instructions for causing the machine to perform any one or more of the methodologies discussed herein can be executed. DETAILED DESCRIPTION

[0030] Embodiments of the present disclosure relate to systems and methods for optimizing substrate placement in a processing chamber using substrate measurements. Processing results of a manufacturing process depend on many factors, including processing recipe, chamber parameter settings, chamber element conditions, and substrate placement within the processing chamber. For example, based on placement of the substrate relative to a processing kit ring, processing results can vary across the surface of the substrate. The gap between the substrate edge and the processing kit ring is generally related to the impact of the substrate processing adjacent to the substrate edge. Additionally, based on the condition of the showerhead, the condition of the substrate support supporting the substrate, the condition of the chamber liner, the condition of the pumps and / or valves, etc., processing results can vary across the surface of the entire substrate. Optimal placement of the substrate can be affected by the condition of the processing chamber elements described herein. For example, any changes to the substrate support, such as an electrostatic chuck, a clamp, a vacuum chuck, a heater, a support including a pocket with a lip on the edge of the support, and / or a substrate support including one or more embedded features (e.g., heaters, cooling plates, electronic components, etc.), etc., over time will slowly affect the substrate results, such as temperature variations through the processed substrate surface, radio frequency (RF) fields at the edge of the substrate, etc. These effects on the substrate results will slowly affect the optimal substrate placement for optimal processing results. Thus, over time, the optimal placement of the substrate on the substrate support will tend to change.

[0031] Substrate results can be particularly affected by substrate placement near the edge of a processed substrate. For example, factors affecting substrate etch rate (e.g., temperature, gas flow, etc.) can cause a tilt near the edge of a processed substrate. "Tilt" refers to a tendency of a substrate feature (e.g., a valley, a wall, a pillar, a mesa, etc.) to not be normal to the substrate surface. Excessive tilt can result in a poor quality substrate and / or a substrate being scrapped. Tilt often affects substrates near the edge of the substrate and particularly near the very edge of the substrate (e.g., within five millimeters of the edge of the substrate). Variations in substrate placement relative to elements of a substrate support (e.g., a process kit ring, etc.) can affect processing of the substrate. For example, variations in the gap between the edge of the substrate and a process kit ring (e.g., the gap around a wafer) can cause a tilt that exceeds a threshold amount of allowed tilt, resulting in a substrate being scrapped.

[0032] Embodiments described herein provide mechanisms for determining an optimized placement of a substrate for processing in a processing chamber. Some embodiments can be used to determine a suggested position of a substrate on a substrate support for processing in a processing chamber. The suggested position can be used to determine an offset of a substrate processing robot (e.g., a transfer chamber robot, etc.) to place the substrate in the processing chamber.

[0033] In some embodiments, a substrate is processed in a processing chamber according to a recipe. The substrate can be processed to deposit or etch a film layer and / or one or more features (e.g., measurable features) on a surface of the substrate. The substrate can be a bare substrate or a test substrate that does not include a product. Alternatively, the substrate can be a product substrate. The features can include features distributed on the surface of the substrate, such as mesas, dots, structures, valleys, walls, lines, trenches, grooves, fiducials, etc. In some examples, during processing, the substrate can be supported by a substrate support (e.g., an electrostatic chuck, etc.) of the processing chamber. In some examples, after processing, the substrate can have a surface profile (e.g., a thickness profile, a tilt profile, etc.). In examples of an etch processing recipe, the surface profile can be indicative of an etch rate across the surface of the substrate during the etch processing (e.g., an etch rate profile). In some examples, the etch rate can be related to tilt, particularly at locations on the substrate surface near the edge of the substrate.

[0034] In some embodiments, after depositing or etching a film and / or features on the substrate, the substrate can be removed from the processing chamber and input into a substrate measurement system. In some embodiments, a profile map of the substrate is generated by the substrate measurement system based on the surface profile of the substrate. The substrate measurement system can be, for example, a reflectometry measurement system or other measurement system that measures film thickness of a film and / or features at multiple locations of the substrate. The thickness information can be used to generate a profile map of the substrate. Alternatively, one or more other profile maps of the substrate (e.g., optical constants or roughness, particle counts, etc.) can be generated from other measurement data.

[0035] The film and / or feature thickness information (e.g., thickness profile maps) or other film and / or feature information (e.g., other profile maps such as optical constant profile maps, particle count profile maps, etc.) can then be processed using a model (e.g., a trained machine learning model, a physics-based model, a statistical model, and / or an image processor, etc.) to identify changes in one or more properties of the film. The model can output an estimated substrate placement value for placing a substrate relative to one or more elements in a processing chamber (e.g., an electrostatic chuck, a process kit ring, etc.). In embodiments, the model can output a suggested placement position for a substrate, can output one or more predicted film properties associated with the suggested placement position (e.g., a predicted profile map for a substrate processed from the suggested placement position), etc.

[0036] In some embodiments, etch rates for a plurality of locations on a substrate surface are determined from the feature thickness information. In some embodiments, data associated with the etch rates can be processed using a model to determine an estimated placement position for a substrate on a substrate support for processing. For example, normalized etch rates for a plurality of test substrates (each test substrate processed at a different substrate placement) can be processed by a numerical model at specified locations on each substrate to determine the impact of different substrate placements. The model can output an estimated substrate placement and / or an estimated substrate placement value relative to one or more elements in a processing chamber.

[0037] In some embodiments, a suggested placement for a substrate on a substrate support is determined based on the estimated substrate placement value. In some examples, the suggested placement is an optimized placement position on a substrate support (e.g., an electrostatic chuck, etc.) for processing a substrate to meet a target substrate specification (e.g., a tilt near an edge of the substrate is less than a threshold amount, etc.). In some examples, the optimized position for a substrate on a substrate support results in a uniform gap around the substrate between an edge of the substrate and an inner diameter of a process kit ring. However, in some examples, the optimized position for a substrate does not provide a uniform gap. In such examples, the optimized position for a substrate takes into account processing chamber variables, such as a condition of one or more chamber elements (e.g., a process kit ring condition, a showerhead condition, an electrostatic chuck condition, etc.). In some embodiments, robot settings are determined to place a substrate at the suggested placement position, and those settings are used to place the substrate onto a substrate support.

[0038] Accordingly, embodiments described herein add new detection capabilities to processing chambers without adding cost to those processing chambers. These new detection capabilities can be utilized in some embodiments to improve the quality and / or quantity of processed substrates that meet threshold specifications (e.g., threshold tilt specifications of processed substrates, etc.). Additionally or alternatively, embodiments described herein can be used to reduce the amount of scrap product (e.g., scrap substrates) due to scrapping product that does not meet threshold specifications. Embodiments described herein can be used to automatically optimize certain processing variables (e.g., substrate placement during processing) to quickly and more efficiently process higher quality substrates when compared to conventional substrate processing systems. In particular, embodiments described herein can be used to produce substrates with better edge properties (e.g., tilt, etc.) when compared to substrates produced by conventional systems. This in turn can lead to higher quality substrates, less scrap product, etc.

[0039] Figure 1 An illustrative computer system architecture 100 according to aspects of the present disclosure is depicted. The computer system architecture 100 includes a client device 120, a manufacturing apparatus 122, a substrate measurement system 126, a prediction server 112 (e.g., for generating prediction data, for providing model adaptation, for using a knowledge base, etc.), and a data store 150. The prediction server 112 can be part of a prediction system 110. The prediction system 110 can further include server machines 170 and 180. In some embodiments, the computer system architecture 100 can include or be part of a manufacturing system for processing substrates, such as the manufacturing system 200 of Figure 2A FIG. 1. In additional or alternative embodiments, the computer system architecture 100 can include or be part of a substrate placement prediction system (e.g., that assesses conditions of one or more chamber elements in a processing chamber). Further details regarding substrate placement prediction systems are provided with respect to Figures 3-4

[0040] ​The elements of client device 120, manufacturing equipment 122, substrate measurement system 126, prediction system 110, and / or data store 150 can be coupled to one another via network 140. In some embodiments, network 140 is a public network that provides 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 that provides 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 can include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet network), wireless networks (e.g., an 802.11 network or a Wi-Fi network), cellular networks (e.g., a Long Term Evolution (LTE) network), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.

[0041] Client device 120 can include a computing device such as a personal computer (PC), a laptop computer, a mobile phone, a smart phone, a tablet computer, a notebook computer, a networked television (“smart television”), a networked media player (e.g., a Blu-ray player), a set-top box, an over-the-top (OTT) streaming device, an operating box, and / or the like.

[0042] Manufacturing equipment 122 can produce products in accordance with a recipe. In some embodiments, manufacturing equipment 122 can include or be part of a manufacturing system that includes one or more stations (e.g., processing chambers, transfer chambers, load locks, factory interfaces, and / or the like) configured to perform different operations on substrates.

[0043] Substrate measurement system 126 can be an element of a manufacturing system that can be used to measure substrates before and / or after the substrates are processed in one or more processing chambers. Substrate measurement system 126 can be configured to generate optical emission spectroscopy data, reflectometry data, and / or other metrology data. Substrate measurement system 126 can include one or more elements configured to collect and / or generate measurement data associated with one or more portions of a profile of a surface of a substrate after the substrate has been removed from a processing chamber.

[0044] In some embodiments, the substrate measurement system 126 can be configured to generate metrology data associated with substrates processed by the other manufacturing equipment 122. The metrology data can include values for one or more of film property data (e.g., wafer spatial film properties such as thickness), dimensions (e.g., thickness, height, etc.), dielectric constant, dopant concentration, density, defects, etc. The metrology data can be data for finished or semi-finished products, or data for test substrates (e.g., blanket wafers). Some embodiments are discussed with reference to using reflectometry data and thickness profiles to determine a condition of a chamber component. However, it should be understood that the principles and embodiments set forth herein with respect to reflectometry data and thickness profiles are also applicable to other types of metrology data. For example, measurements can be made of particle counts, optical constants of a coating, surface roughness of a coating, material composition of a coating, etc. Many areas on a substrate can be measured in this manner, and can be used to generate profiles of particle counts, optical constants, surface roughness, material composition, etc. on the measured substrate.

[0045] The substrate measurement system 126 can be configured to generate metrology data associated with substrates before and / or after processing of the substrates. The substrate measurement system 126 can be integrated with a station of a manufacturing system that includes the manufacturing equipment 122. In some embodiments, the substrate measurement system 126 can be coupled to or part of a station of a processing tool (e.g., a processing chamber, a transfer chamber, etc.) that is maintained under vacuum. Such a substrate measurement system 126 can be referred to as on-board metrology equipment. Thus, the substrate measurement system 126 can be used to measure a substrate while the substrate is under vacuum. For example, after a substrate processing (e.g., an etching process, a deposition process, etc.) is performed on a substrate, metrology data for the processed substrate can be generated by the substrate measurement system 126 without removing the processed substrate from the vacuum. 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, etc.) that is not maintained under vacuum. Such a substrate measurement system 126 can be referred to as on-board metrology equipment.

[0046] Instead of the substrate measurement system 126 being included in the manufacturing system (e.g., attached to a factory interface or a transfer chamber), the substrate measurement system 126 can be a device separate (i.e., external) from the manufacturing equipment 122. For example, the substrate measurement system 126 can be a standalone device that is not coupled to any station of the manufacturing equipment 122. To obtain measurements of a substrate using the separate substrate measurement system 126, a user of the manufacturing system (e.g., an engineer, an operator) can cause a substrate being processed at the manufacturing equipment 122 to be 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 metrology data generated for a substrate to a client device 120 coupled to the substrate measurement system 126 (e.g., for presentation to a user of the manufacturing, such as an operator or an engineer) via the network 140. In other or similar embodiments, a user of the manufacturing system can obtain metrology data for a substrate from the substrate measurement system 126 and can provide the metrology data to the computer system architecture via a graphical user interface (GUI) of the client device 120.

[0047] The data storage 150 can be a memory (e.g., a random access memory), a drive (e.g., a hard disk drive, a flash drive), a database system, or another type of element or device capable of storing data. The data storage 150 can include multiple storage elements (e.g., multiple drives or multiple databases) that can span multiple computing devices (e.g., multiple server computers). The data storage 150 can store profiles (e.g., generated from reflectometry data, from spectroscopy data, etc.), such as film thickness profiles and / or other substrate profiles. The film thickness profiles and / or other substrate profiles can include historical profiles and / or current profiles.

[0048] One or more portions of the data storage 150 can be configured to store data that is inaccessible to a user of the manufacturing system. In some embodiments, a user of the manufacturing system can not have access to all data stored at the data storage 150. In other or similar embodiments, a portion of the data stored at the data storage 150 is inaccessible to the user, while another portion of the data stored at the data storage 150 is accessible by the user. In some embodiments, the inaccessible data stored at the data storage 150 is encrypted with an encryption mechanism that is unknown to the user (e.g., the data is encrypted using a private key). In other or similar embodiments, the data storage 150 can include multiple data storages, where the data that is inaccessible to the user is stored in a first data storage, and the data that is accessible to the user is stored in a second data storage.

[0049] In some embodiments, the prediction system 110 includes a server machine 170 and a server machine 180. The server machine 170 includes a training set generator 172 that is capable of generating training data sets (e.g., a set of data inputs and a set of target outputs) to train, validate, and / or test a machine learning model 190 or a set of machine learning models 190. Further details regarding the training set generator 172 are described below with reference to Figure 4 Some operations of the training set generator 172 are described in detail. In some embodiments, the training set generator 172 can split the training data into training sets, validation sets, and test sets.

[0050] The server machine 180 can include a training engine 182. An engine can refer to hardware (e.g., circuitry, special purpose logic, programmable logic, microcode, processing apparatus, etc.), software (e.g., instructions running on a processing apparatus, general purpose computer system, or special purpose machine), firmware, microcode, or combinations thereof. The training engine 182 is capable of training a machine learning model 190 or a set of machine learning models 190. The machine learning model 190 can refer to a model artifact established by the training engine 182 using training data that includes training inputs and corresponding target outputs (correct answers for the respective training inputs). The training engine 182 can find patterns in the training data that map the training inputs to the target outputs (the answers to be predicted) and provide a machine learning model 190 that captures these patterns. The machine learning model 190 can include a linear regression model, a partial least squares regression model, a Gaussian regression model, a random forest model, a support vector machine model, a neural network, a ridge regression model, etc.

[0051] The training engine 182 is also capable of validating the trained machine learning model 190 using the corresponding feature sets of the validation sets of the training set generator 172. In some embodiments, the training engine 182 can assign a performance rating to each of the set of trained machine learning models 190. The performance rating can correspond to an accuracy of the corresponding trained model, a speed of the corresponding model, and / or an efficiency of the corresponding model. In accordance with the embodiments described herein, the training engine 182 can select a trained machine learning model 190 that has a performance rating that meets a performance criterion used by the prediction engine 114. Further details regarding the training engine 182 are described below with reference to Figure 5 provided.

[0052] The prediction server 112 includes a prediction engine 114 that is capable of providing data from the substrate measurement system 126 (e.g., a film thickness profile) as an input to the trained machine learning model 190 and running the trained model 190 on the 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 to have a performance rating that meets a performance criterion. As described with reference to Figure 6Further described, in some embodiments, the prediction engine 114 processes input data using the model 190 to evaluate a substrate placement for processing a substrate in a processing chamber.

[0053] It should be noted that, in some other implementations, the functionality of the server machines 170 and 180 and the prediction server 112 can be provided by a larger or smaller number of machines. For example, in some embodiments, the server machines 170 and 180 can be integrated into a single machine, while in some other or similar embodiments, the server machines 170 and 180 and the prediction server 112 can be integrated into a single machine. Generally speaking, functionality described in one implementation as being carried out by the server machines 170, the server machines 180, and / or the prediction server 112 can also be carried out on the client device 120. Moreover, functionality attributed to a particular element can be carried out by different or multiple elements operating together.

[0054] In embodiments, a "user" can be represented as a single individual. However, other embodiments of the present disclosure encompass a "user" as an entity controlled by multiple users and / or automated sources. For example, a group of individual users that collectively act as a group of administrators can be considered a "user."

[0055] FIG. 2 is a top-down schematic view of an example manufacturing system 200 according to aspects of the present disclosure. The manufacturing system 200 can perform one or more processes on a substrate 202. According to aspects of the present disclosure, the substrate 202 can be any suitable rigid, fixed-size planar article, such as a silicon-containing disk or wafer, a patterned wafer, a glass plate, etc. suitable for having electronic devices or circuit elements fabricated thereon. In some embodiments, the manufacturing system 200 can be a cluster tool, a single-wafer processing tool, a batch processing tool, etc. according to aspects of the present disclosure. Figure 1 According to the described embodiments, the manufacturing system 200 can include or be part of the computer system architecture 110.

[0056] The manufacturing system 200 can include a processing tool 204 and a factory interface 206 coupled to the processing tool 204. The processing tool 204 can include a housing 208 having a transfer chamber 210 therein. The transfer chamber 210 can include one or more processing chambers (also referred to as process chambers) 214, 216, 218 arranged about and coupled thereto. The processing chambers 214, 216, 218 can be coupled to the transfer chamber 210 by respective ports, such as slit valves, etc. The transfer chamber 210 can also include a transfer chamber robot 212 configured to transfer the substrate 202 between the processing chambers 214, 216, 218, a load lock 220, etc. The transfer chamber robot 212 can include one or more arms, with each arm including one or more end effectors at an end of each arm. The end effectors can be configured to manipulate a particular object, such as a wafer.

[0057] In some embodiments, the transfer chamber 210 can also include metrology equipment attached thereto, such as a substrate measurement system 126. The substrate measurement system 126 can be configured to generate metrology data associated with the substrate 202 before or after substrate processing while the substrate is maintained in a vacuum environment. As shown, the substrate measurement system 126 can be attached to or disposed within the transfer chamber 210. If the substrate measurement system 126 is disposed within or coupled to the transfer chamber 210, metrology data associated with the substrate 202 can be generated without removing the substrate 202 from the vacuum environment (e.g., transferred to the factory interface 206). Figure 2A

[0058] The processing chambers 214, 216, 218 can be adapted to perform any number of processes on the substrate 202. The same or different substrate processes can be performed in each of the processing chambers 214, 216, 218. The substrate processes can include atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, curing, pre-cleaning, metal or metal oxide removal, etc. Other processes can be performed on the substrate therein.

[0059] 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 interface and couple with the transfer chamber 210 on one side and the factory interface 206 on the other side. The load lock 220 can have an atmosphere that is controlled, which in some embodiments can change from a vacuum environment (in which substrates can be transferred to and from the transfer chamber 210) to an inert gas environment at (or near) atmospheric pressure (in which substrates can be transferred to and from the factory interface 206).

[0060] The factory interface 206 can be any suitable enclosure, such as an equipment front end module (EFEM). The factory interface 206 can be configured to receive substrates 202 from substrate carriers 222 (e.g., front opening unified pods (FOUPs)) docked at various load ports 224 of the factory interface 206. A factory interface robot 226 (shown in dashed line) can be configured to transfer the substrates 202 between the substrate carriers 222 (also referred to as containers) 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.

[0061] ​In some embodiments, manufacturing system 200 may include a substrate measurement system 126 attached to factory interface 206. The substrate measurement system 126 attached to the factory interface may be configured to generate measurement data associated with substrate 202 before substrate 202 is placed in a vacuum environment (e.g., transferred to loading lock 220) and / or after substrate 202 is removed from the vacuum environment (e.g., removed from loading lock 220).

[0062] Manufacturing system 200 can also be connected to client devices (e.g., Figure 1 The client device 120 is configured to provide information about the manufacturing system 200 to a user (e.g., an operator). In some embodiments, the client device may provide information to the user of the manufacturing system 200 via one or more graphical user interfaces (GUIs). For example, the client device may provide information via the GUI about one or more chamber condition measurements (e.g., during substrate processing) of the processing chambers 214, 216, 218.

[0063] Manufacturing system 200 may also include or be coupled to system controller 228. System controller 228 may be and / or include computing devices such as personal computers, server computers, programmable logic controllers (PLCs), microcontrollers, etc. System controller 228 may include one or more processing devices, which may be general-purpose processing devices, such as microprocessors, central processing units, etc. More specifically, the processing device may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction set (VLIW) microprocessor, or a processor that implements other instruction sets or a combination of instruction sets. The processing device 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), network processors, etc. System controller 228 may include data storage devices (e.g., one or more disk drives and / or solid-state drives), main memory, static memory, network interfaces, and / or other elements. System controller 228 may execute instructions to perform any one or more of the methods and / or embodiments described herein. In some embodiments, system controller 228 may execute instructions to perform one or more operations at manufacturing system 200 according to a processing recipe. The instructions may be stored on a computer-readable storage medium, which may include main memory, static memory, secondary storage, and / or processing means (during execution of the instructions).

[0064] In some embodiments, system controller 228 can receive data from substrate measurement system 126 based on measurements of substrates that have undergone processing in processing chambers 214, 216, 218. The data received by system controller 228 can include spectral data, reflectometry data, and / or other data for all or a portion of substrate 202. The data received from substrate measurement system 126 can be stored in data storage 250. Data storage 250 can be included as an element within system controller 228, or can be a separate element from system controller 228. In some embodiments, data storage 250 can be or include a portion of data storage 150 as described with respect to Figure 1

[0065] Figure 2B One embodiment of a substrate measurement system 251 that can be used to measure a processed substrate is shown. Substrate measurement system 251 can be an integrated measurement and / or imaging system (e.g., an integrated reflectometry (IR) system) configured to measure film properties (e.g., such as thickness) across a surface of a substrate 264 after the substrate 264 has been processed in a processing chamber. Reflectometry is a measurement technique that uses variations in the light reflected from an object to determine the geometric and / or material properties of the object. A reflectance spectrometer measures the reflected light intensity over a range of wavelengths. For dielectric films, these intensity variations can be used to determine the thickness of the film.

[0066] ​For example, the substrate measurement system 251 can be used to monitor the results of processing on one or more substrates for etching and deposition processes, including film thickness. An integrated measurement and / or imaging system can be used to measure the surface of the substrate 264 while the substrate 264 is still in the device manufacturing 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 positioned inside a factory interface or a transfer chamber. The substrate measurement system 251 can also be a standalone system that is not connected to the manufacturing system. The substrate measurement system 251 can be mechanically isolated from the factory interface and the external environment to protect the substrate measurement system 251 from external vibrations. In some embodiments, the substrate measurement system 251 and elements included therein can provide analytical measurements (e.g., thickness measurements) that can provide a uniformity profile (referred to herein as a profile map) across the surface of the substrate 264. A computing device can process data from the substrate measurement system 251 and provide feedback to a user. The substrate measurement system 251 can be a component capable of measuring film thickness and / or other film properties (e.g., optical constants, particle counts, roughness, etc.) on a portion of a substrate or an entire substrate after the substrate has been processed in a chamber. The measurement results can be used to determine when to perform maintenance on a processing chamber, when to perform further testing on a substrate, when to flag a substrate as out of specification, placement of a substrate during substrate processing, etc.

[0067] When the substrate 264 is lowered and secured to the substrate support 256 (e.g., a chuck), 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 (the center of the chuck corresponds to an axis of rotation about which the chuck rotates) 256 and apply the one or more coordinate transformations to correct for the offset, as described in more detail below.

[0068] The substrate measurement system 251 can include a rotational actuator 252 and a linear actuator 254. The rotational actuator 252 can be a motor, a rotational actuator (e.g., an electric rotational actuator), etc. The linear actuator 254 can be an electric linear actuator that can convert rotational motion in a motor to linear or straight line motion along an axis. The substrate measurement system 251 can include a substrate support 256, a camera 258, a sensor 260, and a processing device 262.

[0069] The substrate support 256 can be a vacuum chuck, an electrostatic chuck, a magnetic chuck, a mechanical chuck (e.g., a four-point chuck, a three-point chuck, an edge / ring chuck, etc.), or other types of chucks. The substrate support 256 can secure a substrate 264 (e.g., a wafer). The rotary actuator 252 can rotate the substrate support 256 about the first axis 253. The rotary actuator 252 can be controlled by a servo controller and / or a servo motor, which can allow precise control of the rotational position, velocity, and / or acceleration of the rotary actuator and, thus, can allow precise control of the rotational position, velocity, and / or acceleration of the substrate support 256. The linear actuator 254 can linearly move the substrate support 256 along the second axis 255. The linear actuator 254 can be controlled by a servo controller and / or a servo motor 272, which can allow precise control of the linear position, velocity, and acceleration of the linear actuator 254 and, thus, can allow precise control of the linear position, velocity, and acceleration of the substrate support 256.

[0070] The camera 258 can be positioned above 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 other suitable types of cameras. The sensor 260 can also be positioned above the substrate support 256 and can measure at least one target position on the substrate at a time (e.g., can generate a reflection measurement or other measurement of the target position). The camera 258 and the sensor 260 can be fixed in a fixed position on the substrate measurement system 251, while the substrate support 256 can be moved in r-0 motion by the rotary actuator 252 and the linear actuator 254.

[0071] In some embodiments, the processing device 262 can determine that the substrate 264 is not centered on the substrate support 256 based on one or more images of the substrate 264 generated by the camera 258. When the substrate 264 is initially placed on the substrate support 256, the substrate 264 can not be located at the center of the substrate support 256. The robot blade 270 can place the substrate 264 on a transfer station 268 (e.g., a set of lift pins). The substrate support 256 can be moved in a first direction along the second axis 255 such that the substrate support 256 is positioned at the transfer station 268. The transfer station 268 can be located on a lift mechanism 266 (or can be a set of lift pins), which can move the transfer station 268 up and down in a vertical direction (i.e., perpendicular to the second axis 255 and parallel to the first axis 253). When the substrate support 256 is positioned at the transfer station 268, the substrate 264 can be received by the substrate support 256. The substrate 264 can not be located at the center of 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 an edge of the substrate 264 is at a target position.

[0072] The substrate support can be rotated 360 degrees and images can be generated during rotation of the substrate support. One or more measurements and / or image generation can be taken at various different theta values with the chuck, and the detected edge position can vary. Variations in the detected edge can indicate that the substrate (which can be a circular substrate) is off-center. Furthermore, the determined variations in the detected edge can be used to calculate the amount of misalignment.

[0073] In one embodiment, the parameters (r, Q) determine the misalignment of the substrate with respect to the stage. With these parameters, the motion system can establish forward and inverse transforms that convert (r, Q) coordinates of the stage to (r, Q) coordinates of the substrate. Then, the motion system can calculate a trajectory in space of the substrate while sending commands to move the motors attached to the substrate support 256. In one embodiment, the motion system can calculate a trajectory in arbitrary space because it runs real-time control software of the motion drivers (e.g., over an Ethernet network) connected to the linear and rotary actuators. The processing device 262 can calculate the corrected trajectory and transmit the commanded positions to the motion drivers in real-time (e.g., at a rate of 1 kHz).

[0074] In some embodiments, the rotary actuator 252 used to measure the target positions causes an offset between the field of view of the sensor 260 and the target positions on the substrate 264 due to the substrate 264 not being centered on the substrate support 256. In this case, the linear actuator 254 can linearly move the substrate support 256 along the second axis to correct the offset. Then, the sensor 260 can measure the target positions on the substrate 264. Once measurements of all target points on the substrate are measured, the processing device 262 can determine a uniformity profile across the surface of the substrate 264 based on the measurements.

[0075] 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 about which the substrate support 256 rotates) and the center of the substrate 264 applied during rotation of the substrate support 256 to correct the misalignment.

[0076] Figure 2C is a cross-sectional schematic side view of a substrate measurement sub-system 282 according to aspects of the present disclosure. The substrate measurement sub-system 282 can be configured to obtain measurements of one or more portions of a substrate, such as a substrate 202, before or after processing the substrate 202 at a processing chamber. The substrate measurement sub-system 282 can correspond to the substrate measurement sub-system 282 in embodiments. Figure 2A Figure 2A ​The substrate measurement system 126 can include a substrate measurement subsystem 282. The substrate measurement subsystem 282 can obtain measurements of a portion of the substrate 202 by generating data associated with the portion of the substrate 202. In some embodiments, the substrate measurement subsystem 282 can be configured to generate spectral data, position data, and / or other attribute data related to the substrate 202.

[0077] The substrate measurement subsystem 282 can be configured to generate one or more types of data for the substrate, including spectral data, position data, substrate attribute data, etc. The substrate measurement subsystem 282 can generate data for the substrate in response to a request to obtain one or more measurements of the substrate prior to or after processing the substrate at a manufacturing system. The substrate measurement subsystem 282 can include one or more elements that facilitate generating data for the substrate. For example, the substrate measurement subsystem can include spectral sensing elements for sensing a spectrum or spectra from a portion of the substrate and generating spectral data for the substrate. In some embodiments, the spectral sensing elements can be interchangeable elements that can be configured based on a type of processing being performed at a manufacturing system or a target type of measurement to be obtained at the substrate measurement subsystem. For example, one or more elements of the spectral sensing elements can be exchanged at the substrate measurement subsystem to enable collection of reflectance spectral data, ellipsometry spectral data, hyperspectral imaging data, chemical imaging (e.g., X-ray photoelectron spectroscopy (XPS), energy dispersive X-ray spectroscopy (EDX), X-ray fluorescence (XRF), etc.) data, etc.

[0078] The substrate measurement subsystem 282 can 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 can include a substrate sensing element 284 configured to detect when the substrate 202 is delivered to the substrate measurement subsystem 282. The substrate sensing element 284 can include any element configured to detect when the substrate 202 is delivered to the substrate measurement subsystem 282. For example, the substrate sensing element 284 can include an optical sensing element that transmits a beam of light through an entrance to the substrate measurement subsystem 282. When the substrate 202 is placed within the substrate measurement subsystem 282, the substrate sensing element 284 can detect that the substrate 202 has been delivered to the substrate measurement subsystem 282 in response to the substrate 202 blocking the beam of light transmitted through the entrance to the substrate measurement subsystem 282. In response to detecting that the substrate 202 has been delivered to the substrate measurement subsystem 282, the substrate sensing element 284 can deliver an indication to the controller 283 that the substrate 202 has been delivered to the substrate measurement subsystem 282.

[0079] In some embodiments, the substrate sensing element 284 can be further configured to detect identification data associated with the substrate 202. In some embodiments, when the substrate 202 is transported to the substrate measurement sub-system 282, the substrate 202 can be embedded within a substrate carrier (not shown). The substrate carrier can include one or more registration features that can identify the substrate 202. For example, an optical sensing element of the substrate sensing element 284 can detect that the substrate 202 embedded within the substrate carrier has blocked a light beam transmitted through an entrance to the substrate measurement sub-system 282. The optical sensing element can further detect one or more registration features included on the substrate carrier. In response to detecting the one or more registration features, the optical sensing element can generate an optical signature associated with the one or more registration features. The substrate sensing element 284 can transmit the optical signature generated by the optical sensing element to the controller 283 along with an indication that the substrate has been placed within the substrate measurement sub-system 282. In response to receiving the optical signature from the sensing element 284, the controller 283 can analyze the optical signature to determine identification information associated with the substrate 202. The identification information associated with the substrate 202 can include an identifier of the substrate 202, an identifier of a process for the substrate 202 (e.g., a lot number or a process run number), an identifier of a type of the substrate 202 (e.g., a wafer, etc.), etc.

[0080] The substrate measurement sub-system 282 can include one or more elements configured to determine a position and / or orientation of the substrate 202 within the substrate measurement sub-system 282. The position and / or orientation of the substrate 202 can be determined based on identification of a reference position of the substrate 202. The reference position can be a portion of the substrate 202 that includes an identification feature associated with a particular portion of the substrate 202. For example, the substrate 202 can have a reference tag embedded in a central portion of the substrate 202. In another example, the substrate 202 can have one or more structural features included on a surface of the substrate 202 at a central portion of the substrate 202. The controller 283 can determine the identification feature associated with the particular portion of the substrate 202 based on the determined identification information of the substrate 202. For example, in response to determining that the substrate 202 is a wafer, the controller 283 can determine one or more identification features that are typically included in a portion of a wafer.

[0081] The controller 283 can identify a reference position of the substrate 202 using one or more camera elements 285 configured to capture image data of the substrate 202. The camera elements 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 an identifying feature associated with the reference position of the substrate 202. The controller 283 can further determine a position and / or orientation of the substrate 202 as depicted in the image data based on the identified identifying feature of the substrate 202. The controller 283 can determine a position and / or orientation of the substrate 202 based on the identified identifying feature of the substrate 202 and the determined position and / or orientation of the substrate 202 as depicted in the image data.

[0082] 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 can include one or more coordinates (e.g., Cartesian coordinates, polar coordinates, etc.) each associated with a portion of the substrate 202, where each coordinate is determined based on a distance from the reference position of 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 a portion of the substrate 202 including the reference position, where the first position data includes Cartesian coordinates of (0, 0). The controller 283 can generate second position data associated with a second portion of the substrate 202 relative to the reference position. For example, a portion of the substrate 202 located about 2 nanometers (nm) east of the reference position can be assigned Cartesian coordinates of (0, 1). In another example, a portion of the substrate 202 located about 5 nm north of the reference position can be assigned Cartesian coordinates of (1, 0).

[0083] The controller 283 can determine one or more portions of the substrate 202 to measure based on the determined position data for the substrate 202. In some embodiments, the controller 283 can receive one or more operations of a process recipe 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 the one or more operations of the process recipe. For example, the controller 283 can receive an indication to perform an etching process on the substrate 202, where several structural features are etched onto a surface of the substrate 202. As a result, the controller 283 can determine one or more structural features to measure and an expected location of these features on various portions of the substrate 202.

[0084] The substrate measurement subsystem 282 can include one or more measurement elements for measuring the substrate 202. In some embodiments, the substrate measurement subsystem 282 can include one or more spectral sensing elements 287 configured to generate spectral data for one or more portions of the substrate 202. As previously discussed, the spectral data can correspond to the intensity of the detected energy wave for each wavelength of the detected wave (i.e., the intensity or amount of energy).

[0085] In one embodiment, a plurality of wavelengths can be included in the reflected energy waves received by the substrate measurement subsystem 282. Each reflected energy wave can be associated with a different portion of the substrate 202. In some embodiments, the intensity of each reflected energy wave received by the substrate measurement subsystem 282 can be measured. Each intensity can be measured for each wavelength of the reflected energy waves received by the substrate measurement subsystem 282. The association between each intensity and each wavelength can form the basis of the spectral data. In some embodiments, one or more wavelengths can be associated with an intensity value that is outside of an expected range of intensity values. In such embodiments, the intensity value that is outside of the expected range of intensity values can be an indication that a defect exists at a portion of the substrate 202.

[0086] The measurement elements for measuring the substrate 202 can also include non-spectral sensing elements configured to collect and generate non-spectral data. For example, the measurement elements can include eddy current sensors or capacitance sensors. Although some embodiments of the present description can refer to collecting and using spectral data for the substrate 202, embodiments of the present description can be applicable to non-spectral data collected for the substrate 202.

[0087] The spectral sensing elements 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 elements 287 can include a wave generator 288 and a reflected wave receiver 291. In some embodiments, the wave generator 288 can be a light wave generator configured to generate a beam of light toward a portion of the substrate 202. In such embodiments, the reflected wave receiver 291 can be configured to receive the reflected beam of light from the portion of the substrate 202. The wave generator 288 can be configured to generate an energy stream 289 (e.g., a beam of light) and transmit the energy stream 289 to a portion of the substrate 202. A reflected energy wave 290 can be reflected from the portion of the substrate 202 and received by the reflected wave receiver 291. Although a single energy wave is shown being reflected from the surface of the substrate 202, a plurality of energy waves can be reflected from the surface of the substrate 202 and received by the reflected wave receiver 291. Figure 2C Although a single energy wave is shown being reflected from the surface of the substrate 202, a plurality of energy waves can be reflected from the surface of the substrate 202 and received by the reflected wave receiver 291.

[0088] In response to the reflected energy wave 290 received by the reflected wave receiver 291 from a portion of the substrate 202, the spectral sensing element 287 can measure the wavelength of each wave contained in the reflected energy wave 289. The spectral sensing element 287 can further measure the intensity of each measured wavelength. In response to measuring each wavelength and each wavelength intensity, the spectral sensing element 287 can generate spectral data for that portion of the substrate 202. The spectral sensing element 287 can transmit the generated spectral data to the controller 283. The controller 283 can, in response to receiving the generated spectral data, generate a mapping between the received spectral data and the position data of the measured portion of the substrate 202.

[0089] The substrate measurement subsystem 282 can be configured to generate a specific type of spectral data based on the type of measurement to be obtained at the substrate measurement subsystem 282. In some embodiments, the spectral sensing element 287 can be a first spectral sensing element configured to generate one type of spectral data. For example, the spectral sensing element 287 can be configured to generate reflectance spectral data, elliptic polarized spectral data, hyperspectral imaging data, chemical imaging data, thermal spectral data, or conduction spectral data. In such embodiments, the first spectral sensing element can be removed from the substrate measurement subsystem 282 and replaced with a second spectral sensing element configured to generate different types of spectral data (e.g., reflectance spectral data, elliptic polarized spectral data, hyperspectral imaging data, or chemical imaging data).

[0090] The controller 283 can determine the data type (i.e., spectral data, non-spectral data) generated for the substrate 202 based on the type of measurements obtained for one or more portions of the substrate 202. In some embodiments, the controller 283 can be based on measurements obtained from... Figure 2A The system controller 228 receives a notification to determine one or more types of measurements. In other or similar embodiments, the controller 283 may determine one or more types of measurements based on instructions to generate measurements on a portion of the substrate 202. In response to determining one or more types of measurements to be obtained, the controller 283 may determine the type of data to be generated for the substrate 202. For example, the controller 283 may determine that spectral data is to be generated for the substrate 202, and that a second spectral sensing element is the optimal sensing element for obtaining measurements of the determined type for one or more portions of the substrate 202. In response to determining that the second sensing element is the optimal sensing element, the controller 283 may send a notification to the system controller indicating that the first spectral sensing element should be replaced by the second spectral sensing element and that the second spectral sensing element should be used to obtain one or more types of measurements on 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, wherein the client device may provide the notification to a user (e.g., an operator) of the manufacturing system via a GUI.

[0091] In other or similar embodiments, the optical spectral sensing element 287 can be configured to produce multiple types of spectral data. In such embodiments, in accordance with previously described embodiments, the controller 283 can cause the optical spectral sensing element 287 to produce a particular type of spectral data based on a type of measurement to be obtained for one or more portions of the substrate 202. In response to determining the type of measurement to be obtained, the controller 283 can determine that a first type of spectral data is to be produced by the optical spectral sensing element 287. Based on determining that the first type of spectral data is to be produced by the optical spectral sensing element 287, the controller 283 can cause the optical spectral sensing element 287 to produce the first type of spectral data for one or more portions of the substrate 202.

[0092] As previously described, the controller 283 can determine one or more portions of the substrate 202 to be measured at the substrate measurement subsystem 282. In some embodiments, one or more measurement elements, such as the optical spectral sensing element 287, can be fixed elements within the substrate measurement subsystem 282. In such embodiments, the substrate measurement subsystem 282 can include one or more positioning elements 295 configured to modify a position and / or orientation of the substrate 202 relative to the optical spectral sensing element 287. In some embodiments, the positioning elements 295 can be configured to translate the substrate 202 along a first axis and / or a second axis relative to the optical spectral sensing element 287. In other or similar embodiments, the positioning elements 295 can be configured to rotate the substrate 202 about a third axis relative to the optical spectral sensing element 287.

[0093] As the optical spectral sensing element 287 produces spectral data for one or more portions of the substrate 202, the positioning elements 295 can modify the position and / or orientation of the substrate 202 in accordance with one or more determined portions of the substrate 202 to be measured. For example, prior to the optical spectral sensing element 287 producing spectral data for the substrate 202, the positioning elements 295 can position the substrate 202 at Cartesian coordinates (0, 0) and the optical spectral sensing element 287 can produce first spectral data for the substrate 202 at the Cartesian coordinates (0, 0). In response to the optical spectral sensing element 287 producing the first spectral data for the substrate 202 at the Cartesian coordinates (0, 0), the positioning elements 240 can translate the substrate 202 along a first axis such that the optical spectral sensing element 287 is configured to produce second spectral data for the substrate 202 at Cartesian coordinates (0, 1). In response to the optical spectral sensing element 287 producing the second spectral data for the substrate 202 at the Cartesian coordinates (0, 1), the controller 283 can rotate the substrate 202 along a second axis such that the optical spectral sensing element 287 is configured to produce third spectral data for the substrate 202 at Cartesian coordinates (1, 1). This process can occur multiple times until spectral data is produced for each determined portion of the substrate 202.

[0094] In some embodiments, one or more layers 297 of material can be included on a surface of the substrate 202. The one or more layers 297 can include etching material, photoresist material, masking material, deposited material, etc. In some embodiments, the one or more layers 297 can include etching material that has been etched according to an etching process performed in a processing chamber. In such embodiments, spectral data can be collected from one or more portions of the etching material of the layer 297 deposited on the substrate 202 that have not been etched according to previously disclosed embodiments. In other or similar embodiments, the one or more layers 297 can include etching material that has been etched according to an etching process in a processing chamber. In such embodiments, one or more structural features (e.g., lines, pillars, openings, etc.) can be etched into the one or more layers 297 of the substrate 202. In such embodiments, spectral data can be collected for the one or more structural features etched into the one or more layers 297 of the substrate 202.

[0095] In some embodiments, the substrate measurement subsystem 282 can include one or more additional sensors configured to capture additional data of the substrate 202. For example, the substrate measurement subsystem 282 can include additional sensors configured to determine a thickness of the substrate 202, a thickness of a film deposited on a surface of the substrate 202, etc. Each sensor can be configured to transmit the captured data to the controller 283.

[0096] According to embodiments described herein, in response to receiving at least one of spectral data, positional data, or attribute data of the substrate 202, the controller 283 can transmit the received data to the system controller 228 for processing and analysis.

[0097] In some embodiments, the substrate measurement subsystem 282 includes one or more image capture devices 299, such as cameras (e.g., including complementary metal-oxide-semiconductor (CMOS) sensors or charge-coupled device (CCD) sensors), connected to the controller 283. The image capture devices 299 can produce images (e.g., two-dimensional (2D) color images, infrared (IR) images, near-IR images, etc.). In embodiments, the images can be processed by one or more trained machine learning models, along with spectral data produced by the spectral sensing elements 287, to make determinations about one or more chamber elements.

[0098] Figure 3An illustrative system architecture 300 for substrate placement prediction of a processing chamber is depicted in accordance with aspects of the present disclosure. In some embodiments, system architecture 300 can include or be part of one or more elements of computer architecture 100 and / or manufacturing system 200. System architecture 300 can include one or more elements of manufacturing equipment 122 (e.g., substrate measurement system 126), server machine 320, and server machine 350.

[0099] As previously described, manufacturing equipment 122 can produce a product in accordance with a recipe or by performing operations over a period of time. Manufacturing equipment 122 can include a processing chamber 310 configured to perform substrate processing on a substrate in accordance with a substrate processing recipe. In some embodiments, the recipe can be a blanket wafer recipe that deposits a film on a test substrate. In some embodiments, the recipe can be a blanket wafer recipe that etches a surface of a test substrate. In some embodiments, processing chamber 310 can be any of the processing chambers 214, 216, 218 described with reference to Figure 2A Manufacturing equipment 122 can also include substrate measurement system 126, as described herein.

[0100] Manufacturing equipment 122 can be coupled to server machine 320. Server machine 320 can include a processing device 322 and / or data storage 332. In some embodiments, processing device 322 can be configured to execute one or more instructions to perform operations at manufacturing equipment 122. For example, processing device 322 can include or be part of system controller 228 described with reference to Figure 2A In some embodiments, data storage 332 can include or be part of data storage 150 and / or data storage 250.

[0101] Processing device 322 can be configured to receive data from one or more elements of manufacturing equipment 122 (i.e., via a network). For example, processing device 322 can receive surface profile data (e.g., wafer maps, thickness profile data, etc.) 336 collected by substrate measurement system 126 after a substrate has been processed in a processing chamber. In another example, processing device 322 can receive metrology data collected by other metrology equipment before and / or after substrate processing of a substrate. Metrology data can include metrology measurements generated for a substrate by an integrated metrology device. In some embodiments, processing device 322 can store received spectral data, film thickness profile data, substrate thickness profile data, and / or received metrology data at data storage 332.

[0102] Processing device 352 may include substrate placement engine 330. Substrate placement engine 330 at processing device 322 may be configured to determine a recommended placement (e.g., a recommended location) of a substrate for processing in processing chamber 310. Processing chamber 310 may be a processing chamber for processing a substrate being measured. Substrate placement engine 330 may determine one or more substrate placement metrics for processing one or more substrates in processing chamber 310 based on substrate surface profile data 336. In some embodiments, substrate placement metrics may include a series of values ​​(e.g., vectors, matrices, etc.) indicating the correlation of specific data combinations, correlations, patterns, and / or relationships present in sensor data. For example, substrate placement metrics may include feature vectors comprising binary values ​​indicating the presence or absence of specific features in the data.

[0103] The substrate placement metric can be compared with known patterns and / or combinations of substrate placement metrics (e.g., a target substrate placement metric). The target substrate placement metric can be associated with an ideal substrate placement for optimal substrate processing. In response to a substrate placement metric determining one or more substrate placement positions satisfying one or more substrate placement criteria (e.g., a condition related to a processed substrate that meets threshold specifications such as a threshold tilt specification), the substrate placement engine 330 can determine a recommended placement of the substrate for processing and / or may provide an alert to the user. In some embodiments, the substrate placement criteria may include a specified combination of values ​​indicated by the substrate placement metrics. Once a recommended substrate placement is determined, the substrate placement engine 330 can output instructions for the robotic arm to use specific settings (e.g., x-settings and / or y-settings) when placing the substrate on the substrate support according to the coordinates of the recommended substrate placement.

[0104] like Figure 3 As shown, in some embodiments, the processing device 322 may include a training set generator 324 and / or a training engine 326. In some embodiments, the training set generator 324 may correspond to training set generator 172 and / or the training engine 326 may correspond to training engine 182, as shown in reference 182. Figure 1The training set generator 324 can be configured to generate a training set 340 to train the machine learning model 334 or a set of machine learning models 334. For example, the training set generator 324 can generate training inputs based on historical substrate surface profile data 336. The substrate surface profile data 336 (e.g., thickness profile maps) can be associated with etch rates and / or tilts on the surface of one or more substrates processed in a processing chamber. In some embodiments, the training set generator 324 can retrieve historical substrate surface profile data 336 (e.g., substrate thickness profile data) from the data store 332 to generate training inputs. The training set generator 324 can generate target outputs indicative of estimated substrate placement values (e.g., substrate placement metrics) for the training inputs based on the historical substrate surface profile data 336. The training set generator 324 can include the generated training inputs and the generated target outputs in the training set 340. For each historical substrate thickness profile, the training set 340 can additionally include an indicator of substrate placement (e.g., relative to an inner diameter of a processing kit ring on a substrate support such as an electrostatic chuck). Referring to Figure 5 Further details are provided regarding generating the training set 340.

[0105] The training engine 326 can be configured to train, validate, and / or test the machine learning model 334 or the set of machine learning models 334. The training engine 326 can provide the training set 340 to train the machine learning model 334 and store the trained machine learning model 334 at the data store 332. In some embodiments, the training engine 326 can validate the trained machine learning model 334 using a validation set 342. The validation set 342 can include surface profile data 336 and associated chamber element condition metrics. The training set generator 324 and / or the training engine 326 can generate the validation set 342 based on historical surface profile data 336. In some embodiments, the validation set 342 can include historical surface profile data 336 that is different from the historical surface profile data 336 included in the training set 340.

[0106] The training engine 326 can provide the historical surface profile data 336 as input to the trained machine learning model 334 and can extract one or more substrate placement metrics for processing a substrate in a processing chamber from one or more outputs of the trained model 334. The input can additionally include lifetimes of one or more elements of the processing chamber and / or one or more images of the substrate used to generate the historical surface profile data 236. The training engine 326 can assign a performance score to the trained model 334 based on an accuracy of the substrate placement metrics. The training engine 326 can select the trained model 334 to use to evaluate substrate placement for processing a substrate in a processing chamber based on surface profile data of substrates processed by the processing chamber.

[0107] As previously discussed, in some embodiments, the training set generator 324 and / or the training engine 326 can be elements of the processing device 322 at the server 320. In additional or alternative embodiments, the training set generator 324 and / or the training engine 326 can be elements of the processing device 352 at the server 350. The server 350 can include or be part of a computing system separate from the manufacturing system 200. In some embodiments, as discussed with reference to Figure 2A The server 320 can include or be part of the system controller 228, as discussed with reference to FIG. 1. In such embodiments, the server 350 can include or be part of a computing system coupled to (i.e., via a network) but separate from the system controller 228. For example, a user of the manufacturing system 200 can be provided access to data stored at one or more portions of the data storage 332 or one or more processes executed at the processing device 322. However, the user of the manufacturing system 200 can not be provided access to any data stored at one or more portions of the data storage 354 or any processes executed at the processing device 352.

[0108] The processing device 352 can be configured to execute the training set generator 324 and / or the training engine 326 in a similar manner as the processing device 322. In some embodiments, the server 350 can be coupled to the manufacturing equipment 122 and / or the inline metrology equipment 130 via a network. Accordingly, in accordance with embodiments described with respect to the processing device 322, the processing device 352 can obtain surface profile data 336 and substrate placement metrics corresponding to the surface profile data 336 for use by the training set generator 324 and / or the training engine 326 to generate the training set 340 and the validation set 342. In other or similar embodiments, the server 350 is not coupled to the manufacturing equipment 122 and / or the external metrology equipment 130. Accordingly, the processing device 352 can obtain the surface profile data 336 from the processing device 322.

[0109] The training set generator 324 at the processing device 352 can generate the training set 340 in accordance with previously described embodiments. In accordance with previously described embodiments, the training engine 326 at the processing device 352 can train and / or validate the machine learning model 334. 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 the server machine 320. In accordance with embodiments described herein, the processing device 352 can obtain surface profile data 336 from the other manufacturing equipment and / or the other server machines. In some embodiments, the training set 340 and / or the validation set 342 can be generated based on surface profile data 336 obtained for substrates processed at the processing chamber 310 and other surface profile data obtained for other substrates processed at processing chambers of other manufacturing systems.

[0110] In response to the training engine 326 selecting a trained model 334 to use, the processing device 352 can transmit the trained model 334 to the processing device 322. The substrate placement engine 330 can use the trained model 334 to provide substrate placement recommendations, as previously described.

[0111] Figure 4 Model training workflow 405 and model application workflow 417 for substrate placement determination are shown, in accordance with one embodiment. The model training workflow 405 and model application workflow 417 can be carried out by processing logic executed by a processor of a computing device. One or more of these workflows 405, 417 can be implemented, for example, by one or more machine learning models implemented on a processing device and / or other software and / or firmware executed on a processing device.

[0112] The model training workflow 405 is used to train one or more machine learning models (e.g., deep learning models) to determine optimal substrate placement for processing a substrate in a processing chamber. The model application workflow 417 is used to apply one or more trained machine learning models to carry out substrate placement evaluation. Each contour map 412 can be associated with a thickness profile or other surface profile of a substrate being processed. For example, each of the contour maps 412 can reflect a thickness of a respective substrate after a processing operation (e.g., an etching operation, etc.) is performed on the substrate.

[0113] Various machine learning outputs are described herein. Specific numbers and arrangements of machine learning models are described and shown. However, it should be understood that the numbers and types of machine learning models used and the arrangement of such machine learning models can be modified to achieve the same or similar end results. Thus, the arrangement of machine learning models described and shown is merely an example and should not be construed as limiting.

[0114] In some embodiments, one or more machine learning models are trained to perform one or more substrate placement estimation tasks. Each task can be performed by a separate machine learning model. Alternatively, a single machine learning model can perform each task or a subset of tasks. For example, a first machine learning model can be trained to determine substrate placement (e.g., a position on a substrate support for substrate processing), and a second machine learning model can be trained to determine substrate switch offsets (e.g., substrate handler robot switch offsets). Additionally or alternatively, different machine learning models can be trained to perform different combinations of tasks. In an example, one or several machine learning models can be trained, where the trained machine learning (ML) model is a single shared neural network with multiple shared layers and multiple higher-level different output layers, where each output layer outputs a different prediction, classification, identification, etc. For example, a first higher-level output layer can determine substrate placement relative to a first type of chamber component (e.g., an electrostatic chuck), and a second higher-level output layer can determine substrate placement relative to a second type of chamber component (e.g., a process kit ring).

[0115] 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. Artificial neural networks generally include feature representation elements with classifier or regression layers that map features to an output space of a target. For example, a convolutional neural network (CNN) hosts multiple layers of convolutional filters. Pooling is performed at lower layers, and can address non-linearities, with multiple layers of perceptrons typically appended on top of the lower layers, mapping top-level features extracted by the convolutional layers to decisions (e.g., classification outputs). Deep learning is a class of machine learning algorithms that uses a series of nonlinear processing units in multiple layers to perform feature extraction and transformation. Each successive layer uses the output of the previous layer as input. Deep neural networks can learn in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. Deep neural networks include a hierarchy of 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 composite representation. It is noted that the deep learning process can learn on its own which features are best placed at which level. The “depth” in “deep learning” refers to the number of layers of data transformation. More precisely, deep learning systems have a fairly large credit assignment path (CAP) depth. The CAP is the chain of transformations from input to output. The CAP describes the potential causal relationships between inputs and outputs. For feedforward neural networks, the depth of the CAP can be the depth of the network, and can be the number of hidden layers plus one. For recurrent neural networks, where signals can propagate through layers more than once, the CAP depth is potentially infinite.

[0116] Training of a neural network can be achieved in a supervised learning fashion, involving feeding a training dataset consisting of labeled inputs through the network, observing its outputs, defining an error (by measuring the difference between the outputs and the label values), and adjusting the weights of the network at all its layers and nodes using techniques such as deep gradient descent and backpropagation to minimize the error. In many applications, repeating this process over many labeled inputs in the training dataset results in a network that can produce correct outputs when the inputs are different from those present in the training dataset.

[0117] For the model training workflow 405, a training dataset should be formed using a training dataset that includes profile maps 412 for hundreds, thousands, ten-thousands, hundreds of thousands, or more substrates. The data can include, for example, uniformity profiles determined using a given number of measurements, each measurement associated with a particular target location. This data can be processed to produce one or more training datasets 436 for training one or more machine learning models. The training data items in the training dataset 436 can include a profile map 412, a substrate placement for the substrate that was measured to produce the profile map, and / or one or more images of the substrate.

[0118] To implement the training, the processing logic inputs the training dataset 436 into one or more untrained machine learning models. The machine learning models can be initialized prior to inputting the first input into the machine learning model. The processing logic trains the untrained machine learning models based on the training dataset to produce one or more trained machine learning models that perform various operations as described above. The training can be implemented by inputting input data into the machine learning one at a time, the input data being, for example, one or more profile maps 412 (e.g., thickness profile maps, spectral profile maps, roughness profile maps, particle count profile maps, optical constant profile maps, etc.), images of the element, and / or lifetime information.

[0119] The machine learning model processes the input to produce an output. An artificial neural network includes an input layer composed of values in data points. The next layer is called a hidden layer, with each node of the hidden layer receiving one or more input values. Each node contains parameters (e.g., weights) that are applied to the input values. Thus, each node essentially inputs the input values into a multivariable function (e.g., a non-linear mathematical transformation) to produce an output value. The next layer can be another hidden layer or an output layer. In either case, the nodes of the next layer receive the output values from the nodes of the previous layer, and each node applies weights to these values and then produces its own output value. This can be performed at each layer. The last layer is an output layer, with one node for each class, prediction, and / or output that the machine learning model can produce.

[0120] Accordingly, the output can include one or more predictions or inferences (e.g., an estimate of a substrate placement in a processing chamber for processing a substrate in the processing chamber in which the measured substrate was processed). The processing logic can compare the estimated substrate placement of the output to historical substrate placements. The processing logic determines an error (i.e., a classification error) based on a difference between the estimated substrate placement and a target substrate placement. The processing logic adjusts the weights of one or more nodes in the machine learning model according to the error. An error term or delta can be determined for each node in the artificial neural network. According to this error, the artificial neural network adjusts one or more of its parameters (weights of one or more inputs to a node) of one or more of its nodes. The parameters can be updated in a backpropagation manner, such that nodes of a highest layer are updated first, then nodes of a next layer are updated, and so on. The artificial neural network includes multiple layers of “neurons,” where each layer receives values from neurons of a previous layer as inputs. The parameters of each neuron include weights associated with values received from each neuron of a previous layer. Accordingly, adjusting the parameters can include adjusting the weights assigned to each input of one or more neurons of one or more layers in the artificial neural network.

[0121] Once the model parameters are optimized, model validation can be performed to determine whether the model has improved and to determine a current accuracy of the deep learning model. After one or more rounds of training, the processing logic can determine whether a stopping criterion has been met. The stopping criterion can be a target accuracy level, a target number of processed images from the training data set, a target amount of parameter variation for one or more previous data points, a combination thereof, and / or other criteria. In one embodiment, the stopping criterion is met when at least a minimum number of data points have been processed and at least a threshold accuracy is reached. The threshold accuracy can be, for example, 70%, 40%, or 90% accuracy. In one embodiment, the stopping criterion is met if the accuracy of the machine learning model has stopped improving. If the stopping criterion has not been met, further training is performed. If the stopping criterion is met, the training can be complete. Once the machine learning model is trained, a reserved portion of the training data set can be used to test the model. Once one or more trained machine learning models 438 are generated, they can be stored in the model storage 445 and can be incorporated into the substrate placement engine 330.

[0122] For the model application workflow 417, according to one embodiment, input data 462 can be input to one or more substrate placement judges 467, each of which can include a trained neural network or other model. Additionally or alternatively, one or more substrate placement judges 467 can apply image processing algorithms to determine chamber component conditions. The input data can include a profile map (e.g., of a polymer layer on a substrate measured using an integrated reflectometry measurement device or substrate measurement system). The input data can additionally optionally include one or more images of the substrate. Based on the input data 462, the substrate placement judges 467 can output one or more estimated substrate placements 469. The estimated substrate placements 469 can include a substrate placement relative to a substrate support structure used to process a substrate within a processing chamber (e.g., optionally as an offset from a center of a substrate support such as an electrostatic chuck). In some examples, a coordinate system based on the substrate support (e.g., based on a center point of the substrate support) can be used to associate the estimated substrate placement (e.g., position) relative to the substrate support.

[0123] The action judges 472 can determine one or more actions 470 to be taken based on the substrate placements 469. In one embodiment, the action judges 472 compare the substrate placement estimates to one or more substrate placement thresholds. If one or more substrate placement estimates meet or exceed the substrate placement thresholds, the action judges 472 can determine to recommend updating the substrate placement for future substrate placements, and can output a recommendation or notification to update the substrate placement parameters. In some embodiments, the action judges 472 are to automatically update the substrate placement metrics based on substrate placements 469 that meet one or more criteria. In some examples, the substrate placements 469 can include an estimated position of a placed substrate relative to a substrate support (e.g., electrostatic chuck) and / or relative to a component of the substrate support (e.g., a process kit ring) for processing. The estimated position can be an optimized position of the substrate to minimize tilt in the processed substrate, particularly near the edges of the substrate. The substrate placements 469 can include a coordinate position relating a center point of the substrate support to a center point of the substrate. In some examples, the substrate placements 469 can reflect an offset between the center of the substrate and the center of the substrate support. In some embodiments, the offset can be used to determine one or more offsets for a robot used to manipulate the substrate (e.g., to update a position for the robot to place and / or“switch” the substrate, etc.).

[0124] Figure 5is a flowchart of a method 500 for generating a training dataset for training a machine learning model to perform substrate placement assessment according to aspects of the present disclosure. The method 500 is performed by processing logic that can comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computing system or a dedicated machine), firmware, or a combination thereof. In one implementation, the method 500 can be performed by a computer system architecture 100 such as Figure 1 In other or similar implementations, one or more operations of the method 500 can be performed by one or more other machines not depicted in the figures. In some aspects, one or more operations of the method 500 can be performed by the training set generator 324 of the server machine 320 or the server machine 350 described with respect to Figure 3

[0125] At block 510, the processing logic initializes a training set T to an empty set (e.g., {}). At block 512, the processing logic obtains substrate surface data associated with a substrate processed at a processing chamber of a manufacturing system (e.g., reflectometry data of a surface of a film on a substrate, such as a film thickness profile or wafer map).

[0126] At block 514, the processing logic obtains substrate placement information for a substrate processed by the processing chamber. As previously described, the substrate placement information can include coordinate positions of the substrate relative to a substrate support and / or relative to elements of the substrate support.

[0127] At block 516, the processing logic generates a training input based on the sensor data obtained for the substrate at block 512. In some embodiments, the training input can include a normalized sensor dataset (e.g., including surface reflectometry data as described herein).

[0128] At block 518, the processing logic can generate a target output based on the substrate placement information obtained at block 514. The target output can correspond to a substrate placement metric for the substrate processed in the processing chamber (data indicative of a placement of one or more substrates processed in the processing chamber).

[0129] At block 520, the processing logic generates an input / output mapping. The input / output mapping refers to a training input that includes or is based on data of a substrate, and a target output for the training input, where the target output identifies a substrate placement, and where the training input is associated with (or mapped to) the target output. At block 522, the processing logic adds the input / output mapping to the training set T.

[0130] ​At block 524, the processing logic determines whether the training set T includes a sufficient amount of training data to train the machine learning model. It should be noted that in some embodiments, the sufficiency of the training set T can be determined simply based on the number of input / output mappings in the training set, while in some other embodiments, the sufficiency of the training set T can be determined based on one or more other criteria in addition to or instead of the number of input / output mappings (e.g., a measure of diversity of the training examples, etc.). In response to determining that the training set T includes a sufficient amount of training data to train the machine learning model, the processing logic provides the training set T to train the machine learning model. In response to determining that the training set does not include a sufficient amount of training data to train the machine learning model, the method 500 returns to block 512.

[0131] At block 526, the processing logic provides the training set T to train the machine learning model. In some embodiments, the training set T is provided to the training engine 326 of the server machine 320 and / or the server machine 350 to perform the training. In the case of a neural network, for example, the input values of a given input / output mapping (e.g., the spectral data and / or chamber data of a previous substrate) are input to the neural network, and the output values of the input / output mapping are stored in the output nodes of the neural network. The connection weights in the neural network are then adjusted according to a learning algorithm (e.g., backpropagation, etc.), 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 a substrate placement (e.g., a substrate placement metric) for a substrate processed in the processing chamber.

[0132] Figure 6 is a flowchart illustrating an embodiment of a method 600 of training a machine learning model to estimate a substrate placement for processing a substrate in a processing chamber. The method 600 is performed by processing logic that can comprise hardware (e.g., circuitry, dedicated logic, etc.), software (such as is run on a general-purpose computer system or a dedicated machine), firmware, or a combination thereof. In one implementation, the method 600 can be performed by a computer system, such as the computer system architecture 100 of FIG. 1. Figure 1 In other or similar embodiments, one or more operations of the method 600 can be performed by one or more other machines not depicted in the figure. In some aspects, one or more operations of the method 600 can be performed by the server machine 320 or the training engine 326 of the server machine 350 described with respect to FIG. 2. Figure 3

[0133] ​At block 602 of the method 600, processing logic collects a training data set, which can include data from a plurality of substrate profiles (e.g., film thickness profiles indicating film thickness of a polymer film at a plurality of locations on a substrate, a substrate surface profile, a substrate thickness profile, etc.). Each data item of the training data set can include one or more labels. The data items in the training data set can include input-level (e.g., image-level) labels indicating a presence or absence of one or more substrate features associated with a substrate placement. For example, some data items can include a label of a tilt present in a processed substrate. In some embodiments, each data item includes a substrate map, which can be an image of a substrate (e.g., a thermal map of a substrate). Colors in the thermal map can indicate substrate thickness and / or other parameter values (e.g., such as tilt, etc.). Alternatively, actual thickness values can be used for each of a number of coordinates on a surface of a substrate (e.g., thickness values of a substrate).

[0134] At block 604, data items from the training data set are input into an untrained machine learning model. At block 606, the machine learning model is trained based on the training data set to produce a trained machine learning model that classifies or estimates one or more substrate placements for processing a substrate in a processing chamber. The machine learning model can also be trained to output one or more other types of predictions, coordinate-level classifications, decisions, etc.

[0135] In one embodiment, at block 610, an input of a training data item is input into the machine learning model. The input can include data from a substrate profile (e.g., a substrate surface profile) indicating one or more surface characteristics (e.g., thickness, optical constant, particle count, roughness, material characteristics, etc.) across a substrate. In embodiments, the data can be input as an image or as a feature vector. At block 612, the machine learning model processes the input to produce an output. The output can include one or more substrate placements (e.g., substrate placement values, etc.). The substrate placements can be suggested substrate placements for placing future substrates on a substrate support. The output can additionally or alternatively include one or more substrate switching offsets for a robot arm. For example, a switching offset can associate an initial robot switching orientation with an updated robot switching orientation to correspond to a predicted substrate placement for substrate processing.

[0136] At block 614, processing logic compares an output probability and / or value of a substrate placement metric to a known best substrate placement associated with the input. At block 616, processing logic determines an error based on a difference between the output and the known placement. At block 618, processing logic adjusts a weight of one or more nodes in the machine learning model based on the error.

[0137] At block 620, the processing logic determines whether the stopping criteria are met. If the stopping criteria have not been met, the method returns to block 610 and another training data item is input into 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.

[0138] In one embodiment, one or more ML models are trained for application across multiple processing chambers, which can be processing chambers of the same type or model. The trained ML model can then be further adjusted for a particular instance of a processing chamber. The further adjustment can be performed through the use of additional training data items that include surface profile maps of substrates processed by the processing chamber in question. Such adjustment can address chamber mismatches between chambers and / or some processing chamber's particular hardware processing suite. Additionally, in some embodiments, further training is performed to adjust the ML model for a processing chamber after maintenance is performed on the processing chamber and / or after changes are made to the hardware of the processing chamber.

[0139] Figure 7 A flowchart of a method 700 of determining a recommended substrate placement according to aspects of the present disclosure is shown. The method 700 is performed by processing logic that can comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computing system or a dedicated machine), firmware, or some combination thereof. In one implementation, the method 700 can be performed by a computer system architecture 100 such as Figure 1 In other or similar implementations, one or more operations of the method 700 can be performed by one or more other machines not depicted in the figures. In some aspects, one or more operations of the method 700 can be performed by the substrate placement engine 330 of the server machine 320 described with respect to Figure 3

[0140] At block 706, the processing logic (e.g., of a processing device) determines a center of a substrate support (e.g., electrostatic chuck) of a processing chamber of a substrate processing system (also referred to as a manufacturing system). In some embodiments, the center of the substrate support is determined from data collected by a multi-function wafer. For example, a multi-function wafer (e.g., a camera wafer) can collect an image of an electrostatic chuck in a processing chamber. The processing device can process the image to determine a center of the electrostatic chuck. The center of the substrate support can correspond to a reference point (e.g., (0, 0) point) of the substrate support.

[0141] ​At block 708, the substrate is aligned with a center of the substrate support. In some embodiments, processing logic causes a robotic arm (e.g., of a substrate handling robot) to place the substrate on the substrate support such that a center of the substrate is aligned with a center of the substrate support. In some embodiments, the substrate is centered on the substrate support to obtain baseline measurements regarding placement of the substrate. In some embodiments, the substrate is a proxy substrate (e.g., a test substrate, etc.).

[0142] At block 710, the processing chamber processes the substrate. For example, the processing chamber can perform an etch process to partially remove a film on a surface of the substrate, or can perform a film deposition process to deposit 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). The processing of the substrate can be performed according to a recipe (e.g., an etch process recipe, a deposition process recipe, etc.). During processing, the substrate can be supported by a substrate support (e.g., an electrostatic chuck) of the processing chamber. The substrate can be placed in an initial position in the processing chamber prior to processing (e.g., at block 708) and can be held in the initial position during processing. In some embodiments, after processing, the substrate includes a surface profile (e.g., a thickness profile, etc.).

[0143] At block 712, one or more robots transfer the substrate from the processing chamber to the substrate measurement system. If the substrate measurement system is connected to or included in a transfer chamber, a transfer chamber robot can remove the substrate from the processing chamber and insert the substrate into the substrate measurement system. If the substrate measurement system is connected to or included in a factory interface, a transfer chamber robot can remove the substrate from the processing chamber and place the substrate in a load lock. A factory interface robot can then remove the substrate from the load lock and insert the substrate into the substrate measurement system. The substrate measurement system can be any of the aforementioned substrate measurement systems, such as an integrated reflectometer (IR) device.

[0144] At block 714, the substrate measurement system generates measurements for a number of locations on a surface of the substrate. Each location can have a unique set of coordinates.

[0145] At block 716, the substrate measurement system and / or the computing device can generate a profile map of the surface profile of the substrate (e.g., a substrate surface profile map) based on the measurements of the substrate measurement system. The profile map can be or include an image, where each pixel in the image corresponds to a coordinate on the substrate. Each pixel can have an intensity value corresponding to a measurement value (e.g., a thickness value) at the coordinate of the substrate associated with that pixel. In some embodiments, the profile map indicates a thickness profile, which indicates an etch rate profile. For example, the profile map can reflect a thickness profile of a substrate that has experienced an etch process for a predetermined amount of time. In some embodiments, the profile map indicates a substrate surface defect, such as a tilt. In some embodiments, the tilt is related to the etch rate. In some embodiments, the profile can be or include a feature vector, where each entry in the feature vector is associated with a coordinate of the substrate, and where each entry can have a value representing a value of the surface profile at the coordinate of the substrate. In some embodiments, processing logic (e.g., of the measurement system and / or of the computing device) can determine an etch rate profile map corresponding to an etch rate proximate to an edge of the substrate.

[0146] At block 718, the computing device processes data from the profile map (e.g., a thickness profile map, a particle map, an optical constant map, a roughness map, an etch rate profile map, etc.) using a model. In some embodiments, the computing device processes data from the profile map using one or more trained machine learning models. In some embodiments, the trained machine learning models are trained using data collected from a plurality of substrates processed according to a recipe. For example, a set of substrates are processed according to a recipe at various placements (e.g., positions) on a substrate support. Profile maps of the processed substrates can be generated. The profile maps and corresponding substrate placements can be input as training data to train the machine learning models.

[0147] In some embodiments, the model is a physics-based model or a statistical model. In some embodiments, the model outputs an estimated substrate placement value (e.g., an estimated substrate placement metric) for placing a substrate relative to one or more elements of a substrate support (e.g., an electrostatic chuck). In some examples, the model outputs a placement metric (e.g., a coordinate position of a substrate on a substrate support) corresponding to a placement of a substrate relative to a substrate support (e.g., an electrostatic chuck) and / or one or more elements of the substrate support (e.g., a process kit ring). In some embodiments, the output of the model is based on the etch rate profile map determined at block 716. In some embodiments, the output of the model indicates that the substrate is not placed at an optimal position for processing. In one embodiment, instead of or in addition to using a trained ML model, one or more computer vision algorithms are used to process the profile map.

[0148] At block 720, processing logic (e.g., of the computing device) determines a suggested placement of the substrate on the substrate support based on the estimated placement value (e.g., output by the model at block 718). In some examples, the computing device can use the estimated substrate placement value to determine a position on the substrate support for placing the substrate. Specifically, at block 718, the computing device can determine a coordinate position based on the value (or values) output from the model. In some embodiments, the computing device can determine a gap that exists between an edge of the substrate and an inner diameter of the process kit ring. In some embodiments, the gap around the substrate is not uniform. For example, the gap on one side of the substrate can be larger than the gap on an opposite side of the substrate. Thus, the gap can be related to the placement of the substrate (e.g., on the electrostatic chuck, within the inner diameter of the process kit ring, etc.). The processing logic can then place another substrate in the processing chamber according to the suggested placement (e.g., via a substrate handling robot), as described below.

[0149] Figure 8 A flowchart of a method 800 to compare a second estimated substrate placement to a first estimated substrate placement according to aspects of the disclosure is shown. In some embodiments, the method 800 is performed in conjunction with the method 700 described above. For example, the method 700 can correspond to processing a first substrate, while the method 800 can correspond to processing a second substrate and comparing results of the second substrate to results of the first substrate. The method 800 is performed by processing logic that can comprise hardware (circuitry, dedicated logic, etc.), software (such as is run on a general purpose computing device or a dedicated machine), firmware, or some combination thereof. In one implementation, the method 800 is performed by a computer system architecture 100, such as Figure 1 In other or similar implementations, one or more operations of the method 800 can be performed by one or more other machines not depicted in the figures. In some aspects, one or more operations of the method 800 can be performed by the substrate placement engine 330 of the server machine 320 described with respect to Figure 3

[0150] ​At block 808, a substrate (e.g., a second substrate) is placed in the processing chamber according to the suggested placement (e.g., determined at block 720 of method 700). In some embodiments, the processing logic causes a robot to place the substrate on a substrate support (e.g., an electrostatic chuck) within the processing chamber. According to the suggested placement, the robot can place the substrate at a position such that the center of the substrate is offset from the center of the substrate support. For example, the suggested placement can specify that the center of the substrate is to be offset from the center of the electrostatic chuck by a specified amount in a specified direction. The amount and direction of the offset can be a difference between the first robot switch orientation and the second robot switch orientation. In some examples, the first robot switch orientation is a baseline (e.g., a preset, etc.) orientation and the second robot switch orientation is an updated orientation based on the offset. The robot can place the substrate on the electrostatic chuck at the specified offset amount from the center of the electrostatic chuck in the offset direction.

[0151] At block 810, the substrate is processed in the processing chamber according to a recipe (e.g., the recipe of block 710 of method 700). The recipe can be an etch processing recipe and / or a deposition processing recipe. Similar to the substrate processed at block 710 of method 700, the substrate includes a processed surface profile. The surface profile can be a thickness profile.

[0152] At block 812, the substrate is transferred (e.g., via one or more transfer robots) from the processing chamber to a substrate measurement system (e.g., similar to at block 712 of method 700).

[0153] At block 814, the substrate measurement system measures a surface of the substrate (e.g., similar to at block 714 of method 700).

[0154] At block 816, the substrate measurement system and / or a computing device can generate a profile map of the surface profile of the substrate based on the measurements of the substrate measurement system (e.g., similar to at block 716 of method 700). In some embodiments, the profile can be an image or can include an image. The profile map can indicate an etch rate profile.

[0155] At block 818, the computing device processes data from the profile map (e.g., a thickness profile map, a grain map, an optical constant map, a roughness map, etc.) using a model (e.g., a trained machine learning model, a physics-based model, a statistical model, etc.). In some embodiments, the model outputs an estimated substrate placement value (e.g., an estimated substrate placement metric, etc.).

[0156] At block 820, processing logic (e.g., of a computing device) compares the estimated substrate placement value output from the model at block 818 to another estimated substrate placement value (e.g., the estimated substrate placement value of block 718 of method 700). The processing logic can determine whether to update the suggested placement based on the comparison. For example, in response to the estimated substrate placement values being the same (e.g., substantially the same, within a threshold difference of each other, etc.), the processing logic can determine that the suggested placement is appropriate. In response to the estimated substrate placement values being different (e.g., outside a threshold difference of each other, etc.), the processing logic can determine that the suggested placement should be updated.

[0157] At block 822, the processing logic updates the suggested placement based on the comparison at block 820.

[0158] Figure 9 A profile map 900 of a processed substrate according to aspects of the disclosure. The profile map 900 is a heat map showing temperatures at different locations on the substrate during processing in a processing chamber. The temperatures can be based on thicknesses of different locations of a film deposited or etched during processing. A key 902 is provided to show how to interpret the profile map 900. As shown, hot spots 905 are uneven around the edge of the substrate. In some embodiments, the methods described herein can provide a substrate without hot spots (e.g., substantially without hot spots, substantially uniform temperatures, etc.). Uneven hot spots can indicate an increase in etch rate, which is related to an increase in substrate tilt. During processing, on the sides near the hot spots 905, the substrate can be too close to a processing kit ring (e.g., of a substrate support in a processing chamber). For example, at near the hot spots 905, the gap between the edge of the substrate and the inner diameter of the processing kit ring can be too small or too large. The uneven hot spots 905 around the edge of the substrate can indicate that there is excessive tilt in features of the processed substrate near the edge of the substrate. Thus, according to some embodiments described herein, the placement of the substrate for processing should be changed.

[0159] Figures 10A-10B A flowchart of a method of determining an optimal substrate placement according to aspects of the disclosure. Figure 10A A method 1000A for determining an optimal substrate placement is shown, according to some embodiments. Figure 10B A method 1000B for determining an optimal substrate placement is shown, according to some embodiments. In some embodiments, the method 1000A and / or 1000B can be a method for performing an all-angle fit (AAF) algorithm to suggest wafer placement adjustments. In some embodiments, the method 1000A and / or 1000B is used to optimize etch profile uniformity, particularly near the outermost edges of the substrate.

[0160] Referring to Figure 10AAt 1002, a design of experiment (DOE) is performed. In some embodiments, a plurality of substrates (e.g., test substrates) are processed (e.g., etched) at different locations on a substrate support in a processing chamber. In some embodiments, a first substrate is placed at a first placement location (e.g., a first position) on the substrate support. The first substrate can be processed, and the processed first substrate can have a first surface profile. In some embodiments, the first surface profile is measured using the substrate measurement system described above. In some embodiments, a second substrate is placed at a second placement location on the substrate support that is different from the first placement location. The second substrate is then processed, and a surface profile of the processed second substrate is measured. A third substrate can be processed at a third placement location, and a surface profile of the processed third substrate can be measured subsequently.

[0161] Referring to Figure 11A An example plot 1100A showing substrate placement for a DOE is shown. Referring to Figure 13A An example plot 1300A showing substrate placement in a DOE is shown. In some embodiments, plot 1100A can correspond to plot 1300A. In some embodiments, substrates can be placed at different locations on a substrate support to perform a DOE. Again referring to Figure 11A In some embodiments, a substrate is placed and processed at each placement location 1102-1118. In some embodiments, location 1118 can correspond to a center of the substrate support. In some embodiments, a first substrate can be placed and processed at first location 1102, a second substrate can be placed and processed at second location 1104, a third substrate can be placed and processed at third location 1106, and so on. Although Figure 11A Nine possible placement locations for processing substrates are shown, but a DOE can be performed with fewer than nine processed substrates at different placement locations. In some embodiments, no more than two placement locations can lie on the same line in a two-dimensional space. In some embodiments, at least one placement location of the DOE deviates from a straight line formed between two other placement locations. In some embodiments, a DOE can be performed with more than nine processed substrates at different placement locations. Including a greater number of processed substrates at different placement locations can improve the accuracy of the DOE. Higher accuracy can be obtained if the DOE placement locations cover more azimuthal angles. For example, Figure 13A The placement locations shown form an “X” pattern. Adding additional placement locations to the DOE to form another “X” pattern at different angles between the “X” legs (e.g., shallower angles or deeper angles) can make the prediction more accurate. In some embodiments, a DOE can be performed with as few as three processed substrates at three different placement locations.

[0162] Again referring toFigure 10A Data processing is performed at block 1004. In some embodiments, an etch rate is determined at a location near the edge of each substrate processed by DOE (e.g., at block 1002). For example, the etch rate can be determined by subtracting the processed thickness of the substrate from the unprocessed thickness and then dividing by the processing time. In some embodiments, the etch rate is determined at locations with multiple azimuth angles at radial positions adjacent to the edge of the first processed substrate. (Refer to...) Figure 11B The diagram illustrates a radial plot of the substrate etch rate relative to an azimuth angle θ. In some embodiments, the etch rate is determined for all azimuth angles surrounding the center of the treated substrate. In some embodiments, the etch rate is determined for several azimuth angles surrounding the center of the treated substrate, such that the determined etch rate represents the entire etch rate profile. The determined etch rate may represent only all azimuth angles. For example, the etch rate may be determined at azimuth angles such as 0°, 15°, 30°, and 45° around the treated substrate near the substrate edge. However, it may be determined in relation to... Figure 11B The etching rate can be determined at various azimuth angles with different scales as shown. For example, the etching rate can be determined at 0°, 10°, 20°, 30°, etc., or 0°, 30°, 60°, 95°, etc. Figure 11B The etching rates of the treated substrate at various azimuth angles around the center of the substrate near the substrate edge are shown.

[0163] Refer again Figure 10A In block 1004A, the etch rate can be normalized. Normalizing the etch rate can help account for differences in etching chamber conditions. Normalizing the etch rate can also help account for substrate differences, such as slight differences in composition or defects. Substrate differences can lead to inconsistencies in substrate processing, which in turn can result in inconsistent data. Normalizing the etch rate can help eliminate the effects of remote defects in the substrate or defects and / or variations in the processing from the dataset. In some embodiments, the etch rate is normalized using an average etch rate. In some examples, for a first etch rate profile (e.g., ...), Figure 11B or Figure 12A The etch rate profile shown is used to determine a first average etch rate. Each value of the first etch rate profile is divided by the first average etch rate to determine a first normalized etch rate profile. In some similar examples, for a second etch rate profile, a second average etch rate is determined. Each value of the second etch rate profile is divided by the second average etch rate to determine a second normalized etch rate profile. Other methods for normalizing etch rate data can be used. See below. Figure 10B Let's discuss more details about data normalization.

[0164] Reference Figure 12A, an example plot of substrate etch rate versus azimuthal angle Θ is shown. An etch rate profile 1202A corresponding to the etch rate of a first substrate proximate to a first substrate edge is shown. An etch rate profile 1204A corresponding to the etch rate of a second substrate proximate to a second substrate edge is shown. Referring to Figure 12B , an example plot of normalized substrate etch rate versus azimuthal angle Θ is shown. A normalized etch rate profile 1202B can correspond to the etch rate profile 1202A of Figure 12A . A normalized etch rate profile 1204B can correspond to the etch rate profile 1204A of Figure 12A . With the normalized etch rate profiles, etch rate values for discrete azimuthal angles (e.g., discrete values of Θ) can be compared by eliminating at least some inconsistencies between the processed substrates from the data.

[0165] At block 1004B, a linear fit is performed using respective values of normalized etch rate of the substrates processed as part of the DOE (e.g., normalized etch rate values of the processed substrates at the same azimuthal angle). Referring to Figure 13B , an example graphical representation 1300B of normalized substrate etch rate of substrates processed at different placement positions is shown. For a particular azimuthal angle 1354 (45° as shown in Figure 13B ), the normalized etch rate is retrieved from the normalized etch rate profile 1352 of each processed substrate. Referring to Figure 13C , a plot 1300C of the linear fit of normalized substrate etch rate is shown. The normalized etch rate 1372 of each processed substrate at the azimuthal angle 1354 is plotted relative to different positions along the vector 1310 of Figure 13A . The vector 1310 in Figure 13A will be described in detail below. A linear fit 1380 is formed by the plotted normalized etch rates 1372. The linear fit 1380 can represent a predicted etch rate of a processed substrate at the different positions of the respective azimuthal angle along the vector 1310. The linear fit of normalized etch rate can be determined at each azimuthal angle.

[0166] Referring again to Figure 10AAt block 1006, an optimal substrate placement position is determined using the linear fit data determined at block 1004B. In some embodiments, the linear fit data can be used to predict the etch rate profile for any substrate placement. By predicting the etch rate profile for different placement positions, an optimal position can be determined that satisfies one or more metrics. For example, a placement position that minimizes the etch rate range (e.g., the etch rate range near the edge of the substrate) or the etch rate standard deviation (e.g., the etch rate standard deviation near the edge of the substrate) can be determined. Minimizing the etch rate range and / or the etch rate standard deviation can result in a more consistent processed substrate, particularly near the edge of the substrate. Once the optimal substrate placement position is determined, the optimal placement position is suggested to the substrate processing system. The substrate can be placed on a substrate support in the processing chamber according to the suggested placement.

[0167] Referring to Figure 10B , a method 1000B for determining an optimal substrate placement for processing in a processing chamber is shown. The method 1000B is described with reference to an etch process, but can also be used for performing a deposition process. Thus, any description regarding etching, etch rate, etch rate profile, etc. also applies to deposition, deposition rate, deposition rate profile, etc. The method 1000B can also be applied to any other metric that exhibits a linear or higher order response to substrate placement at a particular azimuthal angle. For example, the etch rate gradient across a substrate (e.g., the difference in etch rate for a 146 mm radius and a 148 mm radius) and the etch tilt profile of a patterned substrate can also exhibit a similar response to substrate placement. At block 1010, a substrate is placed at a placement position on a substrate support. The substrate can be a test substrate that is processed as part of a DOE. The substrate can be placed at one of the positions 1102-1118 shown in Figure 11A At block 1020, an etch operation can be performed on the substrate to remove material from the substrate (or a deposition operation can be performed to add material to the substrate). At block 1030, using a substrate measurement system, the etch rate or deposition rate can be measured near the edge of the substrate at a radial distance from the center of the substrate at a plurality of azimuthal angles. In some examples, the etch rate or deposition rate is measured around the center axis of the substrate at a distance of about 140 mm and 150 mm from the center of the substrate. The etch rate or deposition rate measured at the various azimuthal angles can be used to construct an etch rate profile or deposition rate profile (e.g., as shown in Figure 11B for a single substrate, as shown in Figure 12A for two substrates). At block 1040, the etch rate profile or deposition rate profile is normalized. In some embodiments, the etch rate profile or deposition rate profile is normalized using the average etch rate of the etch rate profile or the average deposition rate of the deposition rate profile. For example, with reference to Figure 12AFor the etch rate profile 1202A, an average etch rate from an azimuthal angle of 0° to an azimuthal angle of 360° can be determined. Each etch rate value on the etch rate profile can be divided by the average etch rate to calculate a normalized etch rate profile 1202B of Figure 12B .

[0168] Referring again to Figure 10B , blocks 1010-1040 can be repeated a number of times for a number of substrates processed as part of the DOE. Each time block 1010 is repeated, an associated substrate is placed at a different placement position on the substrate support. In some embodiments, blocks 1010-1040 are repeated nine times to process nine test substrates at nine different placement positions. In some embodiments, blocks 1010-1040 are repeated as few as three times to process as few as three test substrates at as few as three different placement positions. Including more or fewer substrates in the DOE (and thus repeating blocks 1010-1040 more or fewer times) depends on the precision threshold of the DOE. For example, if the precision threshold is higher, more test substrates are processed. Similarly, if the precision threshold is lower, fewer test substrates can be processed.

[0169] At block 1050, the normalized etch rate profile of each substrate processed at blocks 1010-1040 (e.g., during each repetition of blocks 1010-1040) at various substrate placements is compiled. Referring to Figure 13B , a graphical representation 1300B of the normalized etch rate profiles is shown. The normalized etch rate profile 1352 can represent the normalized etch rate profile of a substrate processed at each placement position 1102-1118 shown in Figure 11A and / or each placement position 1302 shown in Figure 13A . Referring again to Figure 10B , at block 1055, using the normalized etch rate profiles compiled at block 1050, a normalized etch rate for each processed substrate at each azimuthal angle is determined. Referring to Figure 13B , the normalized etch rate is determined from the corresponding normalized etch rate profile 1352 at an azimuthal angle 1354. The azimuthal angle 1354 is shown as 45°. However, in some embodiments, depending on the density of the substrate etch rate sampling map, several normalized etch rates for each processed substrate are determined from the normalized etch rate profiles 1352 at all azimuthal angles.

[0170] Referring again to Figure 10B , at block 1060, a linear fit is performed for each azimuthal angle for which a normalized etch rate is determined at block 1055. The linear fit can be performed using various linear and non-linear higher order fitting methods known to those skilled in the art. Referring to Figure 13A, an example plot 1300A of substrate placement in a DOE is shown. Substrates can be processed at each substrate placement location 1302 as placed on a substrate support. A vector 1310 can correspond to Figure 13B the azimuth angle 1354. The angle of the vector 1310 with the X-axis can be equal to the value of the azimuth angle 1354. The placement locations 1302 can be projected onto the vector 1310, and the position of each projected placement location on the vector 1310 is converted to a value y’. As shown, y’ is at a 45° angle with the X-axis of the plot 1300A, and the corresponding angle Θ at which the azimuth angle 1354 lies is likewise 45°. Referring to Figure 13C , a normalized etch rate 1372 at the azimuth angle 1354 is shown with respect to dy’. The normalized etch rate 1372 is plotted for each processed substrate. A linear fit 1380 is calculated to fit the plotted normalized etch rates 1372. The process of projecting the placement locations 1302 onto the vector 1310, converting the position of each projected placement location to y’ on the vector 1310, and plotting the normalized etch rate 1372 with respect to dy’ can be repeated for each data corresponding to each of several azimuth angles.

[0171] Referring again to Figure 10B , at block 1070, a predicted etch rate profile is calculated for the estimated substrate placement location using the one or more linear fits calculated at block 1060. For example, using the etch rate data and / or the linear fits, an estimated location of a substrate that can produce a substrate that meets one or more threshold criteria can be determined. The threshold criteria can include a threshold etch rate profile range, a threshold etch rate profile minimum, a threshold etch rate profile maximum, and / or a threshold etch rate profile standard deviation. Using the linear fit 1380 corresponding to a particular azimuth angle, a predicted value of the etch rate at the particular azimuth angle can be calculated for the estimated substrate placement location. Once the predicted etch rates for all azimuth angles are determined, the predicted etch rates can be combined to form a predicted etch rate profile. At block 1080, an optimal substrate placement location that produces an optimal etch rate profile is determined. In some embodiments, the processing logic searches all substrate placement locations to determine which location has the optimal etch rate profile (e.g., that meets the one or more threshold criteria described above). The optimal placement location can be determined through experimentation. In some embodiments, a machine learning model described herein can be used to determine the optimal placement location. The machine learning model can be trained with historical data (e.g., historical etch rate profiles and / or historical substrate placement locations) to determine the optimal placement to produce a substrate with an optimal etch rate profile. At block 1090, the substrate can be placed at the optimal placement location for processing.

[0172] Referring again to Figure 12AAccording to aspects of the present disclosure, an example plot 1200A of substrate etch rate versus azimuthal angle Θ is shown. In some embodiments, the etch rate profile 1202A is an etch rate profile of a substrate processed at an optimal position on a substrate support. The optimal position can be determined using one or more methods described herein. In contrast, the etch rate profile 1204A is an etch rate profile of a substrate processed at a non-optimal position on a substrate support.

[0173] Referring to Figure 12B According to aspects of the present disclosure, an example plot 1200B of normalized etch rate versus azimuthal angle Θ is shown. In some embodiments, the normalized etch rate profile 1202B is a normalized etch rate profile of a substrate processed at an optimal position on a substrate support. In contrast, the normalized etch rate profile 1204B is a normalized etch rate profile of a substrate processed at a non-optimal position on a substrate support. As shown in plot 1200B, the normalized etch rate profile 1202B has a more consistent normalized etch rate at all azimuthal angles Θ. Thus, the etch rate of a substrate processed at an optimal position is more consistent, and a substrate processed at an optimal position will yield a better substrate than a substrate processed at a non-optimal position.

[0174] Figure 14 is a flowchart of a method 1400 of determining an optimal substrate placement according to aspects of the present disclosure. The method 1400 can be performed by processing logic that comprises hardware (e.g., circuitry, dedicated logic, etc.), software (such as is run on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one implementation, the method 1400 can be performed by a computer system, such as the computer system architecture 100 of FIG. 1. Figure 1 In other or similar implementations, one or more operations of the method 1400 can be performed by one or more other machines not shown in the figure.

[0175] At block 1410, a first substrate is processed in a processing chamber. For example, the processing chamber can perform an etch process to partially remove a film on a surface of the substrate, or perform a film deposition process to deposit 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). The processing of the substrate can be performed according to a recipe (e.g., an etch process recipe, a deposition process recipe, etc.). During processing, the substrate can be supported by a substrate support (e.g., an electrostatic chuck) of the processing chamber. The substrate can be placed at an initial position in the processing chamber prior to processing, and can be held at the initial position during processing. After processing, in some embodiments, the substrate includes a surface profile (e.g., a thickness profile, etc.).

[0176] At block 1412, one or more robots transfer the substrate from the processing chamber to the substrate measurement system. If the substrate measurement system is connected to or included in a transfer chamber, the transfer chamber robot can remove the substrate from the processing chamber and insert the substrate into the substrate measurement system. If the substrate measurement system is connected to or included in a factory interface, the transfer chamber robot can remove the substrate from the processing chamber and place the substrate in a load lock. Then, a factory interface robot can remove the substrate from the load lock and insert the substrate into the substrate measurement system. The substrate measurement system can be any of the substrate measurement systems described above, such as an integrated reflectometer (IR) device.

[0177] At block 1414, the substrate measurement system generates measurements for a number of locations on the surface of the substrate. Each location can have a unique set of coordinates.

[0178] At block 1416, the substrate measurement system and / or a computing device can generate a profile map of the surface profile of the substrate (e.g., a substrate surface profile map) based on the measurements of the substrate measurement system. The profile map can be or include an image, where each pixel in the image corresponds to a coordinate on the substrate. Each pixel can have an intensity value that corresponds to a measurement value (e.g., a thickness value) at the coordinate of the substrate associated with the pixel. In some embodiments, the profile map indicates a thickness profile, which indicates an etch rate profile. For example, the profile map can reflect a thickness profile of a substrate that has undergone an etch process for a predetermined amount of time. In some embodiments, the profile map indicates a substrate surface defect, such as a tilt. In some embodiments, the tilt is related to the etch rate. In some embodiments, the profile can be or include a feature vector, where each entry in the feature vector is associated with a coordinate of the substrate, and where each entry can have a value that represents a value of the surface profile at the coordinate of the substrate. In some embodiments, processing logic (e.g., of the measurement system and / or the computing device) can determine an etch rate profile that corresponds to the etch rate near the edge of the substrate.

[0179] At block 1418, the computing device can determine a plurality of etch rates that correspond to a plurality of locations on the substrate. The computing device can use the surface profile map generated at block 1416 to determine an etch rate profile of the etch rate near the edge of the substrate at several azimuthal angles around the center of the substrate.

[0180] At block 1420, processing logic processes data associated with the plurality of etch rates determined at block 1418. In some embodiments, the computing device processes the data using a model. The model can include a trained machine learning model, a mathematical model, a linear fit model, and / or a statistical model. The model can output one or more estimated surface profiles associated with one or more estimated placement positions on the substrate support. For example, the model can use a numerical method to estimate a predicted surface profile of a substrate processed at the estimated placement position. In some embodiments, the model normalizes the etch rate profile (e.g., determined at block 1418) and determines a normalized etch rate for various values of the azimuth angle Θ. In some embodiments, the model performs a linear fit using the normalized etch rates from several processed substrates for each value of the azimuth angle Θ. Using the linear fit, the model can determine one or more estimated placement positions on the substrate support to process a substrate having the best etch rate profile.

[0181] At block 1422, processing logic (e.g., of the computing device) determines a suggested placement of a substrate on the substrate support based on the one or more estimated placement positions (e.g., output by the model at block 1420). In some examples, the computing device can use the estimated placement positions to determine a position on the substrate support for placing a substrate. In particular, the computing device can determine a coordinate position based on the value (or values) output from the model at block 1420. Processing logic can then place another substrate in the processing chamber according to the suggested placement described herein (e.g., by a substrate handling robot).

[0182] Figure 15An illustrative representation of a machine in example form is depicted which is a computing device 1500 in which an instruction set can be executed for causing the machine to perform any one or more of the methodologies discussed herein. In alternative embodiments, the machine can be connected (e.g., networked) to other machines in a Local Area Network (LAN), an intranet, an extranet, or the Internet. The machine can operate in the capacity of a server or a client machine in client-server network environment, or as a peer machine (peer-to-peer (or distributed) network environment). The machine can be a personal computer (PC), a tablet computer, a set-top box (STB), a Personal Digital Assistant (PDA), a cellular telephone, a web appliance, a server, a network router, switch or bridge, or any other machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term "machine" shall also be taken to include any collection of machines (e.g., computers) that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein. In an embodiment, the computing device 1500 can correspond to one or more of the server machine 170, the server machine 180, the prediction server 112, the system controller 228, the server machine 320, or the server machine 350 as described herein.

[0183] The example computing device 1500 includes a processing device 1502, a main memory 1504 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM), etc.), a static memory 1506 (e.g., flash memory, static random access memory (SRAM), etc.), and a secondary memory (e.g., a data storage device 1528), which communicate with each other via a bus 1508.

[0184] The processing device 1502 can be representative of one or more general -purpose processors such as microprocessors, central processing units, or the like. More particularly, the processing device 1502 can be a complex instruction set computing (CISC) microprocessor, reduced instruction set computing (RISC) microprocessor, very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or processors implementing a combination of instruction sets. The processing device 1502 can also be one or more special-purpose processing devices such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), network processor, or the like. The processing device 1502 can also be or include a system on a chip (SoC), a programmable logic controller (PLC), or other types of processing devices. The processing device 1502 is configured to execute processing logic for performing the operations discussed herein.

[0185] The computing device 1500 can further include a network interface device 1522 for communicating with a network 1564. The computing device 1500 can also include a video display unit 1510 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1512 (e.g., a keyboard), a cursor control device 1514 (e.g., a mouse), and a signal generation device 1520 (e.g., a speaker).

[0186] The data storage device 1528 can include a machine-readable storage medium (or more specifically a non-transitory computer-readable storage medium) 1524 on which is stored one or more sets of instructions 1526 embodying any one or more of the methodologies or functions described herein. For example, the instructions 1526 can include instructions for the substrate placement engine 330. A non-transitory storage medium is a storage medium that is not a carrier wave. The instructions 1526 can also reside completely, or at least partially, within the main memory 1504 and / or within the processing device 1502 during execution thereof by the computing device 1500, the main memory 1504 and the processing device 1502 also constituting computer-readable storage media.

[0187] While the computer-readable storage medium 1524 is shown in an example embodiment to be a single medium, the phrase “computer-readable storage medium” should be taken to include a single medium or multiple media (e.g., a central or distributed database, and / or associated caches and servers) that store the one or more sets of instructions. The phrase “computer-readable storage medium” shall also be taken to include any medium that is capable of storing or encoding a set of instructions for execution by a machine and that cause the machine to perform any one or more of the methodologies of the present disclosure. The phrase “computer-readable storage medium” shall accordingly be taken to include, but not be limited to, solid-state memories, and optical and magnetic media.

[0188] The foregoing description sets forth numerous specific details, such as examples of specific systems, elements, methods, etc., in order to provide a thorough understanding of numerous embodiments of the present disclosure. However, it will be apparent to one skilled in the art that at least some embodiments of the present disclosure can be practiced without these specific details. In other instances, well-known components or methods are not described in detail or are presented in simple block diagram format in order to avoid unnecessarily obscuring the present disclosure. Thus, the specific details set forth above are merely exemplary. Particular implementations can vary from these exemplary details and still be contemplated to be within the scope of the present disclosure.

[0189] Throughout the specification, and in each of the claims, "a" or "an" can mean one or more, depending upon the context in which it is used. Similarly, the singular word "the" can mean the singular or the plural, or both, depending upon the context in which it is used. Throughout the specification, unless otherwise indicated, the word "comprising" is synonymous with the word "including," and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps. Throughout the specification, unless otherwise indicated, the word "coupled" means directly or indirectly connected, and can include wired or wireless connection. Throughout the specification, unless otherwise indicated, the word "substantially" means largely but not necessarily wholly what is specified, and includes instances of approximately, nearly or almost, as understood by one of ordinary skill in the art. Throughout the specification, unless otherwise indicated, the word "about" means approximately, nearly, or almost, as understood by one of ordinary skill in the art. Throughout the specification, unless otherwise indicated, the word "or" means any one of the words is inclusive, and not the exclusive "or" as understood by one of ordinary skill in the art.

[0190] Although the operations of the methods herein are shown and described in a particular order, the order of the operations can be altered, so that certain operations can be performed in an inverse order, or so that certain operations can be performed, at least in part, concurrently with other operations. In another embodiment, instructions or sub-operations of distinct operations can be in an intermittent and / or alternating manner.

[0191] It is to be understood that the above description is intended to be illustrative, and not restrictive. Many other embodiments will be apparent to those of skill in the art upon reading and understanding the above description. The scope of the disclosure should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. A method comprising: processing a first substrate in a process chamber of a substrate processing system according to a recipe while the first substrate is supported by a substrate support of the process chamber, wherein the first substrate comprises a first surface profile after the processing; generating a first profile map of the first surface profile of the first substrate using a substrate measurement system of the substrate processing system; processing data from the first profile map using a model, wherein the model outputs a first estimated substrate placement value of the first substrate relative to a placement of one or more elements of the substrate support; and determining a suggested placement of a substrate on the substrate support based on the first estimated substrate placement value.

2. The method of claim 1, further comprising: causing a second substrate to be placed in the process chamber according to the suggested placement, wherein the one or more elements of the substrate support comprise a process kit ring, and wherein the second substrate is positioned within an inner diameter of the process kit ring according to the suggested placement.

3. The method of claim 2, further comprising: processing the second substrate in the process chamber according to the recipe, wherein the second substrate comprises a second surface profile after the processing; generating a second profile map of the second surface profile using the substrate measurement system of the substrate processing system; processing data of the second profile map using the model, wherein the model outputs a second estimated substrate placement value; comparing the second estimated substrate placement value to the first estimated substrate placement value; and updating the suggested placement based on the comparison.

4. The method of claim 1, further comprising: determining a center of the substrate support; and aligning the first substrate with the center of the substrate support prior to the processing of the first substrate.

5. The method of claim 4, wherein a center of the substrate is offset from the center of the substrate support for the suggested placement of the substrate.

6. The method of claim 1, further comprising: determining a substrate handoff offset based on the suggested placement, wherein the substrate handoff offset is an offset of a mechanical arm from a first robot handoff orientation to a second robot handoff orientation.

7. The method of claim 1, wherein the first surface profile comprises a first thickness profile.

8. The method of claim 1, further comprising: determining a first etch rate profile of the first substrate for an etch rate profile proximate to an edge of the first substrate based on the first profile map, wherein the model outputs the first estimated substrate placement value based on the first etch rate profile.

9. The method of claim 1, wherein the model comprises at least one of a trained machine learning model, a physics-based model, or a statistical model.

10. The method of claim 1, wherein the model comprises a trained machine learning model, the method further comprising: ​ training a machine learning model to produce the trained machine learning model, wherein the machine learning model is trained using data of a plurality of processed substrates processed according to the recipe.

11. A system comprising: a process chamber; a substrate measurement tool; a memory; and a processing device coupled to the memory, the processing device to perform: causing a first substrate to be processed in the process chamber according to a recipe while the first substrate is supported by a substrate support of the process chamber, wherein the first substrate comprises a first surface profile after the processing; generating a first profile map of the first surface profile of the first substrate using the substrate measurement tool; processing data of the first profile map using a model, wherein the model outputs a first estimated substrate placement value of a placement of the first substrate relative to one or more elements of the substrate support; and determining a suggested placement of a substrate on the substrate support based on the first estimated substrate placement value.

12. The system of claim 11, wherein the processing device is further to perform: causing a second substrate to be placed in the process chamber according to the suggested placement, wherein the one or more elements of the substrate support comprise a process kit ring, and wherein the second substrate is positioned within an inner diameter of the process kit ring according to the suggested placement.

13. The system of claim 12, wherein the processing device is further to perform: causing the second substrate to be processed in the process chamber according to the recipe, wherein the second substrate comprises a second surface profile after the processing; generating a second profile map of the second surface profile using the substrate measurement tool; processing data of the second profile map using the model, wherein the model outputs a second estimated substrate placement value; comparing the second estimated substrate placement value to the first estimated substrate placement value; and updating the suggested placement based on the comparison.

14. The system of claim 11, the processing device is further to perform: wherein determining a center of the substrate support; and aligning the first substrate with the center of the substrate support prior to the processing of the first substrate.

15. The system of claim 14, wherein a center of the substrate is offset from the center of the substrate support for the suggested placement of the substrate.

16. The system of claim 11, wherein the model comprises a trained machine learning model, and wherein the processing device is further to: train a machine learning model to produce the trained machine learning model, wherein the machine learning model is trained using data of a plurality of processed substrates processed according to the recipe.

17. A computer readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations comprising: causing a first substrate to be processed in a process chamber according to a recipe while the first substrate is supported by a substrate support of the process chamber, wherein the first substrate comprises a first surface profile after the processing; ​ a first profile map of the first surface profile of the first substrate is generated using a substrate measurement system of a substrate processing system; data of the first profile map is processed using a model, wherein the model outputs a first estimated substrate placement value of the first substrate relative to placement of one or more elements of the substrate support; and a suggested placement of a substrate on the substrate support is determined based on the first estimated substrate placement value.

18. The computer-readable medium of claim 17, the operations further comprising: causing a second substrate to be placed in the processing chamber according to the suggested placement, wherein the one or more elements of the substrate support comprise a processing kit ring, and wherein the second substrate is positioned within an inner diameter of the processing kit ring according to the suggested placement.

19. The computer-readable medium of claim 18, the operations further comprising: causing the second substrate to be processed in the processing chamber according to the recipe, wherein the second substrate comprises a second surface profile after the processing; a second profile map of the second surface profile is generated using the substrate measurement system; data of the second profile map is processed using the model, wherein the model outputs a second estimated substrate placement value; the second estimated substrate placement value is compared to the first estimated substrate placement value; and the suggested placement is updated based on the comparison.

20. The computer-readable medium of claim 17, wherein the model comprises a trained machine learning model, the operations further comprising: training a machine learning model to generate the trained machine learning model, wherein the machine learning model is trained using data of a plurality of processed substrates processed according to the recipe. ​

Images