Method and system for spectral libraries in manufacturing systems

Abstract

Spectral data associated with one or more regions of a surface of a substrate is identified. The substrate has been processed according to one or more first operations of a process recipe unknown to a system controller of the manufacturing system. The spectral data is provided as input to a machine learning model trained to predict, based on the given spectral data, a respective process recipe associated with the substrate and one or more operations of the respective process recipe that have already been performed. Based on one or more outputs of the machine learning model, it is determined that the substrate is associated with the process recipe and that one or more second operations have not yet been performed. The substrate is processed according to one or more second operations of the process recipe.

Description

[Technical Field]

[0001] FIELD OF THE DISCLOSURE Embodiments of the present disclosure relate generally to manufacturing systems, and more particularly to methods and systems for spectral libraries in manufacturing systems. [Background technology]

[0002] Substrate processing can include a series of processes performed in one or more process chambers of a manufacturing system. For example, a substrate can be processed according to a deposition process, an etch process, etc. In some cases, a process chamber may stop operating while a substrate process is being performed. For example, a manufacturing system including a process chamber performing a substrate process may lose power, causing the substrate process to terminate unexpectedly in the process chamber. When power to the manufacturing system is restored, a system controller of the manufacturing system may not recognize the particular substrate in the process chamber and / or the substrate process that was being performed on the substrate when power to the process tool was lost.

[0003] In some cases, a conditioning operation may be performed on a process chamber to make the process chamber suitable for substrate processing. For example, performing a substrate process may cause wear on process chamber parts and / or the interior surfaces of the process chamber. Substrate processing may be temporarily suspended in the process chamber for a period of time so that a maintenance process (e.g., a preventive maintenance process, a corrective maintenance process, etc.) can be performed to correct and / or mitigate wear or damage to the process chamber parts and / or the interior of the process chamber. One or more conditioning operations may be performed as part of the maintenance process. Once the maintenance process is completed and the process chamber is restored to a suitable condition for substrate processing, the process chamber may resume substrate processing. Summary of the Invention

[0004] Some of the described embodiments cover a method and system for preventing misprocessing of a substrate in a manufacturing system. The method includes identifying spectral data associated with a substrate processed in the manufacturing system according to one or more first operations of a process recipe. The process recipe is unknown to a system controller of the manufacturing system. The method further includes determining a process recipe associated with the substrate based on the identified spectral data. The method further includes identifying one or more second operations of the process recipe for the substrate. The method further includes processing the substrate according to one or more second operations of the process recipe.

[0005] Some of the described embodiments cover a method and system for monitoring a conditioning process in a process chamber of a manufacturing system. The method includes identifying spectral data associated with a first substrate in the manufacturing system. The first substrate is being processed in the process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber. The method further includes determining a state of the process chamber after performance of the one or more first operations based on the identified spectral data. The method further includes determining whether the state of the process chamber satisfies one or more chamber condition criteria. In response to determining that the state of the process chamber does not satisfy the one or more chamber condition criteria, the method further includes updating one or more second operations of the chamber conditioning process so that the state of the process chamber satisfies the one or more chamber condition criteria. The method further includes processing a second substrate in the process chamber according to the updated one or more second operations.

[0006] The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like reference numerals refer to like elements. It should be noted that different references to "an" or "one" embodiment in the present disclosure are not necessarily to the same embodiment, and such references mean at least one. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 illustrates an exemplary computer system architecture according to aspects of the present disclosure. [Figure 2] FIG. 1 is a top schematic view of an exemplary manufacturing system according to aspects of the present disclosure. [Figure 3] FIG. 2 is a cross-sectional schematic side view of a substrate measurement subsystem according to aspects of the present disclosure. [Figure 4] FIG. 2 illustrates an example process identifier engine according to aspects of the present disclosure. [Figure 5] 1 is a flow diagram of an exemplary method for preventing misprocessing of a substrate, according to an aspect of the present disclosure. [Figure 6] 1 is a flow diagram of an exemplary method for training a machine learning model according to an aspect of the present disclosure. [Figure 7] FIG. 1 illustrates an exemplary chamber conditioning engine according to aspects of the present disclosure. [Figure 8] 1 is a flow diagram of an exemplary method for monitoring one or more conditioning operations in a process chamber, according to an aspect of the present disclosure. [Figure 9] 1 is a flow diagram of another example method for training a machine learning model according to aspects of the present disclosure. [Figure 10] FIG. 1 is a block diagram of an exemplary computer system that operates in accordance with one or more aspects of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0008] The embodiments described herein provide methods and systems for spectral libraries in a manufacturing system. In particular, the embodiments described herein provide methods and systems for preventing misprocessing of substrates in a manufacturing system. The embodiments described herein additionally or alternatively provide methods and systems for monitoring a conditioning process in a process chamber of a manufacturing system.

[0009] Substrate processing can include a series of processes for producing electrical circuits on semiconductors (e.g., silicon wafers, etc.) according to a circuit design. Such substrate processes can be performed using substrates placed in one or more process chambers of a process tool of a manufacturing system. Substrate processes can include deposition processes, etching processes, etc. A deposition process involves depositing one or more material layers on a surface of a substrate placed in a process chamber. An etching process involves transferring a pattern of a mask material to one or more material layers on the surface of the substrate. For example, after placing a substrate in a process chamber, an etching plasma can be flowed into the process chamber to remove (e.g., etch) portions of the material layer on the surface of the substrate that are not covered by the mask material. The portions of the layer remaining after the etching process can have a pattern that is the same as or similar to the pattern of the mask material. A production substrate, as referred to herein, includes any substrate that is processed according to a substrate process (e.g., an etching process, a deposition process, etc.).

[0010] In some cases, a system controller of a manufacturing system may lose track of one or more substrates being processed in the manufacturing system. For example, the manufacturing system may lose power (e.g., due to a weather event, system failure, etc.) while a substrate process is being performed on a substrate in a process chamber. The state of the substrates and / or process chambers in the manufacturing system may not have been stored in the system controller's memory before the power loss. Thus, when power is restored to the manufacturing system, the system controller may not know which substrates were located in each process chamber and / or which substrate process operations were performed on the substrates before the power loss. In another example, the computing system of the system controller may be reset or restarted (e.g., due to a failure in the computing system, to perform an emergency update of software or firmware running on the computing system, etc.) while one or more substrates are being processed in the manufacturing system. After the computing system is reset or restarted, the system controller may not know the state of the substrates and / or process chambers, as described above.

[0011] In some manufacturing systems, substrates may include one or more identification marks that are unique to each substrate. For example, a unique serial number may be printed or otherwise formed on the surface of a substrate to distinguish it from other substrates in the manufacturing system. Such identification marks are referred to herein as substrate identifiers (IDs). As described above, after a system controller loses track of one or more substrates, substrates in a process chamber may be transferred to a substrate identifier station, which may scan the surface of the substrate for a substrate ID. After the substrate ID is identified, the substrate identifier (ID) station may send an indication of the substrate ID to the system controller. The system controller may identify a process recipe for the substrate based on the substrate ID and may determine whether to continue processing the substrate according to the process recipe (e.g., by performing one or more operations that would have been performed if the system controller had not lost track of the substrate) or to terminate processing of the substrate and begin substrate processing for another substrate (e.g., replacing the substrate).

[0012] The internal environment of the process chamber and / or other stations of the process tool can be under vacuum and / or highly controlled to prevent oxygen, moisture, etc. from affecting the quality of the substrates processed in the chamber. The substrate ID station can be located outside the process chamber and / or process tool, and accordingly, the substrate is removed from the internal environment of the process tool when transferred to the substrate ID station. In some cases, exposing a substrate to the external environment of the process tool before each substrate processing operation is completed can damage the substrate. The substrate may become unusable, and the system controller may initiate substrate processing on one or more other substrates to replace the damaged substrate. In other or similar examples, the substrate ID station can be or include a substrate ID scanner device. An operator of the manufacturing system may open the lid of a process chamber containing a substrate and use the substrate ID scanner device to scan the substrate ID on the surface of the substrate. Opening the process chamber lid exposes the substrate and the internal environment of the process chamber to the external environment of the process tool, which can damage the substrate and / or introduce undesirable particles (e.g., oxygen, moisture, etc.) into the process chamber. The substrate may become unusable, as described above. In some cases, one or more conditioning operations may be performed in the process chamber (e.g., after the substrate is removed) to return the process chamber to a condition suitable for processing the substrate. Performing substrate processing on additional substrates (e.g., to replace a damaged substrate) and / or performing process chamber conditioning operations may reduce the overall throughput and efficiency of the manufacturing system and may increase the overall wait time of the manufacturing system.

[0013] In some embodiments, a chamber conditioning operation can be performed on a process chamber as part of an initialization process and / or a maintenance process for the process chamber. An initialization process refers to a process that prepares a process chamber for performing a particular substrate process (e.g., an etch process, a deposition process, etc.). An initialization process can be performed when a process chamber is first used in a process tool and / or when a process chamber has previously been used for a particular substrate process (e.g., a deposition process) and is now being used for a different substrate process (e.g., an etch process). A maintenance process can be performed periodically to correct and / or mitigate wear or damage to process chamber parts and / or the interior of the process chamber. A maintenance process can be a preventive maintenance (PM) process and / or a corrective maintenance (CM) process. A PM process refers to a maintenance process performed according to a regular maintenance schedule to maintain one or more components of a process chamber and / or to mitigate wear on the interior of the process chamber. A CM process refers to a maintenance process to correct or mitigate a failure of one or more process chamber parts detected before, during, or after completion of a substrate process (e.g., between runs of a scheduled PM process).

[0014] A chamber conditioning operation can include etching one or more substrates (e.g., blank silicon wafers or dummy wafers) to make the interior of the process chamber suitable for substrate processing. A substrate processed during a chamber conditioning operation is referred to herein as a chamber conditioning substrate or simply a conditioning substrate. A substrate processed during a substrate process (e.g., for producing electrical devices, etc.) is referred to herein as a production substrate. Some systems use metrology to determine whether the interior of the process chamber is suitable for production substrate processing. For example, after a chamber conditioning operation is completed, the conditioned substrate is transferred from the process chamber to a metrology device. The metrology device can be located external to the process tool that includes the process chamber, and the conditioned substrate is correspondingly removed from the vacuum environment of the process tool during transfer to the metrology device. The metrology device acquires metrology measurements (e.g., critical dimension (CD) measurements, etch rate measurements, etc.) of the conditioned substrate, which can be used (e.g., by a system controller) to determine whether chamber conditions are met (e.g., whether the acquired metrology measurements correspond to target metrology measurements for the chamber). Subsequent conditioning operations may be performed on the process chamber until chamber conditions are met and / or a threshold number of conditioning substrates have been processed, at which point the process chamber is determined to be restored to a condition suitable for substrate processing and the process chamber may be used for production substrate processing.

[0015] The time from taking a process chamber offline to perform a conditioning operation to bringing the process chamber back online for production substrate processing is called the green-to-green (G2G) time. As the G2G time of a process chamber increases, the total number of production substrates processed by the process chamber decreases, which can reduce overall efficiency and throughput and increase overall wait time for the process chamber and the manufacturing system. As described above, for each conditioning operation performed in a process chamber, the conditioned substrate is removed from the process chamber and transferred to a metrology device, which may be located outside the process tool, for measurement. Transferring the conditioned substrate to the metrology device and obtaining metrology measurements of the conditioned substrate can take a significant amount of time, which can increase the G2G time of the process chamber. In some systems, once the first conditioned substrate is transferred from the process chamber and measured by the external metrology device, the process chamber may perform additional conditioning operations on subsequent conditioned substrates. The process chamber may have been restored to a condition suitable for processing production substrates after the conditioning operation on the first conditioning substrate, and therefore, additional conditioning operations initiated while the first conditioning substrate is transferred and measured may be unnecessary. Thus, the additional conditioning operations may increase the G2G time of the process chamber and unnecessarily consume resources (e.g., processing resources, computing resources, etc.) of the manufacturing system, which may further reduce efficiency and throughput and increase the overall latency of the manufacturing system.

[0016] Aspects of the present disclosure address these and other shortcomings by providing methods and systems for a spectral library in a manufacturing system. Spectral data can be collected for a substrate (e.g., a production substrate, a conditioning substrate, etc.) before, during, or after a process is performed in a process chamber. In some embodiments, the spectral data can be collected by a substrate measurement subsystem, such as substrate measurement subsystem 300 of FIG. 3. The spectral data can include or otherwise correspond to the intensity (i.e., the intensity or amount of energy) of each wavelength of detected energy waves reflected from the surface of the substrate. The spectral data can be used to prevent misprocessing of substrates in a manufacturing system and / or to monitor a conditioning process in a process chamber of a manufacturing system, in accordance with embodiments described herein.

[0017] In some embodiments, production substrates can be transferred to a process chamber for processing according to a substrate process (e.g., an etch process, a deposition process, etc.). Before, during, or after performing a substrate process, a system controller of a fabrication system may lose track of the production substrate and may become unaware of the type of production substrate, the process recipe associated with the production substrate, and / or one or more operations of the process recipe that have been performed and / or will be performed on the production substrate. In an illustrative example, the fabrication system may lose power before, during, or after substrate processing. In another example, the system controller may be reset or restarted (e.g., by an operator of the fabrication system) before, during, or after substrate processing. In some embodiments, the substrate measurement subsystem can collect spectral data of one or more regions on the surface of the substrate. In some embodiments, the substrate measurement subsystem can be configured to collect spectral data of the substrate while the substrate is in the process chamber. In other or similar embodiments, the substrate measurement subsystem can be included in or otherwise connected to a region of the process tool (e.g., a transfer chamber) having an internal environment that is in the same or similar condition as the process chamber (e.g., under vacuum, with minimal oxygen, moisture, etc.). The substrate can be transferred from the process chamber to the substrate measurement subsystem for measurement. In such an embodiment, the substrate can be measured in the controlled environment of the process tool and is not removed from the process tool and exposed to the outside environment for measurement.

[0018] The substrate measurement subsystem can collect spectral data for the production substrate and transmit the collected spectral data (e.g., over a network) to a system controller or another computing system of the manufacturing system. The system controller, in some embodiments, can determine a process recipe associated with the production substrate based on the collected spectral data and spectral data from the spectral library. The spectral library can include spectral data collected for previous substrates processed according to one or more previous substrate processes in the manufacturing system. In some embodiments, the system controller can compare the collected spectral data for the production substrate with the spectral data from the spectral library and identify previous spectral data corresponding to the collected spectral data. Upon identifying previous spectral data corresponding to the collected spectral data, the system controller can determine the previous substrate for which the previous spectral data was collected. In some embodiments, the spectral library can indicate a process performed on the previous substrate and / or one or more operations of a process recipe for a process performed and / or not yet performed on the previous substrate. In other or similar embodiments, the spectral library can indicate an identifier associated with the previous substrate. In such embodiments, the system controller can determine a process performed on the previous substrate and / or an operation of a process recipe for that process based on data stored in the memory of the manufacturing system. In response to determining a process recipe associated with the production substrate and / or one or more operations that have been performed and / or have not yet been performed on the substrate based on the collected spectral data, the system controller can cause the substrate to be processed according to the one or more operations that have not yet been performed.

[0019] A spectral library can include spectral data collected for a large number of previous substrates (e.g., thousands, tens of thousands, hundreds of thousands, etc.) processed in a manufacturing system. In some embodiments, the data in the spectral library can be used to train a machine learning model that predicts a process recipe associated with a production substrate and / or one or more operations that have been performed and / or have not yet been performed on a production substrate based on given spectral data collected for the substrate. Further details regarding spectral libraries and machine learning models are provided herein.

[0020] In additional or alternative embodiments, one or more conditioning operations can be performed in a process chamber of a manufacturing system using one or more conditioned substrates. In an illustrative example, the conditioning operation can be performed in some embodiments by etching the conditioned substrate in the process chamber. The substrate measurement system can collect spectral data associated with one or more regions of the conditioned substrate before, during, and / or after the conditioning operation. As described above, the substrate measurement subsystem can be configured to collect spectral data for the substrate while the substrate is disposed in the process chamber. In such embodiments, the substrate measurement subsystem can collect spectral data for the substrate while the substrate is in the process chamber (e.g., before, during, or after the conditioning operation). In other or similar embodiments, the substrate can be transferred (e.g., by one or more robotic arms) to the substrate measurement subsystem for measurement after the conditioning operation in the process chamber. In any embodiment, the substrate measurement subsystem can transmit the spectral data collected for the conditioned substrate to a system controller or another computing system of the manufacturing system.

[0021] A system controller or other computing system can determine the state of the process chamber before, during, or after the conditioning operation based on the spectral data collected for the substrate. The spectral data can indicate the state or state of the substrate before, during, or after the conditioning operation. The state or state of the substrate can correspond to the state or state of the process chamber before, during, or after the operation. Thus, the spectral data for the substrate can indicate the state of the chamber before, during, or after the conditioning operation. In some embodiments, the system controller can determine the state of the process chamber based on the collected spectral data and spectral data from a spectral library. In some embodiments, the spectral library can include spectral data collected for previously conditioned substrates processed according to one or more previous conditioning operations in a process chamber of the manufacturing system. In other or similar embodiments, the spectral library can be the same or similar to the libraries described above. The system controller can compare the collected spectral data for the conditioned substrate with previous spectral data collected for previous conditioned substrates processed according to one or more previous conditioning operations in the manufacturing system and can identify the previous spectral data that corresponds to the collected spectral data. Upon identifying the previous spectral data that corresponds to the collected spectral data, the system controller can determine the previous conditioned substrate from which the previous spectral data was collected. In some embodiments, the spectral library can indicate the condition of the previously conditioned substrate and / or the condition of the process chamber that processed the previously conditioned substrate.The system controller, in some embodiments, may determine the condition of the process chamber used to process the current conditioned substrate based on the condition of the previous conditioned substrate and / or the condition of the process chamber that processed the previous conditioned substrate.

[0022] Upon determining the state of the process chamber based on the spectral data collected for the conditioning substrate processed in the process chamber, the system controller can determine whether the state of the process chamber satisfies one or more chamber condition criteria (e.g., whether the process chamber is in a condition suitable for processing production substrates). If the system controller determines that the one or more chamber condition criteria are met, the system controller can determine that the initialization and / or maintenance processes for the process chamber are complete, and the system controller can begin processing production substrates in the process chamber. If the system controller determines that the criteria are not met, the system controller can update the initialization and / or maintenance processes based on the determined chamber state. In one example, the system controller can determine that one or more additional conditioning operations should be performed on the process chamber based on the collected spectral data. In another example, the system controller can determine that settings for one or more additional conditioning operations should be modified to bring the chamber state into correspondence with a target state based on the collected spectral data. In some embodiments, the system controller can cause the updated initialization and / or maintenance process operations to be performed on the process chamber.

[0023] As indicated above, a spectral library can include spectral data collected for a large number of conditioned substrates (e.g., thousands, tens of thousands, hundreds of thousands, etc.) processed in a manufacturing system. In some embodiments, the data in the spectral library can be used to train a machine learning model that predicts the state of a process chamber before, during, or after a conditioning operation of a conditioned substrate based on given spectral data collected for the conditioned substrate. In some embodiments, a system controller can provide (e.g., continuously) the spectral data collected for conditioning substrates processed in the process chamber as input to the trained machine learning model. When the system controller determines that the state of the process chamber meets one or more chamber state criteria based on one or more outputs of the trained model, the system controller can determine that the initialization process and / or maintenance process of the process chamber is complete, as described above. Further details regarding spectral libraries and machine learning models are provided herein.

[0024] Aspects of the present disclosure address the shortcomings of the prior art by providing techniques for preventing mishandling of substrates in a manufacturing system and for monitoring conditioning processes in process chambers of a manufacturing system. A system controller (or another computing system) that loses track of one or more substrates (e.g., production substrates) in a manufacturing system can use spectral data collected by the one or more substrates (e.g., by a substrate measurement subsystem) to determine the process recipe associated with the substrates and / or one or more operations that have been performed and / or have not yet been performed on the substrates. The spectral data can be collected by the substrate measurement subsystem without the substrates being removed from the process tool and exposed to the external environment of the process tool. The system controller can determine the correct process recipe for the substrates and resume substrate processing without damaging the substrates, resulting in fewer substrate processes being performed on additional substrates in the manufacturing system (e.g., to replace damaged substrates). This improves the overall efficiency and throughput of the manufacturing system and reduces the overall wait time of the manufacturing system.

[0025] According to additional aspects of the present disclosure, a system controller (or another computing system) can use spectral data collected on a substrate (e.g., a conditioning substrate) to determine the state of a process chamber during an initialization process and / or a maintenance process. Based on the determined state, the system controller can further determine whether the initialization process and / or the maintenance process is complete or whether the operation of such a process should be updated to bring the state of the process chamber into correspondence with a target state. As described above, the spectral data can be collected by a substrate measurement subsystem without removing the substrate from the process tool (e.g., for measurement by an external metrology device). By using the spectral data to determine the state of the process chamber, the system controller can determine the state of the process chamber in less time and / or after performing fewer conditioning operations. Therefore, the G2G time of the process chamber is reduced, thereby improving overall efficiency and reducing the overall wait time of the manufacturing system.

[0026] FIG. 1 illustrates an exemplary computer system architecture 100 according to aspects of the present disclosure. In some embodiments, the computer system architecture 100 may be included as part of a manufacturing system for processing substrates, such as the manufacturing system 200 of FIG. 2 . The computer system architecture 100 includes a client device 120, a manufacturing tool 124, a metrology tool 128, a server machine 130, a prediction server 112 (e.g., for generating prediction data, providing model fitting, using a knowledge base, etc.), and / or a data store 140. The prediction server 112 may be part of a prediction system 110. The prediction system 110 may further include server machines 170 and 180. The manufacturing tool 124 may include a sensor 125 configured to capture data on substrates being processed in the manufacturing system. In some embodiments, the manufacturing tool 124 and the sensor 126 may be part of a sensor system including a sensor server (e.g., a field service server (FSS) at a manufacturing facility) and a sensor identifier reader (e.g., a front-opening unified pod (FOUP) radio frequency identification (RFID) reader for the sensor system). In some embodiments, the metrology device 128 may be part of a metrology system that includes a metrology server (eg, metrology database, metrology holder, etc.) and a metrology identifier reader (eg, a FOUP RFID reader for the metrology system).

[0027] The manufacturing tool 124 may perform operations according to a recipe or over a period of time to produce a product (e.g., an electrical device). In some embodiments, the manufacturing tool 124 may include one or more process chambers that process substrates (e.g., production substrates) according to a process recipe. In other or similar embodiments, the process chambers of the manufacturing tool 124 may perform initialization and / or maintenance processes that include performing one or more conditioning operations (e.g., using one or more conditioning substrates) to make the process chamber suitable for processing production substrates. The manufacturing tool 124 may include a substrate measurement subsystem that includes one or more sensors 126 integrated into the substrate measurement subsystem and configured to generate spectral and / or positional data for the substrate. The sensors 126 configured to generate spectral data (referred to herein as spectral sensing components) may include reflectometry sensors, ellipsometry sensors, thermal spectral sensors, capacitive sensors, etc. In some embodiments, the spectral sensing components may be included in the substrate measurement subsystem or in another portion of the manufacturing system. The one or more sensors 126 (e.g., eddy current sensors, etc.) may also be configured to generate non-spectral data for the substrate. Further details regarding the manufacturing tool 124 and the substrate measurement subsystem are provided with respect to FIGS.

[0028] In some embodiments, the sensors 126 can provide sensor data associated with the manufacturing equipment 124. The sensor data may include one or more values ​​of temperature (e.g., heater temperature), spacing (SP), pressure, high frequency radio frequency (HFRF), electrostatic chuck (ESC) voltage, current, flow rate, power, voltage, etc. The sensor data can be associated with or indicative of hardware parameters, such as settings or components (e.g., size, type, etc.) of the manufacturing equipment 124, or manufacturing parameters, such as process parameters of the manufacturing equipment 124. The sensor data may be collected while the manufacturing equipment 124 is performing a manufacturing process (e.g., equipment readings as a product is processed). The sensor data may vary from substrate to substrate.

[0029] The metrology tool 128 can provide metrology data associated with substrates (e.g., wafers, etc.) processed by the manufacturing tool 124. The metrology data may include values ​​of one or more of film property data (e.g., wafer-space film properties), dimensions (e.g., thickness, height, etc.), dielectric constant, dopant concentration, density, defects, etc. In some embodiments, the metrology data may further include values ​​of one or more surface profile property data (e.g., etch rate, etch rate uniformity, critical dimensions of one or more features included in the surface of the substrate, uniformity of critical dimensions across the surface of the substrate, edge placement error, etc.). The metrology data may be for finished or semi-finished products. The metrology data may vary from substrate to substrate.

[0030] The client device 120 may include computing devices such as personal computers (PCs), laptops, mobile phones, smartphones, tablet computers, netbook computers, network-connected televisions ("smart TVs"), network-connected media players (e.g., Blu-ray players), set-top boxes, over-the-top (OTT) streaming devices, operator boxes, etc. In some embodiments, metrology data may be received from the client device 120. The client device 120 may display a graphical user interface (GUI) that enables a user to provide as input metrology measurements of substrates processed in a manufacturing system.

[0031] The data store 140 may be a memory (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or another type of component or device capable of storing data. The data store 140 may include multiple storage components (e.g., multiple drives or multiple databases) that may span multiple computing devices (e.g., multiple server computers). The data store 140 may store spectral data, non-spectral data, metrology data, and prediction data. The spectral data may include historical spectral data (e.g., spectral data generated for previous substrates processed in the manufacturing system) and / or current spectra (e.g., spectral data generated for the current substrate being processed in the manufacturing system). The current spectral data may be data from which prediction data is generated. Note that while embodiments of the present disclosure refer to spectral data for training machine learning models, embodiments of the present disclosure may also include non-spectral data used to train machine learning models. In some embodiments, the metrology data may include historical metrology data (e.g., metrology measurements of previous substrates processed in the manufacturing system). The data store 140 may also store context data (eg, recipe name, recipe step number, preventative maintenance indicator, operator, etc.) associated with substrates being processed in the manufacturing system.

[0032] In some embodiments, data store 140 may be configured to store data that is inaccessible to users of the manufacturing system. For example, spectral data, non-spectral data, and / or positional data acquired for substrates being processed in the manufacturing system may be inaccessible to users of the manufacturing system. In some embodiments, all data stored in data store 140 may be inaccessible to users (e.g., operators) of the manufacturing system. In other or similar embodiments, some of the data stored in data store 140 may be inaccessible to users, while other portions of the data stored in data store 140 may be accessible to users. In some embodiments, one or more portions of the data stored in data store 140 may be encrypted using an encryption mechanism unknown to the users (e.g., the data is encrypted using a private encryption key). In other or similar embodiments, data store 140 may include multiple data stores, with data that is inaccessible to users stored in one or more first data stores and data that is accessible to users stored in one or more second data stores.

[0033] 1 , the data store 140, in some embodiments, can store data associated with a spectral library 142. The spectral library 142 can include one or more sets of spectral data collected for a substrate before, during, or after performance of one or more operations on the substrate in the manufacturing tool 124. In some embodiments, the spectral data for the substrate can be collected by a substrate measurement subsystem and / or by other sensors 126 of the manufacturing tool 124, as described herein. In an illustrative example, the spectral library 142 can include one or more sets of spectral data collected before, during, or after performance of a substrate process on a substrate (e.g., a production substrate) in one or more process chambers of the manufacturing tool 124. In another illustrative example, the spectral library 142 can include one or more sets of spectral data collected before, during, or after performance of an initialization process and / or a maintenance process (e.g., a PM process, a CM process, etc.) performed on a process chamber of the manufacturing tool 124. In some embodiments, the spectral library 142 can include additional data associated with the substrate for which the spectral data was collected and / or the process recipe and / or manufacturing tool 124 used to process such substrate. For example, for each set of spectral data collected for a substrate, the spectral library 142 may include indications such as an identifier associated with the substrate, a type associated with the substrate, a process recipe associated with the substrate, one or more operations of the process recipe that were performed on the substrate, an identifier of the process chamber that processed the substrate (e.g., according to the one or more operations), a type associated with the process chamber, one or more settings associated with the process chamber (e.g., before, after, or during the one or more operations), the date and / or time that the process recipe was performed on the substrate, metrology data collected for the substrate after the one or more operations were performed, the state of the process chamber before, during, or after the one or more operations were performed, etc.In other or similar embodiments, the additional data associated with the substrate may be stored in a separate area of ​​the data store 140 (eg, a separate area from the spectral library 142).

[0034] In some embodiments, the system 100 may include one or more server machines 130 that include a spectral data engine 132. In some embodiments, a substrate metrology subsystem that collects spectral data for substrates may transmit the spectral data to the spectral data engine 132 (e.g., over the network 108). The spectral data engine 132 may, in some embodiments, generate a mapping between the spectral data collected for each substrate and additional data associated with the substrate and / or process recipe and / or manufacturing tool 124, as described above. The spectral data engine 132 may store an indication of the mapping in the data store 140 (e.g., with the spectral library 142 and / or in a separate area of ​​the data store 140). In some embodiments, the spectral data engine 132 may provide the spectral data received from the substrate metrology subsystem to the prediction component 114 (e.g., over the network 108). The prediction component 114 may, in some embodiments, provide the spectral data as input to a trained machine learning model 190, as described herein. 1 depicts the spectral data engine 132 as residing on the server machine 130, it should be noted that the spectral data engine 132 may reside on any computing system or component of the system 100. For example, in some embodiments, one or more portions of the spectral data engine 132 may be included in the prediction component 114. In another example, one or more portions of the spectral data engine 132 may be included in a system controller for a manufacturing system of the system 100 (e.g., system controller 228 of FIG. 2 ). In yet another example, one or more portions of the spectral data engine 132 may reside on the client device 120.

[0035] In some embodiments, prediction system 110 includes server machine 170 and server machine 180. Server machine 170 includes a training set generator 172 that can generate training data sets (e.g., a set of data inputs and a set of target outputs) for training, validating, and / or testing machine learning model 190. In some embodiments, training set generator 172 can divide the training data into a training set, a validation set, and a test set. In some embodiments, prediction system 110 generates multiple sets of training data. For example, a first set of training data may correspond to a first type of spectral data (e.g., reflectometry spectral data), and a second set of training data may correspond to a second type of spectral data (ellipsometry spectral data). In some embodiments, training set generator 172 can generate the training data based on data from spectral library 142 according to embodiments described herein.

[0036] The server machine 180 may include a training engine 182, a validation engine 184, a selection engine 185, and / or a test engine 186. An engine may refer to hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, a processing unit, etc.), software (e.g., instructions executing on a processing unit, a general-purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. The training engine 182 may be capable of training a machine learning model 190. The machine learning model 190 may refer to a model artifact created by the training engine 182 using training data including training inputs and corresponding target outputs (correct answers for each training input). The training engine 182 can find patterns in the training data that map the training inputs to the target outputs (predicted answers) and provide a machine learning model 190 that captures these patterns. The machine learning model 190 may use one or more of a support vector machine (SVM), a radial basis function (RBF), clustering, supervised machine learning, semi-supervised machine learning, unsupervised machine learning, a k-nearest neighbor algorithm (k-NN), linear regression, random forests, a neural network (e.g., an artificial neural network), and the like.

[0037] The validation engine 184 may be capable of validating the trained machine learning models 190 using the corresponding set of features of the validation set from the training set generator 172. The validation engine 184 may determine the accuracy of each of the trained machine learning models 190 based on the corresponding set of features of the validation set. The validation engine 184 may discard trained machine learning models 190 that have an accuracy that does not meet a threshold accuracy. In some embodiments, the selection engine 185 may be capable of selecting a trained machine learning model 190 that has an accuracy that meets the threshold accuracy. In some embodiments, the selection engine 185 may be capable of selecting the trained machine learning model 190 with the highest accuracy among the trained machine learning models 190.

[0038] The testing engine 186 may be capable of testing the trained machine learning models 190 using a corresponding set of features of the test set from the training set generator 172. For example, a first trained machine learning model 190 trained using a first set of features of the training set may be tested using a first set of features of the test set. The testing engine 186 may determine the trained machine learning model 190 that has the highest accuracy among all of the trained machine learning models based on the test set.

[0039] The prediction server 112 includes a prediction component 114 that can provide spectral and / or non-spectral data of a portion of a current substrate being processed in the manufacturing system as input to a trained machine learning model 190 and execute the trained machine learning model 190 on the input to obtain one or more outputs. In some embodiments, the machine learning model 190 can be trained to predict, based on given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates. In such embodiments, the prediction component 114 can provide the spectral data associated with the substrates as input to the trained model 190 and obtain one or more outputs of the model 190. The one or more outputs can indicate a process recipe associated with the substrates and one or more operations of the process recipe that have already been performed on the substrates. In additional or alternative embodiments, the machine learning model 190 can be trained to predict, based on given spectral data for the respective substrates (e.g., conditioned substrates), the state of each process chamber in the manufacturing system that processed the substrate. In such an embodiment, the prediction component 114 may provide the spectral data collected for the conditioned substrate as input to the trained model 190 and obtain one or more outputs of the model 190. The one or more outputs may be indicative of the condition of the process chamber that processed the conditioned substrate. Further details regarding the training and use of the model 190 are provided herein.

[0040] Client device 120, manufacturing equipment 124, sensor 126, measurement equipment 128, prediction server 112, server machine 130, data store 140, server machine 170, and server machine 180 may be coupled to each other via network 108. In some embodiments, network 108 is a public network that provides client device 120 with access to prediction server 112, data store 140, and other publicly available computing devices. In some embodiments, network 108 is a private network that provides client device 120 with access to manufacturing equipment 124, measurement equipment 128, data store 140, and other privately available computing devices. Network 108 may include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.

[0041] It should be noted that in some other implementations, the functionality of server machines 130, 170, and 180 and prediction server 112 may be provided by fewer machines. For example, in some embodiments, server machines 130, 170, and 180 may be combined into a single machine, and in some other or similar embodiments, server machines 130, 170, and 180 and prediction server 112 may be combined into a single machine.

[0042] In general, functionality described in one embodiment as being performed by server machine 130, server machine 170, server machine 180, and / or prediction server 112 may also be performed on client device 120. Additionally, functionality attributed to a particular component may be performed by a different component or multiple components operating together.

[0043] In embodiments, a "user" may be represented as a single individual. However, other embodiments of the present disclosure encompass a "user" that is an entity controlled by multiple users and / or automated sources. For example, a collection of individual users consolidated as a group of administrators may be considered a "user."

[0044] 2 is a top schematic view of an exemplary manufacturing system 200 according to aspects of the present disclosure. The manufacturing system 200 can perform one or more processes on a substrate 202. The substrate 202 can be any suitably rigid, fixed-dimension, planar article suitable for fabricating electronic devices or circuit components thereon, such as, for example, a silicon-containing disk or wafer, a patterned wafer, a glass plate, etc. In some embodiments, the substrate 202 can be a production substrate (e.g., a substrate used in the production of a product such as an electronic device), a conditioning substrate (e.g., a substrate used during the performance of one or more conditioning operations, such as an initialization process and / or a maintenance process), and / or any other type of substrate.

[0045] The manufacturing system 200 may include a process tool 204 and a factory interface 206 coupled to the process tool 204. The process tool 204 may include a housing 208 having a transfer chamber 210 therein. The transfer chamber 210 may include one or more process chambers (also referred to as process chambers) 214, 216, 218 disposed around and coupled to the housing 208. The process chambers 214, 216, 218 may be coupled to the transfer chamber 210 through respective ports, such as slit valves. The transfer chamber 210 may also include a transfer chamber robot 212 configured to transfer substrates 202 between the process chambers 214, 216, 218, load locks 220, etc. The transfer chamber robot 212 may include one or more arms, each including one or more end effectors at the end of each arm. The end effectors may be configured to manipulate specific objects, such as wafers.

[0046] The process chambers 214, 216, 218 may be adapted to perform any number of processes on the substrate 202. The same or different substrate processes may be performed in each process chamber 214, 216, 218. In some embodiments, the process chambers 214, 216, 218 may perform substrate processes on one or more substrates 202. The substrate processes may include atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, hardening, pre-cleaning, metal or metal oxide removal, etc. In some embodiments, the substrate processes may include a combination of two or more of atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, hardening, pre-cleaning, metal or metal oxide removal, etc. Other processes may be performed on the substrate in these process chambers. For example, an initialization process may be performed on one or more of the process chambers 214, 216, 218 to prepare the process chambers 214, 216, 218 for substrate processing. In another example, a maintenance process (e.g., a PM process, a CM process, etc.) may be performed to mitigate and / or correct wear or damage to components and / or internals of the process chambers 214, 216, 218. The process chambers 214, 216, 218 may each include one or more sensors configured to capture data of the substrate 202 and / or the environment within the process chambers 214, 216, 218 before, after, or during substrate processing. In some embodiments, the one or more sensors may be configured to capture spectral and / or non-spectral data of a portion of the substrate 202.

[0047] A load lock 220 may also be coupled to the housing 208 and the transfer chamber 210. The load lock 220 may be configured to interface with the transfer chamber 210 on one side and also be coupled to the factory interface 206. The load lock 220, in some embodiments, may have an environmentally controlled atmosphere that may vary from a vacuum environment (where substrates may be transferred to and from the transfer chamber 210) to an inert gas environment at or near atmospheric pressure (where substrates may be transferred to and from the factory interface 206).

[0048] Factory interface 206 may be any suitable enclosure, such as, for example, a front-end equipment module (EFEM). Factory interface 206 may be configured to receive substrates 202 from substrate carriers 222 (e.g., front-opening unified pods (FOUPs)) docked to various load ports of factory interface 206. Factory interface robots 226 (shown in dotted lines) may be configured to transfer substrates 202 between substrate carriers (also referred to as containers) 222 and load locks 220. In other and / or similar embodiments, factory interface 206 may be configured to receive replacement parts from replacement part storage containers 222.

[0049] The manufacturing system 200 may also be connected to a client device (not shown) configured to provide information to a user (e.g., an operator) regarding the manufacturing system 200. In some embodiments, the client device may provide information to a user of the manufacturing system 200 via one or more graphical user interfaces (GUIs). For example, the client device may provide information regarding one or more changes to be made to a process recipe for the substrate 202 via the GUI.

[0050] The manufacturing system 200 may also include a system controller 228. The system controller 228 may be and / or include a computing device such as a personal computer, a server computer, a programmable logic controller (PLC), a microcontroller, etc. The system controller 228 may include one or more processing devices, which may be a general-purpose processing device such as a microprocessor, a central processing unit, 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 word (VLIW) microprocessor, or a processor implementing other instruction sets or combinations of instruction sets. The processing device may 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), a network processor, etc. The system controller 228 may include data storage (e.g., one or more disk drives and / or solid-state drives), main memory, static memory, a network interface, and / or other components. The system controller 228 may execute instructions to perform any one or more of the methodologies and / or embodiments described herein. In some embodiments, the system controller 228 may execute instructions to perform one or more operations in the manufacturing system 200 according to a process recipe. The instructions may be stored in a computer-readable storage medium, which may include a main memory, a static memory, a secondary storage, and / or a processing unit (during execution of the instructions).

[0051] The system controller 228 can receive data from sensors contained on or within various portions of the manufacturing system 200 (e.g., process chambers 214, 216, 218, transfer chamber 210, load lock 220, etc.). Data received by the system controller 228 may include spectral and / or non-spectral data for a portion of the substrate 202. The system controller 228 is described herein as receiving data from sensors contained within the process chambers 214, 216, 218. However, the system controller 228 may receive data from any portion of the manufacturing system 200 and use the data received from that portion in accordance with the embodiments described herein. In an illustrative example, the system controller 228 may receive spectral data from one or more sensors for the process chambers 214, 216, 218 before, after, or during substrate processing in the process chambers 214, 216, 218. The data received from sensors in various portions of the manufacturing system 200 may be stored in a data store 250. Data store 250 may be included as a component within system controller 228 or may be a separate component from system controller 228. In some embodiments, data store 250 may be data store 140 described with respect to FIG.

[0052] The manufacturing system 200 may further include a substrate measurement subsystem 240. The substrate measurement subsystem 240 may acquire spectral measurements of one or more portions of the substrate 202 before, during, or after the substrate 202 is processed in the manufacturing system 200. In some embodiments, the substrate measurement subsystem 240 may acquire spectral measurements of one or more portions of the substrate 202 in response to receiving a request for a spectral measurement from the system controller 228. The substrate measurement subsystem 240 may be integrated within a portion of the manufacturing system 200. In some embodiments, the substrate measurement subsystem 240 may be integrated with the transfer chamber 210, as shown in FIG. 2 . In other or similar embodiments, the substrate measurement subsystem 240 may be integrated within the factory interface 206. In still other or similar embodiments, the substrate measurement subsystem 240 may be integrated into one or more of the process chambers 214, 216, and / or 218. For example, one or more components of the substrate measurement subsystem 240 may be included in the internal environment of the process chambers 214, 216, and / or 218. In another example, the process chambers 214, 216, 218 may include a window (e.g., on the lid of the process chamber, on a sidewall of the process chamber, etc.) that optically exposes the internal environment of the process chamber 214, 216, 218 to an environment external to the process chamber 214, 216, 218. The substrate measurement subsystem 240 may be located outside the process chamber 214, 216, 218 and configured to collect spectral data of the substrate while the substrate is within the process chamber 214, 216, 218. In yet other or similar embodiments, the substrate measurement subsystem 240 may not be integrated with any portion of the manufacturing system 200, but instead may be a stand-alone component. In such an embodiment, the substrate 202 measured in the substrate measurement subsystem 240 may be transferred to or from portions of the manufacturing system 200 before or after the substrate 202 is processed in the manufacturing system 200.

[0053] The substrate measurement subsystem 240 may obtain spectral measurements for a portion of the substrate 202 by generating spectral and / or non-spectral data for the portion of the substrate 202. In some embodiments, the substrate measurement subsystem 240 is configured to generate spectral data, non-spectral data, position data, and other substrate property data for the substrate 202 (e.g., thickness of the substrate 202, width of the substrate 202, etc.). After generating the data for the substrate 202, the substrate measurement subsystem 240 may transmit the generated data to the system controller 228. In response to receiving the data from the substrate measurement subsystem 240, the system controller 228 may store the data in the data store 250. In other or similar embodiments, the substrate measurement subsystem 240 and / or the system controller 228 may provide the data to the spectral data engine 132. As mentioned above, in some embodiments, one or more portions of the spectral data engine 132 may reside in the system controller 228.

[0054] FIG. 3 is a cross-sectional schematic side view of a substrate measurement subsystem 300 according to aspects of the present disclosure. The substrate measurement subsystem 300 may be the same as or otherwise correspond to the substrate measurement subsystem 240 of FIG. 2. The substrate measurement subsystem 300 may be configured to obtain measurements of one or more portions of a substrate, such as the substrate 202 of FIG. 2, before, during, or after processing of the substrate 202 in a process chamber (e.g., process chambers 214, 216, 218). The substrate measurement subsystem 300 may obtain spectral measurements of the portion of the substrate 202 by generating data (e.g., spectral data, non-spectral data, etc.) associated with the portion of the substrate 202. In some embodiments, the substrate measurement subsystem 300 may be configured to generate spectral data, non-spectral data, positional data, and / or other characteristic data associated with the substrate 202. The substrate measurement subsystem 300 may include a controller 330 configured to execute one or more instructions to generate data associated with the portion of the substrate 202.

[0055] The substrate measurement subsystem 300 may detect that the substrate 202 has been transferred to the substrate measurement subsystem 300. In response to detecting that the substrate 202 has been transferred to the substrate measurement subsystem 300, the substrate measurement subsystem 300 may determine the position and / or orientation of the substrate 202. The position and / or orientation of the substrate 202 may be determined based on an identification of a reference position of the substrate 202. The reference position may be a portion of the substrate 202 that includes an identification feature associated with the particular portion of the substrate 202. The controller 228 may determine the identification feature associated with the particular portion of the substrate 202 based on the identification information determined for the substrate 202.

[0056] The controller 330 can identify a reference position of the substrate 202 using one or more camera components 350 configured to capture image data of the substrate 202. The camera components 350 can generate image data for one or more portions of the substrate 202 and transmit the image data to the controller 330. The controller 330 can analyze the image data to identify an identifying feature associated with the reference position of the substrate 202. The controller 330 can further determine a position and / or orientation of the substrate 202 depicted in the image data based on the identified identifying feature of the substrate 202. The controller 330 may determine the 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 depicted in the image data. In response to determining the position and / or orientation of the substrate 202, the controller 330 can generate position data associated with one or more portions of the substrate 202. In some embodiments, the position data may include one or more coordinates (e.g., Cartesian coordinates, polar coordinates, etc.) each associated with a portion of the substrate 202, and each coordinate is determined based on a distance from a reference position of the substrate 202.

[0057] The substrate measurement subsystem 300 may include one or more measurement components for measuring the substrate 202. In some embodiments, the substrate measurement subsystem 300 may include one or more spectral sensing components 320 configured to generate spectral data for one or more portions of the substrate 202. As previously mentioned, the spectral data may correspond to the intensity of the detected energy waves (i.e., the intensity or amount of energy) for each wavelength of the detected waves.

[0058] The spectral sensing component 320 may be configured to detect energy waves reflected from a portion of the substrate 202 and generate spectral data associated with the detected waves. The spectral sensing component 320 may include a wave generator 322 and a reflected wave receiver 324. In some embodiments, the wave generator 322 may be a light wave generator configured to generate a light beam toward the portion of the substrate 202. In such embodiments, the reflected wave receiver 324 may be configured to receive the light beam reflected from the portion of the substrate 202. The wave generator 322 may be configured to generate an energy flow 326 (e.g., a light beam) and transmit the energy flow 326 to the portion of the substrate 202. A reflected energy wave 328 may be reflected from the portion of the substrate 202 and received by the reflected wave receiver 324.

[0059] In response to the reflected wave receiver 324 receiving the reflected energy waves 328 from the portion of the substrate 202, the spectral sensing component 320 can measure the wavelength of each wave included in the reflected energy waves 328. The spectral sensing component 320 can further measure the intensity of each measured wavelength. In response to measuring each wavelength and each wavelength intensity, the spectral sensing component 320 can generate spectral data for the portion of the substrate 202. The spectral sensing component 320 can transmit the generated spectral data to the controller 330. In response to receiving the generated spectral data, the controller 330 can generate a mapping between the received spectral data and position data of the measured portion of the substrate 202.

[0060] The substrate measurement subsystem 300 may be configured to generate a particular type of spectral data based on the type of measurement acquired in the substrate measurement subsystem 300. In some embodiments, the spectral sensing component 320 may be a first spectral sensing component configured to generate one type of spectral data. For example, the spectral sensing component 320 may be configured to generate reflectometry spectral data, ellipsometry spectral data, hyperspectral imaging data, chemical imaging data, thermal spectral data, or conductivity spectral data. In such embodiments, the first spectral sensing component may be removed from the substrate measurement subsystem 300 and replaced with a second spectral sensing component configured to generate a different type of spectral data (e.g., reflectometry spectral data, ellipsometry spectral data, hyperspectral imaging data, chemical imaging data, eddy current spectral data, thermal spectral data, or conductivity spectral data).

[0061] In some embodiments, the substrate measurement subsystem 300 can include a substrate holder 340. The substrate holder 340 can be a chuck, such as a vacuum chuck, an electrostatic chuck, a magnetic chuck, a mechanical chuck (e.g., a four-jaw chuck, a three-jaw chuck, an edge / ring clamp chuck, etc.), or other type of chuck. The substrate holder 340 can also be or include a plate or other surface having a substrate-shaped pocket and / or a set of pins or other features (e.g., three pins) that surround the substrate and prevent the substrate from shifting relative to the substrate holder 340 during movement of the substrate holder 340. The substrate holder 340 can secure the substrate 202. In one embodiment, the substrate holder 340 includes an edge clamp that clamps the substrate from an edge. In one embodiment, the substrate holder 340 is a vacuum chuck. In other embodiments, the substrate holder 340 can be a different type of chuck, such as an electrostatic chuck, a mechanical chuck, or a magnetic chuck.

[0062] The substrate measurement subsystem may include one or more positioning components configured to change the position and / or orientation of the substrate 202 relative to the spectral sensing component 320. In some embodiments, the positioning components may include a first actuator 352 that may move the substrate holder 340 about a first axis (e.g., a rotational axis) and / or a second axis (e.g., a vertical axis). In some embodiments, the first actuator 352 may be controlled by a servo controller and / or a servo motor, which allows for precise control of the rotational position, linear position, velocity, and / or acceleration of the first actuator 352, and thus the substrate holder 340. The substrate holder 340 may have a mass between 1.0 kilograms (kg) and 2.0 kg, which may allow for a rotational speed of 6000 degrees per second. 2 ~14000 degrees / second 2 This allows for linear and / or rotational acceleration of 100 rpm. In additional or alternative embodiments, the positioning component may include a second actuator 354 that can move the substrate holder 340 along a second axis (e.g., a horizontal axis). The second actuator 354, in some embodiments, may be controlled by a servo controller and / or servo motor, which allows for precise control of the linear position, velocity, and / or acceleration of the second actuator 354, and thus the substrate holder 340.

[0063] As the spectral sensing component 320 generates spectral data for one or more portions of the substrate 202, the one or more positioning components can change the position and / or orientation of the substrate 202 depending on the one or more determined portions of the substrate 202 to be measured. For example, before the spectral sensing component 320 generates spectral data for the substrate 202, the positioning component can position the substrate 202 at Cartesian coordinate (0,0), and the spectral sensing component 320 can generate first spectral data for the substrate 202 at the Cartesian coordinate (0,0). In response to the spectral sensing component 320 generating the first spectral data for the substrate 202 at the Cartesian coordinate (0,0), the second actuator 354 can translate the substrate 202 along the horizontal axis, such that the spectral sensing component 320 is configured to generate second spectral data for the substrate 202 at the Cartesian coordinate (0,1). In response to the spectral sensing component 320 generating the second spectral data at the Cartesian coordinate (0,1) of the substrate 202, the first actuator 352 can rotate the substrate 202 along the first axis, such that the spectral sensing component 320 is configured to generate third spectral data at the Cartesian coordinate (1,1) of the substrate 202. This process may be performed multiple times until spectral data has been generated for each determined portion of the substrate 202.

[0064] In some embodiments, one or more material layers 312 may be included on the surface of the substrate 202. The one or more layers 312 may include an etching material, a photoresist material, a mask material, a deposition material, etc. In some embodiments, the one or more layers 312 may include an etching material that is to be etched according to an etching process performed in a process chamber. In such embodiments, spectral data may be collected for one or more portions of the unetched etching material of the layer 312 deposited on the substrate 202 according to previously disclosed embodiments. In other or similar embodiments, the one or more layers 312 may include an etching material that has already been etched according to an etching process in a process chamber. In such embodiments, one or more structural features (e.g., lines, columns, openings, etc.) may be etched in the one or more layers 312 of the substrate 202. In such embodiments, spectral data may be collected for one or more structural features etched in the one or more layers 312 of the substrate 202.

[0065] In response to receiving at least one of the spectral data, positional data, or property data of the substrate 202, the controller 330 can transmit the received data to the system controller 228 for processing and analysis according to embodiments described herein.

[0066] As described above, one or more components of the substrate measurement subsystem 300 can be included in the process chambers 214, 216, 218 and / or can be configured to collect spectral data of substrates within the process chambers 214, 216, 218. In one illustrative example, the process chamber (e.g., process chamber 214) can include a substrate support assembly configured to support the substrate 202 during a process (e.g., a substrate process, etc.). The camera component 350 and / or the spectral sensing component 320 can be included in the process chamber 214 and can generate spectral and / or non-spectral data of the substrate 202 before, during, or after the process, as described herein. The camera component 350 and / or the spectral sensing component 320 can transmit the generated measurements to the controller 330, which may be located outside the process chamber. The controller 330 can transmit the generated measurements to the system controller 228 and / or the spectral data engine 132, as described herein. In another illustrative example, a transparent window may be incorporated into one or more surfaces (e.g., lid, exterior wall, etc.) of the process chamber 214. The camera component 350 and / or the spectral sensing component 320 may, in some embodiments, be configured to generate spectral data by detecting a spectrum through the transparent window from outside the process chamber 214. It should be noted that, according to embodiments described herein, the camera component 350, the spectral sensing component 320, and / or any other component of the substrate measurement subsystem 300 may be configured to generate spectral and / or non-spectral data of the substrate 202 inside or outside the process chamber 214, 216, 218 according to any orientation.

[0067] Referring again to FIG. 1 , in some embodiments, the spectral data engine 132 can include or be otherwise connected to a process identifier engine and / or a chamber conditioning engine. In some embodiments, the system controller 228 may lose track of one or more substrates (e.g., production substrates) in the process tool 204. The process identifier engine can be configured to determine one or more operations of a process recipe associated with the substrate and / or a process recipe performed on the substrate based on the spectral data collected about the substrate (e.g., by the substrate measurement subsystem 300). In additional or alternative embodiments, chamber conditioning operations for initialization processes and / or maintenance processes (e.g., PM processes, CM processes, etc.) can be performed in the process chambers 214, 216, 218 using one or more chamber conditioning substrates. The chamber conditioning engine can determine whether the condition of the process chambers 214, 216, 218 meets one or more chamber condition criteria based on the spectral data collected about the chamber conditioning substrates. If so, the system controller 228 can initiate one or more substrate processes in the process chambers 214, 216, 218 using one or more production substrates. If not, the system controller can update initialization and / or maintenance processes to cause the process chambers 214, 216, 218 to meet the chamber condition criteria. Further details regarding the process identifier engine and chamber conditioning engine are described herein with respect to Figures 4-9.

[0068] FIG. 4 illustrates an exemplary process identifier engine 410 according to aspects of the present disclosure. As mentioned above, the process identifier engine 410 may be part of the spectral data engine 132 in some embodiments. In other or similar embodiments, the process identifier engine 410 may reside on a different computing system than the computing system executing the spectral data engine 132. In such embodiments, the process identifier engine 410 may be communicatively coupled to the spectral data engine 132 and / or the system controller 228 (e.g., via the network 108, via a bus, etc.). As shown in FIG. 4, the process identifier engine 410 may include a spectral data component 412, a process strategy component 414, and / or an operation component 416. The process identifier engine 410, the spectral data engine 132, and / or the system controller 228 may be coupled to a memory 450 (e.g., via the network 108, via a bus, etc.). In some embodiments, the memory 450 may include one or more regions of the data store 140 and / or any other memory of the system 100.

[0069] 5 is a flow diagram of an exemplary method 500 for preventing misprocessing of a substrate according to an embodiment of the present disclosure. Method 500 is performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one embodiment, method 500 may be performed by a computer system, such as computer system architecture 100 of FIG. 1. In other or similar embodiments, one or more operations of method 500 may be performed by one or more other machines not shown in the figure. In some embodiments, one or more operations of method 500 may be performed by process identifier engine 410. In other or similar embodiments, one or more operations of method 500 may be performed by system controller 228.

[0070] In block 510, process logic identifies spectral data associated with a substrate in a manufacturing system. In some embodiments, the substrate 202 may be a production substrate, as described herein. The substrate 202 may be being processed in the process chambers 214, 216, 218 according to one or more operations of a process recipe (e.g., for a substrate process). For purposes of explanation and illustration only, embodiments and examples of the present disclosure may be described with respect to the process chamber 214. However, the embodiments and examples of the present disclosure may be applied to any of the process chambers 214, 216, 218 and / or any other process chambers in the manufacturing system 200 and / or another manufacturing system (e.g., not including the process chambers 214, 216, 218). In an illustrative example, the substrate 202 may be processed in the process chamber 214 according to one or more first operations of the process recipe. Before, during, or after the execution of the one or more first operations, the system controller 228 may lose track of the substrate 202. For example, before, during, or after performing one or more first operations, the process tool 204 and / or the system controller 228 may lose power, and the state of the substrate 202 and / or the process chamber 214 may not be stored in memory (e.g., non-volatile memory) before the power loss. Thus, the process recipe associated with the substrate 202 and / or the one or more first operations performed on the substrate 202 may not be known to the system controller 228.

[0071] In some embodiments, the system controller 228 may determine that the process recipe and / or one or more first operations performed on the substrate 202 are unknown to the system controller 228 (e.g., by determining that data or information indicating an association of a process recipe with the substrate 202 is not present in the memory 450). Upon determining that the process recipe and / or first operation is unknown, the system controller 228 may cause the substrate measurement subsystem 300 to generate spectral data for one or more regions of the substrate 202. As described above, in some embodiments, the substrate measurement subsystem 300 may be included in or otherwise connected to the transfer chamber 210 (e.g., in a vacuum environment). In such embodiments, the system controller 228 may send a signal to a motion controller for the transfer chamber robot 212 to cause the transfer chamber robot 212 to transfer the substrate 202 from the process chamber 214 to the substrate measurement subsystem 300. In other or similar embodiments, the substrate measurement subsystem 300 may be included in or otherwise connected to the factory interface 206 (e.g., outside a vacuum environment). In such embodiments, the system controller 228 sends a first signal to the motion controller for the transfer chamber robot 212, causing the transfer chamber robot 212 to transfer the substrate 202 from the process chamber 214 to the load lock 220, and sends a second signal to the motion controller factory interface robot 226, causing the factory interface robot 226 to transfer the substrate 202 from the load lock 220 to the substrate measurement subsystem 300. In yet other or similar embodiments, one or more components of the substrate measurement subsystem 300 may be included or otherwise configured to collect spectral and / or non-spectral data of the substrate 202 while the substrate 202 is in the process chamber 214. In such embodiments, the substrate 202 remains in the process chamber 214 for measurement.

[0072] In some embodiments, the system controller 228 and / or the process identifier engine 410 send a signal to the controller 330 to cause the substrate measurement subsystem 300 to generate spectral measurements of one or more regions of the surface of the substrate 202. In some embodiments, the signal may include an indication of the region of the surface of the substrate 202 to be measured. In other or similar embodiments, the region of the surface of the substrate 202 may be provided to the controller 330 prior to initialization of the process chamber 214 and / or process tool 204 (e.g., by an operator or engineer of the manufacturing system 200, etc.). In some embodiments, the controller 330 may cause one or more positioning components to move (e.g., rotate, translate) the substrate 202 along one or more axes to a target position. The spectral sensing component 320 and / or camera component 350 may, in some embodiments, collect spectral and / or non-spectral data about the substrate 202 after the substrate 202 has moved to the target position.

[0073] Once the spectral sensing component 320 and / or the camera component 350 collect the spectral and / or non-spectral data of the substrate 202, the controller 330 can transmit the collected data to the process identifier engine 410 (e.g., via the network 108, via a bus, etc.). The spectral data component 412, in some embodiments, can store the spectral data collected for the substrate 202 in memory 450 (e.g., as spectral data 452). The spectral data 452 associated with the substrate 202, in some embodiments, can be identified from signals provided by the controller 330 and / or from the memory 450.

[0074] 5 , at block 520, the process logic determines a process recipe associated with the substrate based on the identified spectral data. In some embodiments, the process recipe component 414 of the process identifier engine 410 can determine the process recipe associated with the substrate 202 by comparing the spectral data 452 collected for the substrate 202 with spectral data collected for other substrates and stored in the spectral library 142. The spectral data 452 can be indicative of the state of the substrate 202 at the time the spectral data 454 was collected (e.g., after execution of a first operation of the process recipe). The process recipe component 414 can compare the spectral data 452 with the spectral data in the spectral library 142 and can identify a set of spectral data that corresponds to (e.g., matches or substantially matches) the spectral data 452. In response to identifying a set of spectral data corresponding to the spectral data 452, the process recipe component 414 can determine (e.g., based on additional data associated with the substrate and / or a process recipe stored in the spectral library 142) a process recipe associated with the identified set of spectral data and / or the substrate from which the set of spectral data was generated. As described herein, the identified set of spectral data can correspond to a previous substrate processed in a process chamber (e.g., process chambers 214, 216, 218, etc.) according to one or more operations of a previous substrate process. The set of spectral data collected for the previous substrate can indicate the state of the previous substrate after execution of one or more operations of the process recipe used to process the previous substrate. The process recipe component 414 can store instructions for the process recipe in the memory 450 as process recipe data 454.

[0075] In some embodiments, a machine learning model (e.g., model 190) can be trained using data from the spectral library 142 to predict a process recipe associated with a substrate and one or more operations of the process recipe that have already been performed on the substrate based on given spectral data collected for the substrate. More details regarding the training model 190 are provided with respect to FIG. 6 . In some embodiments, the process recipe component 414 can provide the spectral data 452 collected for the substrate 202 as input to the trained machine learning model 190 and obtain one or more outputs of the machine learning model 190. The one or more outputs of the model 190 can include an indication of one or more process recipes and, for each of the one or more process recipes, an indication of a confidence level that the respective process recipe is associated with the substrate 202 based on the spectral data 452. In additional or alternative embodiments, the one or more outputs of the model 190 can include an indication of one or more operations of the respective process recipe and, for each of the one or more operations, a confidence level that the operation was performed on the substrate 202. The process recipe component 414 can determine a process recipe associated with the substrate 202 by identifying a process recipe for one or more outputs of the model 190 that has a confidence that meets one or more confidence criteria (e.g., exceeds a threshold confidence, is greater than other confidences associated with other process recipes, etc.).

[0076] Referring again to FIG. 5 , at block 530, the process logic identifies one or more second operations of the process recipe that have not yet been performed on the substrate. As described above, in some embodiments, the process recipe component 414 can determine the process recipe associated with the substrate 202 by comparing the spectral data 452 with spectral data in the spectral library 142. Also, as described above, the spectral data in the spectral library 142 corresponding to the spectral data 452 can be generated for a previous substrate after performance of one or more operations of the process recipe (e.g., the process recipe associated with the substrate 202). In some embodiments, the spectral library 142 can include a mapping between a set of spectral data and an indication of one or more operations of the process recipe that were performed on the previous substrate. The operation component 416, in some embodiments, can determine one or more operations performed on the substrate 202 based on one or more operations performed on the previous substrate.

[0077] In other or similar embodiments, the one or more outputs of the model 190 may include an indication of one or more operations of the process recipe and, for each of the one or more operations, an indication of the confidence that the respective operation has already been performed on the substrate 202, given the spectral data 452 provided as input to the model 190. The operation component 416 may determine the one or more operations that have a confidence that meets one or more confidence criteria (e.g., exceeds a threshold confidence, is greater than other confidences associated with other operations, etc.).

[0078] In some embodiments, a process recipe associated with the substrate 202 can include one or more operations. The operation component 416 can identify a set of operations associated with the process recipe (e.g., from the spectral library 142, the process recipe library in the data store 140, etc.) and determine which operations have not yet been performed on the substrate 202 based on the determined operations that have been performed on the substrate 202.

[0079] 5 , in block 540, the process logic causes the substrate to be processed according to one or more second operations of the process recipe. In some embodiments, the process identifier engine 410 can provide a notification to the system controller 228 (e.g., via the network 108, via a bus, etc.) indicating the process recipe associated with the substrate 202 and one or more operations of the process recipe that have not yet been executed. Upon receiving the notification, the system controller 228 can begin executing one or more operations according to embodiments described herein. As mentioned above, in some embodiments, the substrate 202 may have been transferred to the substrate measurement subsystem 300 for collection of spectral data 452. In such embodiments, the system controller 228 can send one or more signals to a motion controller (e.g., for the FI robot 226, for the transfer chamber robot 212) to transfer the substrate 202 to the process chamber 214 and / or another process chamber (e.g., process chambers 216, 218, another process chamber of another process tool) for execution of the one or more second operations.

[0080] 6 is a flow diagram of an exemplary method 600 for training a machine learning model according to an aspect of the present disclosure. Method 600 is performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic), software (e.g., running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one implementation, method 600 may be performed by a computer system such as computer system architecture 100 of FIG. 1. In other or similar implementations, one or more operations of method 500 may be performed by one or more other machines not shown in the figure. In some aspects, one or more operations of method 600 may be performed by training set generator 172.

[0081] At block 610, process logic initializes a training set T to be empty (e.g., {}). At block 612, process logic identifies spectral data associated with one or more substrates processed in the manufacturing system. In some embodiments, spectral data may be generated by the substrate measurement subsystem 300 for one or more substrates (e.g., substrate 202) processed according to operations of a process recipe in one or more of the process chambers 214, 216, 218 in accordance with the embodiments described above. In some embodiments, components of the substrate measurement subsystem 300 may collect spectral data while a substrate process is being performed in the respective process chamber (e.g., process chamber 214). Thus, the spectral data collected for the substrate 202 may indicate the state of the substrate 202 during each operation of the substrate process. In other or similar embodiments, the one or more substrates 202 may be transferred from the process chamber 214 to the substrate measurement subsystem 300 after completion of one or more operations of the process recipe (e.g., before all operations of the process recipe have been performed), as described above. The spectral data collected by the substrate measurement subsystem 300 can indicate the state of the substrate 202 after completion of one or more operations of the process recipe and / or before completion of all operations. The controller 330 and / or the system controller 228, in some embodiments, can store the spectral data collected for one or more substrates 202 in the data store 140 (e.g., in the spectral library 142).

[0082] At block 614, the process logic determines one or more operations of the process recipe that have been performed on the one or more substrates. The system controller 228 and / or another component of the system 110 may store instructions of the process recipe that have been performed on the substrate 202 in the data store 140, as described herein. As mentioned above, in some embodiments, the system controller 228 may transfer the substrate 202 to the substrate measurement subsystem 300 for measurement after completion of one or more operations of the process recipe. In such embodiments, the system controller 228 and / or another component of the system 110 may store instructions of one or more operations of the completed process recipe in the data store 140, as described above. In additional or alternative embodiments, a component of the substrate measurement subsystem 300 may collect spectral data during each operation of the process recipe (e.g., without the substrate 202 being removed from the process chamber 214). In such an embodiment, the system controller 228 and / or another component of the system 100 can determine a start period and an end period for each operation of the process recipe based on the process recipe data and / or performance logs for the substrate process. The system controller 228 and / or another component can identify a portion of the spectral data collected during each start period and each end period and associate the identified portion of the spectral data with each operation of the process recipe (e.g., by updating a mapping between the portion of the spectral data and an identifier for each operation). The system controller 228 and / or another component can store the associations and / or mappings in the data store 140, as described above.

[0083] At block 616, process logic generates an input / output mapping, where the input is based on the spectral data and the output identifies one or more operations. In some embodiments, training set generator 172 can identify (e.g., in data store 140, spectral library 142, etc.) spectral data collected for a substrate (e.g., as identified at block 610) and instructions for one or more operations performed on the substrate for which the spectral data was collected (e.g., as identified at block 612) to generate the input / output mapping. In some embodiments, the generated input / output mapping can correspond to a mapping between portions of the spectral data and identifiers of respective operations generated by system controller 228 and / or another component of system 100, as described above. At block 618, process logic adds the input / output mapping to training set T. At block 620, process logic determines whether set T is sufficient for training. In response to process logic determining that training set T is not sufficient for training, method 600 returns to block 612. In response to the processing logic determining that the training set T is sufficient for training, the method 600 proceeds to block 622. At block 622, the processing logic provides the training set T for training a machine learning model (e.g., model 190).

[0084] FIG. 7 illustrates an exemplary chamber conditioning engine 710 according to aspects of the present disclosure. As described above, the chamber conditioning engine 710 may be part of the spectral data engine 132 in some embodiments. In other or similar embodiments, the chamber conditioning engine 710 may reside on a computing system different from the computing system executing the spectral data engine 132. In such embodiments, the chamber conditioning engine 710 may be communicatively coupled to the spectral data engine 132 and / or the system controller 228 (e.g., via the network 108, via a bus, etc.). As shown in FIG. 7, the chamber conditioning engine 710 may include a spectral data component 712, a chamber status component 714, and / or a process update component 716. The chamber conditioning engine 710, the spectral data engine 132, and / or the system controller 228 may be coupled to a memory 750 (e.g., via the network 108, via a bus, etc.). In some embodiments, memory 750 may include one or more regions of data store 140 and / or any other memory of system 100 .

[0085] 8 is a flow diagram of an exemplary method 800 for monitoring one or more conditioning operations performed in a process chamber according to an embodiment of the present disclosure. Method 800 is performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one embodiment, method 800 may be performed by a computer system, such as computer system architecture 100 of FIG. 1. In other or similar embodiments, one or more operations of method 800 may be performed by one or more other machines not shown in the figure. In some embodiments, one or more operations of method 800 may be performed by chamber conditioning engine 710.

[0086] At block 810, the process logic identifies spectral data associated with a first substrate processed in a process chamber of the manufacturing system. In some embodiments, the substrate 202 may be a conditioning substrate, as described herein. The substrate measurement subsystem 300 may collect spectral data for the substrate 202 before, during, or after one or more operations of a chamber conditioning process, according to previously described embodiments. In some embodiments, the controller 330 of the substrate measurement subsystem 300 may provide the spectral data to the system controller 228 and / or the chamber conditioning engine 710, according to previously described embodiments. The spectral data component 712 of the chamber conditioning engine 710 may store the spectral data as spectral data 752 in memory 750. In some embodiments, the spectral data component 712 may store the spectral data 752 in the spectral library 142.

[0087] At block 812, the process logic determines the state of the process chamber after performance of the first operation based on the identified spectral data. As described above, the spectral data 752 can be indicative of the state of the substrate 202 before, during, or after performance of one or more operations (e.g., conditioning operations) of a process performed on the substrate in the process chamber 214. The state of the substrate 202 can be dependent on and / or otherwise related to the state of the process chamber 214 before, during, and / or after performance of the one or more operations. Thus, the state of the substrate 202 as indicated by the spectral data 752 can correspond to the state of the process chamber 214. As described above, the spectral library 142 can include one or more sets of spectral data collected for previous substrates processed in the process chamber (e.g., process chambers 214, 216, 218 and / or another process chamber) following one or more previous conditioning operations (e.g., of a previous initialization process and / or a previous maintenance process). Each set of spectral data collected for each previous substrate can be indicative of the state of the respective substrate before, during, and / or after one or more previous conditioning operations have been performed, and thus each set of spectral data can be indicative of the state of the process chamber that performed one or more conditioning operations on the respective substrate.

[0088] In some embodiments, the chamber state component 714 can compare the spectral data 752 to one or more sets of spectral data in the spectral library 142 and can identify a set of spectral data that corresponds to (e.g., matches or substantially matches) the spectral data 752. The identified set of spectral data can indicate the state of a previous substrate before, during, and / or after one or more previous conditioning operations, which can correspond to the state of a process chamber that processed the previous substrate before, during, and / or after the previous conditioning operations. Thus, the chamber state component 714 can determine the state of the process chamber 214 that processed the substrate 202 based on the identified set of spectral data in the spectral library 142. In some embodiments, the chamber state component 714 can store an indication of the process chamber state in the memory 750 as chamber state data 754.

[0089] In some embodiments, a machine learning model (e.g., model 190) can be trained using data from the spectral library 142 to predict the state of each process chamber that processed a substrate based on given spectral data collected for the substrate (e.g., a conditioned substrate). More details regarding the training model 190 are provided with respect to FIG. 9 . In some embodiments, the chamber state component 714 can provide the spectral data 752 collected for the substrate 202 as input to the trained model 190 to obtain one or more outputs of the machine learning model. The one or more outputs of the model 190 can include an indication of one or more process chamber states and, for each of the one or more process chamber states, an indication of a confidence that the respective process chamber state is associated with the process chamber 214 that processed the substrate 202. The chamber state component 714 can determine the chamber state associated with the process chamber 214 by identifying one or more process chamber states in the output of the model 190 that have a confidence that meets one or more confidence criteria (e.g., exceeds a threshold confidence, is greater than other confidences associated with other process chamber states). The chamber state component 714 may store the chamber state associated with the process chamber 214 (eg, determined based on the output of the model 190) in the memory 750 as chamber state data 754, as described above.

[0090] 8 , in block 814, the process logic determines whether the state of the process chamber satisfies one or more chamber condition criteria. In some embodiments, the state of the process chamber 214 may satisfy the chamber condition criteria if, after execution of one or more conditioning operations performed on the process chamber 214, the state of the process chamber 214 corresponds to (e.g., matches or substantially matches) a target state of the process chamber 214. In some embodiments, the target state of the process chamber 214 may be determined based on historical and / or experimental data collected for one or more substrates and / or process chambers of the manufacturing system 200 (and / or other manufacturing systems). In other or similar embodiments, the target state of the process chamber may be provided by an operator or engineer of the manufacturing system 200 (e.g., via the client device 120).

[0091] The chamber state component 714 may compare the determined state of the process chamber 214 to a target state of the process chamber to determine whether the determined state corresponds to the target state. If so, the chamber state component 714 may determine that the chamber state criteria are met. If not, the chamber state component 714 may determine that the chamber state criteria are not met. In an illustrative example, one or more first conditioning operations may be performed on the process chamber 214 with respect to the substrate 202. The chamber state component 714 may determine the state of the process chamber 214 based on the spectral data 752 collected for the substrate 202, as described above, and compare the determined state with the target state of the process chamber 214 after performing the one or more first conditioning operations. Upon determining that the state of the process chamber 214 after performing the one or more first conditioning operations corresponds to the target state of the process chamber 214 after performing the first conditioning operations, the chamber state component 714 may determine that the state of the process chamber 214 meets the chamber state criteria.

[0092] In response to the process logic determining that the chamber state criteria have been met, the method 600 proceeds to block 816. In response to the process logic determining that the chamber state criteria have not been met, the method 600 proceeds to block 820. In block 816, the process logic determines whether the initialization process and / or maintenance process has been completed. In some embodiments, the chamber state component 714 may determine whether the initialization process and / or maintenance process in the process chamber 214 has been completed by determining whether the state of the process chamber 214 corresponds to a target state of a process chamber having completed the initialization process and / or maintenance process. The target state of a process chamber having completed the initialization process and / or maintenance process may be determined based on historical and / or experimental data, as described above, and / or may be provided by an operator or engineer of the manufacturing system 200 (e.g., via the client device 120). Upon determining that the state of the process chamber 214 corresponds to the target state of a process chamber having completed the initialization process and / or maintenance process, the chamber state component 714 may determine that the initialization process and / or maintenance process has been completed.

[0093] In response to the process logic determining that the conditioning process is complete, the method 600 proceeds to block 818. At block 818, the process logic causes a production substrate to be processed in the process chamber. In response to the process logic determining that the conditioning process is not complete, the method 600 proceeds to block 822, described below. In some embodiments, the chamber conditioning engine 710 can send a notification to the system controller 228 indicating that conditions in the process chamber 214 are suitable for processing production substrates. The system controller 228 can, in some embodiments, begin processing the production substrate in response to receiving the notification. In additional or alternative embodiments, the chamber conditioning engine 710 and / or the system controller 228 can send a notification to the client device 120 indicating that conditions in the process chamber 214 are suitable for processing production substrates. The client device 120 can provide the notification to a user of the client device (e.g., an operator, an engineer, etc.) via a graphical user interface (GUI). Based on the notification, an operator and / or engineer can initiate processing of the production substrates (e.g., via interaction with the GUI) and / or have one or more conditioning substrates in the process tool 204 transferred to additional testing equipment (e.g., metrology equipment 128) for further testing and / or validation.

[0094] Referring again to block 814, in response to the process logic determining that the chamber condition criteria are not met, the method 600 proceeds to block 820. In block 820, the process logic updates one or more second operations of the chamber conditioning process (e.g., initialization process and / or maintenance process). In some embodiments, the process updater component 716 can determine one or more updates to make to a process recipe associated with the initialization process and / or maintenance process based on the determined condition for the process chamber 214. In an illustrative example, the spectral data 752 can indicate an etch rate of the substrate 202, which depends on one or more conditions of the process chamber 214. If the prescribed etch rate does not correspond to the target etch rate for the target condition of the process chamber 214, the process updater component can determine one or more settings of the process recipe to modify to bring the etch rate into correspondence with the target etch rate. In some embodiments, the process updater component 716 can store an indication of the updated settings in memory 750 as process recipe data 756. In additional or alternative embodiments, the process updater component 716 can send a notification to the client device 120 indicating that the chamber conditions of the process chamber 214 do not correspond to the target conditions. The client device 120 can provide a notification to an operator and / or engineer of the manufacturing system 200 (e.g., via a GUI of the client device 120). The operator and / or engineer can provide an indication of one or more updates to be made to the settings of the process recipe to bring the conditions of the process chamber 214 into correspondence with the target conditions. The client device 120 can send a notification to the system controller 228 and / or the process updater component 716 indicating the updated settings. The updated settings, in some embodiments, can be stored in memory 750 as process recipe data 756.

[0095] At block 822, the process logic causes one or more second operations of the chamber conditioning process to be performed. The process logic may cause a second substrate (e.g., a conditioning substrate) to be processed according to the one or more second operations. In some embodiments, the process updater component 716 may update one or more settings of the process recipe for the initialization process and / or maintenance process, as described with respect to block 820, to cause the state of the process chamber 214 to correspond to a target state of the process chamber 214. In such embodiments, the system controller 228 may initiate one or more additional conditioning operations to be performed on one or more additional conditioning substrates according to the updated settings. After performing the additional conditioning operations, the substrate measurement subsystem 300 may collect spectral data associated with the conditioning substrate 202, and the chamber state engine 710 may determine the state of the process chamber 214, as described above.

[0096] As described with respect to block 816, in response to the process logic determining that the conditioning process is not complete, the method 600 proceeds to block 822. In such an embodiment, the system controller 228 may perform an additional conditioning operation in the process chamber 214 according to the original (e.g., unupdated) settings for the conditioning operation. After performing the additional conditioning operation, the substrate measurement subsystem 300 may collect spectral data associated with the conditioned substrate 202, and the chamber state engine 710 may determine the state of the process chamber 214, as described above.

[0097] 9 is a flow diagram of an exemplary method for training a machine learning model according to an aspect of the present disclosure. Method 900 is performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (such as running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one implementation, method 900 may be performed by a computer system such as computer system architecture 100 of FIG. 1. In other or similar implementations, one or more operations of method 900 may be performed by one or more other machines not shown in the figure. In some aspects, one or more operations of method 900 may be performed by training set generator 172.

[0098] At block 910, process logic initializes training set T to be empty (e.g., {}). At block 912, process logic identifies spectral data associated with one or more substrates processed in the manufacturing system according to a chamber conditioning process. In some embodiments, the spectral data may be generated by substrate measurement subsystem 300 for one or more substrates (e.g., substrate 202) processed according to the conditioning operations of the initialization and / or maintenance processes in one or more of process chambers 214, 216, 218 according to the embodiments described above. In some embodiments, components of substrate measurement subsystem 300 may collect spectral data while conditioning operations are being performed in the respective process chambers (e.g., process chamber 214). Thus, the spectral data collected for substrate 202 may indicate the condition of substrate 202 and / or the condition of process chamber 214 during the conditioning operation. In other or similar embodiments, one or more substrates 202 may be transferred from process chamber 214 to substrate measurement subsystem 300 after the conditioning operation is completed, as described above.

[0099] At block 914, the process logic obtains metrology data indicative of metrology measurements for one or more substrates. In some embodiments, the system controller 228 may cause one or more robot arms of the manufacturing system 200 to transfer the substrate 202 from the process chamber 214 to the metrology device 128. In some embodiments, the metrology device 128 may be included in or otherwise coupled to the process tool 204. In such embodiments, the robot arm may transfer the substrate 202 from the process chamber 214 to the metrology device 128. In other or similar embodiments, the metrology device 128 may be located external to the process tool 204. In such embodiments, the robot arm may transfer the substrate 202 to the factory interface 206. The substrate 202 may be transferred from the factory interface 206 to the metrology device 128 (e.g., manually by an operator or engineer of the manufacturing system 200). The metrology device 128 may collect metrology data for the substrate 202 and provide the collected metrology data to the system controller 228 and / or another computing system of the system 100, as described herein.

[0100] At block 916, the process logic determines whether the chamber conditions of the process chamber satisfy one or more criteria based on the acquired metrology data. In some embodiments, the process logic determines whether the chamber conditions of the process chamber satisfy the criteria by determining whether the metrology data collected for the substrate corresponds to (e.g., matches or substantially matches) target metrology data collected for a substrate processed in a chamber that satisfies the chamber conditions. The target metrology data may be determined based on historical or experimental data and / or may be provided by an operator and / or engineer of the manufacturing system 200 (e.g., via the client device 120). If the collected metrology data corresponds to the target metrology data, the process logic determines that the criteria have been met. If the collected metrology data does not correspond to the target metrology data, the process logic determines that the criteria have not been met. At block 918, the process logic generates an input / output mapping, where the input is based on the spectral data and the output indicates whether the chamber conditions of the process chamber have been met. In an illustrative example, the input / output mapping may include an indication of the spectral data collected for the substrate 202 mapped to an indication of whether the conditions of the chamber that processed the substrate 202 met one or more criteria.

[0101] At block 920, processing logic adds the input / output mapping to training set T. At block 922, processing logic determines whether set T is sufficient for training. In response to processing logic determining that training set T is not sufficient for training, method 900 returns to block 912. In response to processing logic determining that training set T is sufficient for training, method 900 proceeds to block 924. At block 924, processing logic provides training set T for training a machine learning model (e.g., model 190).

[0102] For ease of explanation, methods 500, 600, 800, and 900 are shown and described as a series of acts. However, acts in accordance with the present disclosure can be performed in various orders and / or simultaneously, as well as with other acts not shown and described herein. Moreover, not all illustrated acts can be performed to implement a method in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that a method can alternatively be represented as a series of interrelated states via a state diagram or events. Furthermore, it should be understood that the methods disclosed herein can be stored on an article of manufacture to facilitate transport and transfer of such methodologies to a computing device. The term article of manufacture, as used herein, is intended to encompass a computer program accessible from any computer-readable device or storage medium.

[0103] FIG. 10 illustrates a block diagram of an exemplary computer system 1000 that operates in accordance with one or more aspects of the present disclosure. In alternative embodiments, the machine may be connected (e.g., networked) to other machines in a local area network (LAN), an intranet, an extranet, or the Internet. The machine may operate as a server or a client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), a tablet computer, a set-top box (STB), a personal digital assistant (PDA), a mobile phone, a web appliance, a server, a network router, a switch, or a bridge, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by the machine. Furthermore, while only a single machine is shown, the term “machine” is intended to include any collection of machines (e.g., computers) that individually or jointly execute a set (or sets) of instructions to perform any one or more of the methodologies discussed herein. In an embodiment, computing device 1000 may correspond to prediction component 114 and / or spectral data engine 132 of FIG. 1, system controller 228 of FIG. 2, or controller 330 of FIG.

[0104] The exemplary computing device 1000 includes a processing unit 1002, a main memory 1004 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM), etc.), a static memory 1006 (e.g., flash memory, static random access memory (SRAM), etc.), and a secondary memory (e.g., data storage device 1028), which communicate with each other via a bus 1008.

[0105] The processing device 1002 may represent one or more general-purpose processors, such as a microprocessor, a central processing unit, or the like. More specifically, the processing device 1002 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing a combination of instruction sets. The processing device 1002 may 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), a network processor, or the like. The processing device 1002 may be or include a system-on-chip (SoC), a programmable logic controller (PLC), or other type of processing device. The processing device 1002 is configured to execute processing logic for performing the operations and steps discussed herein.

[0106] Computing device 1000 may further include a network interface device 1022 for communicating with a network 1064. Computing device 1000 may also include a video display unit 1010 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1012 (e.g., a keyboard), a cursor control device 1014 (e.g., a mouse), and a signal generating device 1020 (e.g., a speaker).

[0107] The data storage device 1028 may include a machine-readable storage medium (or more specifically, a non-transitory computer-readable storage medium) 1024 having stored thereon a set of one or more instructions 1026 that embody any one or more of the methodologies or functions described herein. Here, non-transitory storage medium refers to a storage medium other than a carrier wave. The instructions 1026 may also reside, completely or at least partially, within the main memory 1004 and / or within the processing unit 1002 during execution thereof by the computing device 1000, with the main memory 1004 and the processing unit 1002 also constituting computer-readable storage media.

[0108] The computer-readable storage medium 1024 may also be used to store the model 190 and data used to train the model 190. The computer-readable storage medium 1024 may also store a software library containing methods for invoking the model 190. While the computer-readable storage medium 1024 is shown to be a single medium in the exemplary embodiment, the term "computer-readable storage medium" should be interpreted to include a single medium or multiple media (e.g., centralized or distributed databases, and / or associated caches and servers) that store one or more sets of instructions. The term "computer-readable storage medium" should also be interpreted to include any medium capable of storing or encoding a set of instructions for execution by a machine and causing the machine to perform any one or more of the methodologies of the present disclosure. Accordingly, the term "computer-readable storage medium" should be interpreted to include, but not limited to, solid-state memory, and optical and magnetic media.

[0109] The foregoing description sets forth numerous specific details, such as examples of specific systems, components, methods, etc., to provide a thorough understanding of some embodiments of the present disclosure. However, it will be apparent to those skilled in the art that at least some embodiments of the present disclosure may be practiced without these specific details. In other instances, well-known components or methods have not been described in detail or have been presented in simple block diagram form to avoid unnecessarily obscuring the present disclosure. Thus, the specific details described are merely exemplary. Particular embodiments may vary from these example details and still be construed as being within the scope of the present disclosure.

[0110] Throughout this specification, a reference to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment is included in at least one embodiment. Thus, the appearances of the phrase "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. In addition, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or." When the term "about" or "approximately" is used herein, it is intended to mean that the stated nominal value is accurate to within ±10%.

[0111] Although the operations of the methods herein are shown and described in a particular order, the order of the operations of each method may be changed, certain operations may be performed in reverse order, and certain operations may be performed at least in part concurrently with other operations. In alternative embodiments, instructions or sub-operations of separate operations may be performed intermittently and / or alternately.

[0112] It is to be understood that the above description is intended to be illustrative, and not limiting. Many other embodiments will be apparent to those skilled in the art upon reading and understanding the above description. The scope of the present disclosure should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

[0113] The following section contains examples of embodiments of the present disclosure. Such examples are provided for illustrative purposes only and are not intended to be limiting.

[0114] Example 1 is a method that includes identifying spectral data associated with a substrate in a manufacturing system, where the substrate has been processed according to one or more first operations of a process recipe, the process recipe being unknown to a system controller of the manufacturing system; determining a process recipe associated with the substrate based on the identified spectral data; identifying one or more second operations of the process recipe that have not yet been performed on the substrate; and processing the substrate according to the one or more second operations of the process recipe.

[0115] Example 2 is the method of example 1, wherein determining a process recipe associated with the substrates based on the identified spectral data includes: providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on the given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates; obtaining one or more outputs of the machine learning model; and extracting, from the one or more outputs of the machine learning model, data indicative of the process recipe associated with the substrates and one or more first operations of the process recipe that have already been performed on the substrates.

[0116] Example 3 is the method of example 2, wherein identifying one or more second operations of the process recipe that have not yet been performed on the substrate includes: identifying a set of operations associated with the process recipe, the set of operations including one or more first operations and one or more second operations; and extracting the one or more second operations from the identified set of operations associated with the process recipe.

[0117] Example 4 is the method of example 2, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the process recipe during previous substrate processing performed in the manufacturing system.

[0118] Example 5 is the method of example 1, wherein the spectral data is generated based on one or more measurements of a substrate metrology subsystem for a processing tool that includes a process chamber used to process the substrate.

[0119] Example 6 is the method of example 5, wherein the substrate measurement subsystem is configured to collect spectral data associated with the substrate while the substrate is in the process chamber.

[0120] Example 7 is the method of example 1, wherein causing the substrate to be processed according to one or more second operations of the process recipe includes sending a notification to a system controller of the manufacturing system that includes instructions for the one or more second operations of the process recipe.

[0121] Example 8 is the method of example 1, wherein causing the substrate to be processed in accordance with the one or more second operations of the process recipe includes sending, by the system controller, a signal to at least one component of the manufacturing system to cause the component to begin performing the one or more second operations of the process recipe.

[0122] Example 9 is the method of example 1, further comprising detecting that one or more components of the manufacturing system are inoperable for a period of time prior to identifying the spectral data associated with the substrate in the manufacturing system.

[0123] Example 10 is a system including a memory and a processing device coupled to the memory, the processing device for performing operations including identifying spectral data associated with a substrate in a manufacturing system, the substrate having been processed according to one or more first operations of a process recipe, the process recipe being unknown to a system controller of the manufacturing system; determining a process recipe associated with the substrate based on the identified spectral data; identifying one or more second operations of the process recipe that have not yet been performed on the substrate; and processing the substrate according to the one or more second operations of the process recipe.

[0124] Example 11 is the system of example 10, wherein determining a process recipe associated with the substrate based on the identified spectral data includes providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on the given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates; obtaining one or more outputs of the machine learning model; and extracting data indicative of the process recipe associated with the substrate and one or more first operations of the process recipe that have already been performed on the substrate from the one or more outputs of the machine learning model.

[0125] Example 12 is the system of example 11, wherein identifying one or more second operations of the process recipe that have not yet been performed on the substrate includes identifying a set of operations associated with the process recipe, the set of operations including one or more first operations and one or more second operations; and extracting one or more second operations from the identified set of operations associated with the process recipe.

[0126] Example 13 is the system of example 11, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the process recipe during previous substrate processing performed in the manufacturing system.

[0127] Example 14 is the system of example 10, wherein the spectral data is generated based on one or more measurements of a substrate measurement subsystem for a processing tool that includes a process chamber used to process the substrate.

[0128] Example 15 is the system of example 14, wherein the substrate measurement subsystem is configured to collect spectral data associated with the substrate while the substrate is in a process chamber of the manufacturing system.

[0129] Example 16 is the system of example 10, wherein causing the substrate to be processed according to one or more second operations of the process recipe includes sending a notification to a system controller of the manufacturing system that includes instructions for the one or more second operations of the process recipe.

[0130] Example 17 is a non-transitory computer-readable medium comprising instructions that, when executed by a processing device, cause the processing device to perform operations including identifying spectral data associated with a substrate in a manufacturing system, the substrate having been processed according to one or more first operations of a process recipe, the process recipe being unknown to a system controller of the manufacturing system; determining a process recipe associated with the substrate based on the identified spectral data; identifying one or more second operations of the process recipe that have not yet been performed on the substrate; and processing the substrate according to the one or more second operations of the process recipe.

[0131] Example 18 is the non-transitory computer-readable medium of example 17, wherein determining a process recipe associated with the substrates based on the identified spectral data includes providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on the given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates; obtaining one or more outputs of the machine learning model; and extracting, from the one or more outputs of the machine learning model, data indicative of the process recipe associated with the substrates and one or more first operations of the process recipe that have already been performed on the substrates.

[0132] Example 19 is the non-transitory computer-readable medium of example 18, in which identifying one or more second operations of the process recipe that have not yet been performed on the substrate includes identifying a set of operations associated with the process recipe, the set of operations including one or more first operations and one or more second operations; and extracting the one or more second operations from the identified set of operations associated with the process recipe.

[0133] Example 20 is the non-transitory computer-readable medium of example 18, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the process recipe during previous substrate processing performed in the manufacturing system.

[0134] Example 21 is a method that includes identifying spectral data associated with a first substrate in a manufacturing system, the first substrate being processed in a process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber; determining a state of the process chamber after performance of the one or more first operations based on the identified spectral data; determining whether the state of the process chamber satisfies one or more chamber condition criteria; and, in response to determining that the state of the process chamber does not satisfy the one or more chamber condition criteria, updating one or more second operations of the chamber conditioning process so that the state of the process chamber satisfies the one or more chamber condition criteria; and processing a second substrate in the process chamber according to the updated one or more second operations.

[0135] Example 22 is the method of example 21, in which determining the state of the process chamber after performance of the one or more first operations based on the identified spectral data includes providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict the state of each process chamber of the manufacturing system that processed the substrates based on the given spectral data for the respective substrates; obtaining one or more outputs of the machine learning model; and extracting data indicative of the state of the process chamber that processed the first substrate according to the one or more first operations of the chamber conditioning process from the one or more outputs of the machine learning model.

[0136] Example 23 is the method of example 22, wherein the data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations includes an indication of an amount of drift between the spectral data associated with the first substrate and target spectral data associated with one or more previous substrates processed according to the one or more first operations of the chamber condition process, and wherein each process chamber that processed the one or more previous substrates satisfied one or more chamber condition criteria.

[0137] Example 24 is the method of example 23, wherein determining whether the process chamber condition satisfies one or more chamber condition criteria includes determining whether an amount of drift between the spectral data associated with the first substrate and the target spectral data exceeds a threshold amount of drift.

[0138] Example 25 is the method of example 22, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with a chamber conditioning process performed in the manufacturing system.

[0139] Example 26 is the method of Example 21, further including: in response to determining that the condition of the process chamber satisfies the one or more chamber condition criteria, determining whether the chamber conditioning process is complete based on the condition of the process chamber; and in response to determining that the chamber conditioning process is not complete, processing a second substrate in the process chamber according to one or more second operations.

[0140] Example 27 is the method of example 21, wherein the spectral data is generated based on one or more measurements of a substrate metrology subsystem for a processing tool that includes a process chamber.

[0141] Example 28 is the method of example 27, wherein the substrate measurement subsystem is configured to collect spectral data associated with the substrate while the substrate is in the process chamber.

[0142] Example 29 is a system including a memory and a processing device coupled to the memory, wherein the processing device performs operations including identifying spectral data associated with a first substrate in a manufacturing system, the first substrate being processed in a process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber; determining a state of the process chamber after performance of the one or more first operations based on the identified spectral data; determining whether the state of the process chamber satisfies one or more chamber state criteria; and, in response to determining that the state of the process chamber does not satisfy the one or more chamber state criteria, updating one or more second operations of the chamber conditioning process so that the state of the process chamber satisfies the one or more chamber state criteria; and processing a second substrate in the process chamber according to the updated one or more second operations.

[0143] Example 30 is the system of example 29, wherein determining the state of the process chamber after performance of the one or more first operations based on the identified spectral data includes providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict the state of each process chamber of the manufacturing system that processed the substrates based on the given spectral data of the respective substrates; obtaining one or more outputs of the machine learning model; and extracting data indicative of the state of the process chamber that processed the first substrate according to the one or more first operations of the chamber conditioning process from the one or more outputs of the machine learning model.

[0144] Example 31 is the system of example 30, wherein the data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations includes an indication of an amount of drift between the spectral data associated with the first substrate and target spectral data associated with one or more previous substrates processed according to the one or more first operations of the chamber condition process, and each process chamber that processed the one or more previous substrates met one or more chamber condition criteria.

[0145] Example 32 is the system of example 31, wherein determining whether the process chamber condition satisfies one or more chamber condition criteria includes determining whether an amount of drift between the spectral data associated with the first substrate and the target spectral data exceeds a threshold amount of drift.

[0146] Example 33 is the system of example 30, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with a chamber conditioning process performed in the manufacturing system.

[0147] Example 34 is the system of example 29, wherein the operation further includes, in response to determining that the condition of the process chamber satisfies one or more chamber condition criteria, determining whether the chamber conditioning process is complete based on the condition of the process chamber, and, in response to determining that the chamber conditioning process is not complete, processing a second substrate in the process chamber according to one or more second operations.

[0148] Example 35 is the system of example 29, wherein the spectral data is generated based on one or more measurements of a substrate measurement subsystem for a processing tool that includes a process chamber.

[0149] Example 36 is a non-transitory computer-readable medium containing instructions that, when executed by a processing device, cause a processing device to perform operations including identifying spectral data associated with a first substrate in a manufacturing system, the first substrate being processed in a process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber; determining a state of the process chamber after performance of the one or more first operations based on the identified spectral data; determining whether the state of the process chamber satisfies one or more chamber condition criteria; and, in response to determining that the state of the process chamber does not satisfy the one or more chamber condition criteria, updating one or more second operations of the chamber conditioning process so that the state of the process chamber satisfies the one or more chamber condition criteria; and processing a second substrate in the process chamber according to the updated one or more second operations.

[0150] Example 37 is the non-transitory computer-readable medium of Example 36, in which determining the state of the process chamber after performance of the one or more first operations based on the identified spectral data includes providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict the state of each process chamber of the manufacturing system that processed the substrates based on the given spectral data for the respective substrates; obtaining one or more outputs of the machine learning model; and extracting data indicative of the state of the process chamber that processed the first substrate according to the one or more first operations of the chamber conditioning process from the one or more outputs of the machine learning model.

[0151] Example 38 is the non-transitory computer-readable medium of example 37, in which the data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations includes an indication of an amount of drift between spectral data associated with the first substrate and target spectral data associated with one or more previous substrates processed according to the one or more first operations of the chamber condition process, and each process chamber that processed the one or more previous substrates satisfied one or more chamber condition criteria.

[0152] Example 39 is the non-transitory computer-readable medium of example 38, wherein determining whether the process chamber conditions satisfy one or more chamber condition criteria includes determining whether an amount of drift between the spectral data associated with the first substrate and the target spectral data exceeds a threshold amount of drift.

[0153] Example 40 is the non-transitory computer-readable medium of example 37, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with a chamber conditioning process performed in the manufacturing system.

Claims

1. identifying spectral data associated with substrates processed in a manufacturing system according to one or more first operations of a process recipe, the process recipe identifying spectral data unknown to a system controller of the manufacturing system; determining the process recipe associated with the substrate based on the identified spectral data; identifying one or more second operations of the process recipe for the substrate; processing the substrate according to the one or more second operations of the process recipe; A method comprising:

2. determining the process recipe associated with the substrate based on the identified spectral data; providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates; receiving one or more outputs of the machine learning model; extracting data from the one or more outputs of the machine learning model indicative of the process recipe associated with the substrate and the one or more first operations of the process recipe that have already been performed on the substrate; The method of claim 1 , comprising:

3. identifying the one or more second operations of the process recipe for the substrate; identifying a set of operations associated with the process recipe, the set of operations including the one or more first operations and the one or more second operations; extracting the one or more second operations from the set of identified operations associated with the process recipe; The method of claim 2 , comprising:

4. 3. The method of claim 2, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the process recipe during previous substrate processing performed in the manufacturing system.

5. The method of claim 1 , wherein the spectral data is generated based on one or more measurements of a substrate metrology subsystem for a processing tool that includes a process chamber used to process the substrate.

6. The method of claim 5 , wherein the substrate metrology subsystem is configured to collect the spectral data associated with the substrate while the substrate is in the process chamber.

7. causing the substrate to be processed according to the one or more second operations of the process recipe; sending a notification to the system controller of the manufacturing system, the notification including instructions for the one or more second operations of the process recipe; The method of claim 1 , comprising:

8. causing the substrate to be processed according to the one or more second operations of the process recipe; sending, by the system controller, a signal to at least one component of the manufacturing system to cause the component to begin performing the one or more second operations of the process recipe; The method of claim 1 , comprising:

9. detecting that one or more components of the manufacturing system are inoperable for a period of time prior to identifying the spectral data associated with the substrate at the manufacturing system; The method of claim 1 further comprising:

10. Memory and a processing unit coupled to the memory; a processing unit configured to: identifying spectral data associated with substrates processed in a manufacturing system according to one or more first operations of a process recipe, the process recipe identifying spectral data unknown to a system controller of the manufacturing system; determining the process recipe associated with the substrate based on the identified spectral data; identifying one or more second operations of the process recipe for the substrate; processing the substrate according to the one or more second operations of the process recipe; for performing operations including system.

11. determining the process recipe associated with the substrate based on the identified spectral data; providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates; obtaining one or more outputs of the machine learning model; extracting data indicative of the process recipe associated with the substrate and the one or more first operations of the process recipe that have already been performed on the substrate from one or more outputs of the machine learning model; The system of claim 10, comprising:

12. identifying the one or more second operations of the process recipe for the substrate; identifying a set of operations associated with the process recipe, the set of operations including the one or more first operations and the one or more second operations; extracting the one or more second operations from the set of identified operations associated with the process recipe; The system of claim 11 , comprising:

13. 12. The system of claim 11 , wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the process recipe during a previous substrate process performed in the manufacturing system.

14. The system of claim 10 , wherein the spectral data is generated based on one or more measurements of a substrate metrology subsystem for a processing tool that includes a process chamber used to process the substrate.

15. 15. The system of claim 14, wherein the substrate metrology subsystem is configured to collect the spectral data associated with the substrate while the substrate is in a process chamber of the manufacturing system.

16. causing the substrate to be processed according to the one or more second operations of the process recipe; sending a notification to the system controller of the manufacturing system, the notification including instructions for the one or more second operations of the process recipe; The system of claim 10, comprising:

17. When executed by the processing unit, identifying spectral data associated with substrates processed in a manufacturing system according to one or more first operations of a process recipe, the process recipe identifying spectral data unknown to a system controller of the manufacturing system; determining the process recipe associated with the substrate based on the identified spectral data; identifying one or more second operations of the process recipe for the substrate; processing the substrate according to the one or more second operations of the process recipe; A non-transitory computer-readable medium comprising instructions that cause the processing device to perform operations including:

18. determining the process recipe associated with the substrate based on the identified spectral data; providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on given spectral data, a respective process recipe associated with one or more substrates in the manufacturing system and one or more operations of the respective process recipe that have already been performed on the one or more substrates; obtaining one or more outputs of the machine learning model; extracting data indicative of the process recipe associated with the substrate and the one or more first operations of the process recipe that have already been performed on the substrate from one or more outputs of the machine learning model; 20. The non-transitory computer-readable medium of claim 17, comprising:

19. identifying the one or more second operations of the process recipe for the substrate; identifying a set of operations associated with the process recipe, the set of operations including the one or more first operations and the one or more second operations; extracting the one or more second operations from the set of identified operations associated with the process recipe; 20. The non-transitory computer-readable medium of claim 18, comprising:

20. 20. The non-transitory computer-readable medium of claim 18, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the process recipe during a previous substrate process performed in the manufacturing system.

21. Identifying spectral data associated with a first substrate in a manufacturing system, the first substrate being processed in a process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber; determining a state of the process chamber after performing the one or more first operations based on the identified spectral data; and determining whether the condition of the process chamber satisfies one or more chamber condition criteria; In response to determining that the condition of the process chamber does not satisfy the one or more chamber condition criteria, updating one or more second operations of the chamber conditioning process so that the condition of the process chamber satisfies the one or more chamber condition criteria; processing a second substrate in the process chamber according to the updated one or more second operations; A method comprising:

22. determining the state of the process chamber after performing the one or more first operations based on the identified spectral data; providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on given spectral data for each substrate, a state of each process chamber of the manufacturing system that processed the substrate; obtaining one or more outputs of the machine learning model; extracting, from the one or more outputs of the machine learning model, data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations of the chamber conditioning process; 22. The method of claim 21, comprising:

23. 23. The method of claim 22, wherein the data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations comprises an indication of an amount of drift between the spectral data associated with the first substrate and target spectral data associated with one or more previous substrates processed according to the one or more first operations of the chamber condition process, and wherein each process chamber that processed the one or more previous substrates met the one or more chamber condition criteria.

24. Determining whether the conditions of the process chamber satisfy the one or more chamber condition criteria determining whether the drift between the spectral data associated with the first substrate and the target spectral data exceeds a threshold drift; 24. The method of claim 23, comprising:

25. 23. The method of claim 22, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the chamber conditioning process performed in the manufacturing system.

26. determining whether the chamber conditioning process is complete based on the condition of the process chamber in response to determining that the condition of the process chamber satisfies the one or more chamber condition criteria; and in response to determining that the chamber conditioning process is not complete, processing the second substrate in the process chamber according to the one or more second operations; 22. The method of claim 21 further comprising:

27. 22. The method of claim 21 , wherein the spectral data is generated based on one or more measurements of a substrate metrology subsystem for a processing tool that includes the process chamber.

28. 28. The method of claim 27, wherein the substrate metrology subsystem is configured to collect the spectral data associated with the substrate while the substrate is in the process chamber.

29. Memory and a processing unit coupled to the memory; a processing unit configured to: Identifying spectral data associated with a first substrate in a manufacturing system, the first substrate being processed in a process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber; determining a state of the process chamber after performing the one or more first operations based on the identified spectral data; and determining whether the condition of the process chamber satisfies one or more chamber condition criteria; In response to determining that the condition of the process chamber does not satisfy the one or more chamber condition criteria, updating one or more second operations of the chamber conditioning process so that the condition of the process chamber satisfies the one or more chamber condition criteria; processing a second substrate in the process chamber according to the updated one or more second operations; for performing operations including system.

30. determining the state of the process chamber after performing the one or more first operations based on the identified spectral data; providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on given spectral data for each substrate, a state of each process chamber of the manufacturing system that processed the substrate; obtaining one or more outputs of the machine learning model; extracting, from the one or more outputs of the machine learning model, data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations of the chamber conditioning process; 30. The system of claim 29, comprising:

31. 31. The system of claim 30, wherein the data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations comprises an indication of an amount of drift between the spectral data associated with the first substrate and target spectral data associated with one or more previous substrates processed according to the one or more first operations of the chamber condition process, and wherein each process chamber that processed the one or more previous substrates met the one or more chamber condition criteria.

32. Determining whether the conditions of the process chamber satisfy the one or more chamber condition criteria determining whether the drift between the spectral data associated with the first substrate and the target spectral data exceeds a threshold drift; 32. The system of claim 31 , comprising:

33. 31. The system of claim 30, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the chamber conditioning process performed in the manufacturing system.

34. The operation is determining whether the chamber conditioning process is complete based on the condition of the process chamber in response to determining that the condition of the process chamber satisfies the one or more chamber condition criteria; and in response to determining that the chamber conditioning process is not complete, processing the second substrate in the process chamber according to the one or more second operations; 30. The system of claim 29, further comprising:

35. 30. The system of claim 29, wherein the spectral data is generated based on one or more measurements of a substrate metrology subsystem for a processing tool that includes the process chamber.

36. When executed by the processing unit, identifying spectral data associated with a first substrate in a manufacturing system, the first substrate being processed in a process chamber of the manufacturing system according to one or more first operations of a chamber conditioning process for the process chamber; determining a state of the process chamber after performing the one or more first operations based on the identified spectral data; and determining whether the condition of the process chamber satisfies one or more chamber condition criteria; In response to determining that the condition of the process chamber does not satisfy the one or more chamber condition criteria, updating one or more second operations of the chamber conditioning process so that the condition of the process chamber satisfies the one or more chamber condition criteria; processing a second substrate in the process chamber according to the updated one or more second operations; A non-transitory computer-readable medium comprising instructions that cause the processing device to perform operations including:

37. determining the state of the process chamber after performing the one or more first operations based on the identified spectral data; providing the identified spectral data as input to a machine learning model, the machine learning model being trained to predict, based on given spectral data for each substrate, a state of each process chamber of the manufacturing system that processed the substrate; obtaining one or more outputs of the machine learning model; extracting, from the one or more outputs of the machine learning model, data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations of the chamber conditioning process; 37. The non-transitory computer-readable medium of claim 36, comprising:

38. 38. The non-transitory computer-readable medium of claim 37, wherein the data indicative of the condition of the process chamber that processed the first substrate according to the one or more first operations includes an indication of an amount of drift between the spectral data associated with the first substrate and target spectral data associated with one or more previous substrates processed according to the one or more first operations of the chamber condition process, and wherein each process chamber that processed the one or more previous substrates met the one or more chamber condition criteria.

39. Determining whether the conditions of the process chamber satisfy the one or more chamber condition criteria determining whether the drift between the spectral data associated with the first substrate and the target spectral data exceeds a threshold drift; 39. The non-transitory computer-readable medium of claim 38, comprising:

40. 38. The non-transitory computer-readable medium of claim 37, wherein the machine learning model is trained using historical spectral data collected for one or more previous substrates processed according to at least one operation associated with the chamber conditioning process performed in the manufacturing system.

Images