Process characterization and correction using optical wall process sensor (OWPS)

The optical sensor obtains the spectral data of the chamber wall and updates the process operating parameters, solving the problem of insufficient monitoring of the interior surface conditions of the chamber in traditional systems, improving process efficiency and yield, and ensuring higher accuracy and throughput.

CN118922926APending Publication Date: 2024-11-08APPLIED MATERIALS INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380027752.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-03-16
Filing Date
2023-03-15
Publication Date
2024-11-08

Smart Images

  • Figure CN118922926A_ABST
    Figure CN118922926A_ABST
Patent Text Reader

Abstract

A method includes receiving, by a processing device, first sensor data indicative of a state of a wall corresponding to a first processing chamber. The first sensor data includes optical spectral data. The method further includes determining, by the processing device, a first value based on the first sensor data. The first value corresponds to a first amount of product disposed along a surface of the wall at a first time. The method further includes determining, by the processing device, a first update to a first process operation associated with the first processing chamber based on the first value. The method further includes performing, by the processing device, one or more of (i) preparing a notification indicative of the first update for presentation on a graphical user interface (GUI), or (ii) causing the first process operation to be performed in accordance with the first update.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present disclosure relate to process control and process learning using optical sensors. Background Art

[0002] Variations in conditions on processing chamber surfaces affect various processing parameters. For example, redeposition of etch byproducts on chamber walls may change the etch rate for a given process. As a result, the etch rate (or other process parameters or conditions) may change as substrates are processed in the chamber and result in non-uniform processing between substrates, among other conditions.

[0003] Conventional systems lack reliable methods to monitor surface conditions within a process chamber, including surfaces of liners (or inner walls), lids, electrostatic chucks (ESCs), process rings, and the like. For example, the chemical, physical, and thermal conditions of a process chamber liner are known to affect plasma processes by affecting degassing recombination near the liner. Some monitoring methods, such as capacitive monitoring or resonant frequency monitoring, are being developed, but these methods are affected by thermal or RF noise, especially during processing.

[0004] Further, due to the lack of information, such as optical information regarding the state of the processing chamber walls, substrate processing inefficiencies and / or inaccuracies may occur, particularly as processing equipment ages. For example, process transitions may sometimes result in changes in the amount of deposition (e.g., deposition thickness) or changes in the composition of the deposition, both of which may result in frequent scrapping of unacceptable batches of processed substrates. In another example, knowing when to run a cleaning process within a processing chamber may be based solely on the time in the process, which is little more than a guess. However, running a cleaning process too frequently may affect substrate throughput, while running a cleaning process too infrequently may affect product quality. Other processing inefficiencies or inaccuracies exist, which will be discussed in detail. Summary of the invention

[0005] The following is a simplified summary of the present disclosure to provide a basic understanding of some aspects of the present disclosure. This summary is not an extensive overview of the present disclosure. It is not intended to identify key or important elements of the present disclosure, nor is it intended to delineate any scope of a particular embodiment of the present disclosure or any scope of the claims. Its sole purpose is to present some concepts of the present disclosure in a simplified form as a prelude to a more detailed description presented later.

[0006] Some embodiments described herein encompass a method comprising: receiving, by a processing device, first sensor data indicating a state of a wall corresponding to a first processing chamber. The first sensor data comprises optical spectral data. The method further comprises: determining, by the processing device, a first value based on the first sensor data. The first value corresponds to a first amount of a product disposed along a surface of the wall at a first time. The method further comprises: determining, by the processing device, a first update to a first process operation associated with the first processing chamber based on the first value. The method further comprises: performing, by the processing device, one or more of (i) preparing a notification indicating the first update for presentation on a graphical user interface (GUI), or (ii) causing the first process operation to be performed in accordance with the first update.

[0007] Additional or related embodiments described herein encompass a system comprising: a processing chamber comprising a first wall having an inner surface positioned within the processing chamber. The system further comprises an optical sensor coupled to the processing chamber. The optical sensor is configured to acquire spectral data associated with the inner surface. The system further comprises a memory and a processing device coupled to the memory. The processing device is configured to: receive first sensor data indicating a state of a wall corresponding to a first processing chamber. The first sensor data comprises optical spectral data. The processing device is further configured to: determine a first value based on the first sensor data. The first value corresponds to a first amount of a product disposed along a surface of the wall at a first time. The processing device is further configured to: determine a first update to a first process operation associated with the first processing chamber based on the first value. The processing device is further configured to: perform one or more of the following: (i) prepare a notification indicating the first update for presentation on a GUI, or (ii) cause the first process operation to be performed in accordance with the first update.

[0008] Additional or related embodiments described herein encompass a non-transitory machine-readable storage medium storing instructions that, when executed, cause a processing device to: receive first sensor data indicating a state of a wall corresponding to a first processing chamber. The first sensor data includes optical spectral data. The processing device is further configured to: determine a first value based on the first sensor data. The first value corresponds to a first amount of a product disposed along a surface of the wall at a first time. The processing device is further configured to: determine a first update to a first process operation associated with the first processing chamber based on the first value. The processing device is further configured to: perform one or more of the following: (i) prepare a notification indicating the first update for presentation on a GUI, or (ii) cause the first process operation to be performed in accordance with the first update.

[0009] According to these and other aspects of the present disclosure, numerous other features are provided. Other features and aspects of the present disclosure will become more fully apparent from the following detailed description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The present disclosure is illustrated by way of example and not limitation in the figures of the accompanying drawings in which like reference numerals represent like elements. It should be noted that different references to "a / one" embodiment in the present disclosure do not necessarily refer to the same embodiment, and such references refer to at least one.

[0011] Figure 1 is a schematic top view of an example processing system according to some embodiments.

[0012] Figure 2 is a simplified side view of a system for monitoring a processing chamber in accordance with some embodiments.

[0013] Figure 3 According to some embodiments, Figure 2 An enlarged portion of a side view of a system showing the transparent crystals and deposited transparent films embedded within the walls and liners of the processing chamber.

[0014] 4 is a block diagram illustrating an exemplary system architecture in which embodiments of the present disclosure may operate.

[0015] 5A is a flow chart of a method for determining a process operation based on a chamber wall value update, according to some embodiments.

[0016] 5B is a flow diagram of a method for using optical sensor data to initiate performance of corrective actions, according to some embodiments.

[0017] 5C is a flow chart of a method for maintaining uniformity of a processing chamber, according to some embodiments.

[0018] 5D is a flow chart of a method for detecting and correcting anomalies and / or inconsistencies in a fleet of processing chambers, according to some embodiments.

[0019] 5E is a flow chart of a method for utilizing optical sensor data generated over time, according to some embodiments.

[0020] Fig. 6A -D is a flow chart of a process associated with determining a process operation based on a chamber wall value update, according to some embodiments.

[0021] Figure 7is a flow chart of a method of using one or more optical film properties of a deposited thin film on a transparent film to improve processing within a processing chamber, according to some embodiments.

[0022] Figure 8 A model training workflow and a model application workflow for updated prediction and simulation of chamber operation according to some embodiments are illustrated.

[0023] Fig. 9 A block diagram of an example computing device is depicted in accordance with some embodiments. DETAILED DESCRIPTION

[0024] Embodiments described herein relate to systems and methods for process control and process learning using an optical wall process sensor (OWPS). For example, data from an OWPS can be used in embodiments to update process parameters of one or more processes. These systems and methods can use one or more sensors (e.g., OWPS) that can measure one or more properties of a coating on a chamber wall, and can optionally use machine learning techniques to analyze data from such sensors. In one embodiment, these systems and methods can use interference reflection measurements with reference to a liner (or inner wall) of a processing chamber to determine the condition and / or process state of the processing chamber. For example, reflection measurements can be used to determine one or more optical film properties of a thin film layer (e.g., a process film layer) deposited on a liner surface or on a reflector surface that is close to and substantially represents the liner. The optical properties of the film can be determined before, during, and / or after processes such as deposition processes, etching processes, and cleaning processes. The optical properties of the film (e.g., optical film properties) can include, but are not limited to, thickness, refractive index (n) and extinction coefficient (k) values, chemical properties of the film, and the like.

[0025] In some systems, the characteristics (e.g., one or more property values) of a film deposited on a surface within a processing chamber are monitored. In some embodiments, the deposition or etching of a film on a surface or processing chamber can be used to infer the etching or deposition rate of a material on a substrate. In some embodiments, the performance of a processing operation can be inferred from the properties of the film. In some embodiments, the processing operation can be a cleaning operation, such as a cleaning operation performed between processing of some substrates.

[0026] In conventional systems, a test piece can be used to monitor a thin film deposited on a surface within a processing chamber. A removable object (e.g., a test piece or a thin sheet) can be placed on a surface (e.g., an internal sidewall, a liner, etc.) within the processing chamber. After a certain amount of time, the test piece can be removed from the chamber and the film on the test piece surface can be analyzed. This approach can be cumbersome in many ways. Placing the test piece into the processing chamber and removing the test piece from the chamber may include exposing the interior of the processing chamber to ambient atmospheric conditions. In some embodiments, returning to processing conditions (e.g., a clean interior, a vacuum interior, etc.) may be expensive in terms of time consumption, energy consumption, wear and tear of components, etc. In some embodiments, returning to an operable state may include performing a cleaning operation, a chamber conditioning operation, a chamber coating operation, and / or a chamber aging operation after the interior of the processing chamber is exposed to ambient conditions. Using a test piece-based approach may limit the frequency of analysis of film development. In addition, the test piece can be measured between processing operations, but not during the operation of the processing chamber (e.g., during a deposition process, an etching process, or a cleaning process).

[0027] Embodiments of the present disclosure may address at least some of these deficiencies of conventional solutions. In some embodiments, measurements may be made by an optical sensor configured to detect signals from a thin film on an inner surface of a processing chamber. In some embodiments, determining the state of the processing chamber is performed during processing of a substrate in the processing chamber. In some embodiments, determining the state of the processing chamber is performed before, during, and / or after processing a substrate. In some embodiments, determining the state of the processing chamber is performed before, during, and / or after a cleaning operation or other operation performed without a substrate in the processing chamber.

[0028] In some embodiments, determining the state of the processing chamber may involve adapting the analysis to account for the presence of plasma in the processing chamber, which may be a source of electromagnetic radiation. Plasma includes corrosive gases used to etch substrates. In some embodiments, one or more optical film properties are determined before or after processing, and no compensation for the existing plasma is required during the active process.

[0029] In various embodiments, a condition of a portion of a liner of a processing chamber is generally indicative of a condition of a surface of the processing chamber. For example, a condition may indicate a process shift (or drift) that should be corrected (e.g., via calibration, cleaning, and / or replacement of processing equipment or other structures of the processing chamber). A process shift may affect the performance and yield of processed substrates. By using OWPS data to track such process shifts, the systems and methods described herein may use such process shift (e.g., drift) information to compensate for the drift and / or determine when and / or how to perform maintenance (e.g., cleaning operations).

[0030] In another example, the thickness of the deposited film layer on the inner surface of the processing chamber can be related to the deposited film on the substrate being processed (e.g., measured during processing). The characterization of the thin film deposited on one or more inner surfaces of the processing chamber can be extended to identify and manage deposition rates, deposition amounts, process start and / or end conditions, process frequencies, etc. For example, the surface conditions within the processing chamber can be used to determine the optimal frequency and duration of the cleaning procedure. The frequency and duration of other operations (e.g., chamber conditioning operations, chamber coating operations, chamber aging operations, etc.) can also be determined. Other advantages include reduced damage to the plasma volume when performing reflection measurements (e.g., during substrate processing). The present embodiment can include one or more of the above advantages with minimal modifications to existing liner and processing chamber designs.

[0031] Some embodiments of the disclosed systems and methods include a light source (e.g., a broadband light source), a spectrometer (or other device for recording and measuring incident light as an analytical method), and an optical coupling device such as a collimator or a mirror. These embodiments and the processing chamber as part of the disclosed system may further include a transparent material (e.g., a crystal), at least a portion of which is embedded in the wall and liner of the processing chamber. The transparent material may be made of a transparent ceramic material, or may be made of a transparent material such as sapphire, diamond, quartz, or silicon carbide. A transparent film (or substrate) may be formed on the surface of the transparent material exposed to the interior of the processing chamber. The transparent material may serve as a window for the processing chamber. The transparent film may be deposited via deposition using atomic layer deposition (ALD), chemical vapor deposition (CVD), plasma vapor deposition (PVD), or another method of depositing a thin film on a surface. The transparent film may include yttrium oxide, aluminum oxide, zirconium oxide, a combination of the above, or another transparent material (e.g., a transparent crystalline material). In one embodiment, the transparent film is substantially flush with the surface of the liner. The transparent film may also have a chemical property that substantially matches the liner.

[0032] In various embodiments, during processing in the chamber, the light coupling device can direct light from the light source through the transparent material, through the transparent film, and toward the process film layer deposited on the transparent film. In this manner, light passing through both the transparent crystal and the transparent film will be reflected back from their respective surfaces and combined with light reflected from the surface of the process film layer deposited on the transparent film. This reflected light can include a first spectrum consistent with the conditions of the process film layer deposited on the liner.

[0033] In some embodiments, the optical coupling device directs reflected light (e.g., a combination of light reflected from various surfaces) toward a light detection device. In some embodiments, the optical coupling device focuses this combination of reflected light into a fiber optic cable coupled to a spectrometer. The spectrometer can detect a first spectrum of the focused light, which represents the process film layer and can be used to determine optical film properties. The spectrometer can also detect a second instance (e.g., a spectrum) of the focused light when the light source is turned off during processing in the chamber. This second measurement can be used for background subtraction, normalization, etc. In some embodiments, the second light measurement corresponds to an optical emission spectrum (OES) of a plasma in the processing chamber during the measurement and can be removed from the first measurement to produce a reflection measurement signal that isolates the effects of the process film layer deposited on the surface of the liner (and, for example, optionally, a transparent material and / or transparent film beneath the process film layer).

[0034] Embodiments of the systems and methods may also include a processing device (or controller) coupled to the spectrometer. The processing device may be adapted to receive a first measurement (e.g., a spectrum) and a second measurement and calculate reflectance measurement data by subtracting the second data from the first data. The processing device may calculate a reflectance measurement signal by dividing the reflectance measurement data by a reference spectrum, which normalizes the reflectance measurement data. The reference spectrum may be obtained under known conditions, such as during initial installation of the system, shortly after a maintenance or aging event, and the like.

[0035] Subsequently, the processing equipment may fit the reflection measurement signal to a thin film optical model to determine information including one or more properties (e.g., optical film properties) of the process film layer. Such properties may include, but are not limited to, thickness (e.g., optical thickness), values ​​of refractive index (n) and extinction coefficient (k), reflectivity and / or transmittance (any of which may be wavelength dependent), and / or constituent materials. The one or more properties may in turn inform the determination of certain endpoints associated with the process performed within the processing chamber, including, but not limited to, whether to adjust the deposition rate, when to stop chemical or plasma deposition, when to start cleaning the processing chamber, when to stop cleaning the processing chamber, and the amount of process drift or consumed chamber life. In some embodiments, the one or more film properties may indicate anomalies in the processing equipment, such as abnormal or damaged input materials, hardware failures, aging or drift of processing equipment, etc.

[0036] In one embodiment, for example, the processing equipment (or controller) compares the thickness of the process film layer to a baseline measurement (e.g., taken when the processing chamber is first put into operation). If the change in this value exceeds a threshold change (e.g., a threshold change in the thickness of the process film layer), the processing equipment can trigger a process in the processing chamber to correct the rate of deposition or etching of the process film layer (e.g., a cleaning process, a calibration process, etc.). The processing equipment can also change the process to restore the process state, and / or can alert the user of the processing chamber of a process change, as well as other actions to be discussed.

[0037] In some embodiments, the processing device (or controller) determines the characteristics of the process film layer on the transparent film and / or the chamber wall / liner and the corresponding dimensionless value. The characteristics of the film may include thickness, optical properties, chemical properties, etc. The dimensionless value may indicate the trend of the characteristics of the process film layer. For example, a higher dimensionless value may correspond to a larger thickness. The dimensionless value may correspond to a ratio and / or percentage, such as a ratio of a maximum acceptable value. The dimensionless value may be associated with a critical wall thickness (e.g., a critical thickness of the process film layer). The dimensionless value may associate the process film layer thickness with a maximum and / or minimum allowable thickness (e.g., by ratio or percentage). The dimensionless value may be an index corresponding to the process film layer characteristic. The processing device may determine an update to the process operation based on the dimensionless value. The processing device may calibrate one or more sensors of the processing chamber based on the dimensionless value. In addition, the processing device may receive sensor data corresponding to a second processing chamber (e.g., another processing chamber) and determine a second dimensionless value corresponding to the characteristics of the process film layer in the second processing chamber. The first dimensionless value and the second dimensionless value may be normalized to each other by a processing device. The processing device may normalize the first dimensionless value and the second dimensionless value by a data normalization process. The processing device may further calibrate sensors of two processing chambers based on the first value and the second value. The processing device may further update a recipe operation (e.g., a cleaning operation, a deposition operation, an etching operation, etc.) based on the first value and the second value.

[0038] Embodiments of the present disclosure address the deficiencies of current technology by providing a technique for updating process parameters based on optical chamber wall values. The updating of process parameters can improve the optimization of process operations associated with the processing chamber of the manufacturing system (e.g., substrate yield). Embodiments of the present disclosure provide techniques for updating process operations at a better speed than conventional processing systems. The provided techniques can make the operation of the processing chamber more efficient and increase throughput. In addition, embodiments of the present disclosure provide closer monitoring of the conditions of the processing chamber, and the method is to determine a dimensionless value based on sensor data indicating the thickness of the material on the inner surface of the processing chamber. Closer monitoring of the processing chamber conditions can reduce the downtime of the processing chamber by scheduling maintenance, cleaning, aging, adjustment, etc. when the chamber quality changes or the chamber performance deteriorates. Compared with conventional systems, the reduction in downtime can improve the overall throughput and efficiency of the processing chamber.

[0039] In addition, embodiments of the present disclosure provide improvements in sensor accuracy by calibrating one or more sensors of a processing chamber based on dimensionless values ​​derived from measurements of an OWPS sensor. Calibrating a sensor based on a dimensionless value can take into account sensor drift. In some embodiments, sensor drift may occur due to material accumulation on the inner surface of the processing chamber. The improved sensor accuracy can in turn improve the accuracy of implementing process operations and / or procedures. In general, the present disclosure allows greater control over process operations and parameters, which can improve process accuracy, consistency, and throughput. Therefore, the system controller can consume fewer computing resources (e.g., processing cycles, memory space, etc.) to perform substrate processes at the manufacturing system, which improves overall efficiency and reduces overall latency of the computing system associated with the manufacturing system.

[0040] The method of the present disclosure includes: receiving, by a processing device, first data from an optical sensor of a processing chamber. The method further includes: processing the first data to obtain second data. The second data includes an indication of a condition of a coating on an inner surface of the processing chamber. The method further includes: generating an indication of the performance of a processing operation of the processing chamber based on the second data. The method further includes: causing the execution of a corrective action based on the indication of the performance of the processing chamber. The corrective action may include adjusting a processing recipe, such as adjusting a processing recipe endpoint, scheduling maintenance, alerting a user, etc.

[0041] Another aspect of the present disclosure includes a non-transferable machine-readable storage medium. The storage medium stores instructions. The instructions, when executed by a processing device, cause the processing device to perform an operation. The operation includes: receiving, by the processing device, first data from an optical sensor of a processing chamber, wherein the first data is associated with a first operation of the processing chamber. The operation further includes: receiving second data from the optical sensor of the processing chamber, wherein the second data is associated with a second operation of the processing chamber. The operation stored in the storage medium further includes: generating a first indication and a second indication of performance of the processing chamber based on the first data and the second data. The operation stored in the storage medium further includes: determining one or more differences in performance between the first operation and the second operation. The operation stored in the storage medium further includes: causing the execution of a corrective action based on the one or more differences. The corrective action may include an adjustment to a process recipe, such as an adjustment to a process recipe endpoint. The corrective action may include scheduling maintenance, such as corrective or preventive maintenance. The corrective action may include providing an alert to a user.

[0042] Another aspect of the present disclosure includes a method, the method comprising: receiving, by a processing device, first data from an optical sensor of a first processing chamber. The method further comprises: receiving second data from an optical sensor of a second processing chamber. The method further comprises: generating a first indication of performance of the first processing chamber based on the first data, and generating a second indication of performance of the second processing chamber based on the second data. The method further comprises: determining one or more differences in performance between the first processing chamber and the second processing chamber. The method further comprises: causing execution of a corrective action based on the one or more differences. The corrective action may include updating a process recipe, such as updating a process recipe endpoint, scheduling maintenance, providing an alert to a user, and the like.

[0043] Figure 1 1 is a top view of an example processing system 100 according to one aspect. The processing system 100 includes a transfer chamber robot 101 and a factory interface robot 121, each adapted to transfer a substrate 110 (sometimes referred to as a "wafer" or "semiconductor wafer") from an electronic device processing system (such as Figure 1The processing system 100 shown in the figure) is picked up or placed to a destination in the electronic device processing system. However, any type of electronic device substrate, mask or other silicon oxide-containing substrate (generally referred to as "substrate" herein) can also be transported and transferred by the disclosed robot. For example, the destination of the substrate 110 can be one or more processing chambers 103, and / or one or more of the load lock devices 107A, 107B, which can be distributed around the transfer chamber 114 and coupled to the transfer chamber 114. As shown, for example, the transfer of the substrate can pass through a slit valve 111.

[0044] The processing system 100 may further include a host 102, which includes a transfer chamber 114 and at least two processing chambers 103. The housing of the host 102 includes the transfer chamber 114 therein. The transfer chamber 114 may include a top wall (not shown), a bottom wall (floor) 139, and side walls, and in some embodiments, for example, may be maintained in a vacuum. In the depicted embodiment, the transfer chamber robot 101 is mounted on the bottom wall (floor) 139. However, the transfer chamber robot 101 may be mounted elsewhere, such as on the top wall.

[0045] In various embodiments, the processing chamber 103 can be adapted to implement any number of processes on the substrate 110. These processes may include deposition, oxidation, nitridation, etching, polishing, cleaning, lithography, metrology (e.g., integrated metrology), or similar processes. Other processes may also be implemented. For example, the load locks 107A, 107B may be adapted to interface with a factory interface 117 or other system components, which may receive the substrate 110 from a substrate carrier 119 (e.g., a front opening wafer transfer box (FOUP)), which may be docked at the load port of the factory interface 117. A factory interface robot 121 (shown in dotted lines) may be used to transfer the substrate 110 between the substrate carrier 119 and each load lock 107A, 107B. The transfer of the substrate 110 may be implemented in any order or direction. In some embodiments, the factory interface robot 121 may be the same (or similar) as the transfer chamber robot 101, but may further include a mechanism for allowing the factory interface robot to move in any lateral direction indicated by arrow 123. Any other suitable robot may be used as the factory interface robot 121. In some embodiments, the system 100 may be coupled (eg, interfaced) with a metrology system (eg, an integrated metrology system).

[0046] In an embodiment, and by way of illustration of any robot, the transfer chamber robot 101 includes at least one arm 113 (e.g., a robotic arm) and at least one end effector 115 coupled to the arm 113. The end effector 115 can be controlled by the transfer chamber robot 101 to pick up a substrate 110 from a load lock 107A or 107B, guide the substrate 110 through one of the slit valves 111 of the process chamber 103, and accurately place the substrate 110 on a substrate support of the process chamber 103.

[0047] In various embodiments, one or more of the processing chambers 103 may include a transparent window 120 (e.g., a transparent material such as a transparent crystal), at least a portion of which is embedded in the wall and liner 124 (e.g., an inner wall) of the processing chamber 103. In the disclosed embodiments, light can be collimated and directed to the transparent window 120 to produce reflected light. The reflected light can then be transmitted back through the transparent window 120, as will be described in more detail with reference to Figure 2-3 . The reflected light may then be collected for further analysis. In some embodiments, the reflected light may be focused into a fiber optic cable coupled to a spectrometer 125 for analysis, such as spectral analysis. Whether during or after substrate processing, the spectrometer 125 may perform reflection measurements to determine one or more spectra of the focused light, which may be used to determine at least one optical property of a process film layer deposited on the transparent window 120 (e.g., on a transparent film of the transparent window). The transparent window 120 and the spectrometer 125 may include an optical sensor associated with the processing chamber 103. The transparent window 120 and the spectrometer 125 may include an optical wall process sensor (OWPS). The optical sensor may detect at least one property of a material on an inner surface of the processing chamber 103 (e.g., on a transparent window). In some embodiments, the optical sensor may provide sensor data indicating the thickness of the material on the inner surface of the processing chamber 103 through thin film optical calculations performed by a processing device (e.g., processing device 130).

[0048] The controller 109 (e.g., a tool and equipment controller) can control various aspects of the processing system 100, such as the gas pressure in the processing chamber 103, the individual gas flow rates, the spatial flow rate ratios, the temperature of various chamber components, and the radio frequency (RF) or electrical state of the processing chamber 103. The controller 109 can receive signals from and send commands to the factory interface robot 121, the transfer chamber robot 101, one or more sensors, and / or other processing components of the processing system 100. Thus, the controller 109 can control the start and stop of processing, can adjust the deposition rate, type or mix of deposition compositions, and similar functions. The controller 109 can further receive and process sensory data from various sensors.

[0049] In various embodiments, the controller 109 includes (or is coupled to) a processing device 130 and is coupled to the spectrometer 125. The processing device 130 can be configured to receive and process sensor data, including reflectance measurements performed by the spectrometer 125. The processing device 130 can be configured to receive a variety of measurements and perform signal processing, for example, configured to isolate the effects of the thin film (e.g., from the optical effects generated by the plasma) by receiving data collected with and without an external light source. The processing device 130 can calculate a reflectance measurement signal by subtracting a second measurement from a first measurement (e.g., subtracting a second spectrum from a first spectrum). Subsequently, the processing device can fit the reflectance measurement signal to a thin film optical model to determine information including one or more optical film properties of the process film layer. Based on the thin film optical model, the processing device 130 can determine one or more characteristics of the process product deposited on the wall of the process chamber 103. The characteristics can be calculated by the processing device 130, including thickness (e.g., optical thickness), reflectivity, chemical characteristics, etc. Depending on the results of analyzing the one or more optical film properties, the processing device 130 (e.g., the controller 109) can perform one or more actions. For example, the processing device 130 can direct the processing chamber 103 to make process changes or adjustments. For example, the controller 109 can adjust the processing parameters or settings, such as the deposition rate, the type or mixture of deposition components, the timing of performing a cleaning process in the processing chamber, and other parameters that will be referenced. Figure 7 Further, in some embodiments, the controller 109 may cause calibration of one or more sensors of the processing chamber 103 based on the results of analyzing the one or more optical film properties, may perform chamber matching based on the results of analyzing the one or more optical film properties, may identify chamber drift based on the results of analyzing the one or more optical film properties, and the like.

[0050] In some embodiments, the processing device 130 may determine a dimensionless value based on the result of analyzing the one or more optical film properties. In some embodiments, the dimensionless value corresponds to a first amount of a process product deposited on a transparent film (e.g., the thickness of a process film layer). The dimensionless value may indicate the amount of a process product deposited on an inner surface of the processing chamber 103. The processing device 130 may update a recipe (e.g., a cleaning recipe, an etching recipe, a deposition recipe, etc.) based on the dimensionless value. In addition, the dimensionless value may be normalized by the processing device 130 with a dimensionless value corresponding to one or more other processing chambers 103 to further update the recipe or calibrate one or more sensors. Based on the normalized dimensionless value, the recipes of other processing chambers 103 may be updated and the sensors may be calibrated.

[0051] The controller 109 and / or the processing device 130 may be and / or may include a computing device such as a personal computer, a server computer, a programmable logic controller (PLC), a microcontroller, etc. The controller 109 and / or the processing device 130 may include (or be) one or more processing devices, which may be a general-purpose processing device such as a microprocessor, a central processing unit, or the like. In more detail, 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 that implements other instruction sets or a processor that implements a combination 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, or the like. The controller 109 and / or the processing device 130 may include a data storage device (e.g., one or more disk drives and / or solid-state hard disks), a main memory, a static memory, a network interface, and / or other components. The processing device 130 may execute instructions to perform any one or more of the methodology and / or embodiments described herein. The instructions may be stored on a computer readable storage medium which may include main memory, static memory, secondary storage and / or a processing device (during execution of the instructions).

[0052] Figure 2 A simplified side view of a system 200 for monitoring a processing chamber is illustrated according to one aspect. The system 200 may include an optical sensor that generates optical data indicating one or more characteristics of a material on an interior surface of the processing chamber. In some embodiments, the optical sensor may collect spectral data. In some embodiments, the optical sensor may be configured to generate data indicating an amount of material deposited on a transparent window (e.g., a transparent material or a transparent crystal) proximate to the optical sensor. In some embodiments, the optical sensor may be configured to generate data indicating a film thickness deposited on an interior surface of the processing chamber, the data being related, for example, to a measurement indicating an amount of material.

[0053] System 200 may include, for example, a processing chamber 103 including a liner 124, such as Figure 1As shown. The processing chamber 103 may include a wall 222 to which a liner 124 is attached. The liner 124 is specifically designed to have high chemical resistance to the process chemistry to be performed in the chamber 103. The chemical resistance of the liner 124 can minimize physical or chemical changes to the liner 124. The chemical resistance of the liner 124 can increase the life of the liner 124 (e.g., time before maintenance, time before replacement). In addition, as shown, at least a portion of the transparent window 120 can be embedded in the wall 222 and the liner 124. The window 120 can be made of a transparent ceramic material, or it can be made of a durable transparent material, such as sapphire, diamond, quartz, silicon carbide, or a combination of the above.

[0054] In an embodiment, the system 200 further includes a light source 201 (e.g., a broadband light source or other electromagnetic radiation source), an optical coupling device 204 (e.g., a collimator or a mirror), a spectrometer 225, a controller 109, and a processing device 130. The light source 201 and the spectrometer 225 may be optically coupled to the optical coupling device 204 via one or more couplers 232, which may include fiber optic cables.

[0055] In some embodiments, the light source 201 is a flash lamp. The light source 201 can produce predictable illumination, for example, predictable in terms of power, color, spectrum, etc. In some embodiments, the wavelength produced by the light source 201 can be limited, for example, the wavelength can be limited to a wavelength range of 200-800 nanometers (nm). In some embodiments, the light source 201 can include a pulsed light source, such as a pulsed flash lamp. In some embodiments, the light source 201 is a pulsed xenon source. In some embodiments, the power of the light source 201 can be predictable, for example, the light source 201 can be a flash lamp that exhibits a sigma flash power variation of less than 0.5% over the full wavelength range in an open loop. In other embodiments, the wavelength range and power variation can be different from those discussed above and still be within the scope of the present disclosure. In some embodiments, the wavelength emitted can include near infrared wavelengths. For some practical light sources, the variable power control of the light source 201 can be between 45% and 100%, while providing a power drop of less than 8% over a lifetime of at least one year, which can provide more than 1 billion flashes. Additional or different embodiments are contemplated for light sources 201 with different capabilities and wavelengths.

[0056] In various embodiments, the light coupling device 204 can be adapted to collimate or otherwise transmit light in two directions along the light path. The first direction can include light from the light source 201, which is to be collimated and transmitted through the transparent window 120 into the processing chamber 103. The second direction can be reflected light from the transparent window 120, which is transmitted back to the light coupling device 204, which will be referred to as Figure 3Discussed in more detail. The reflected light can be focused into a coupler 232 (e.g., a fiber optic cable) to be directed along the optical path in a second direction toward the spectrometer 225. Further, the coupler 232 can be coupled between the spectrometer 225 and the light source 201 to efficiently transmit the light from the light source 201 to the transparent window 120 and back to the spectrometer 225.

[0057] In one embodiment, spectrometer 225 also has a detection range similar to the wavelength range of the light source. In some embodiments, spectrometer 225 can have a detection range (e.g., the wavelength range over which the spectrometer operates, such as distributed in visible light, ultraviolet light, infrared light, etc.), a wavelength resolution (e.g., a description of how proficient spectrometer 225 is at separating light of different wavelengths), a sampling rate, etc. The properties of spectrometer 225 can be adjusted to analyze chamber types, formulation types, coating types, etc. Spectrometer 225 can be calibrated according to the same schedule as light source 201, or can be calibrated and / or maintained more or less frequently. Spectrometer 225 can be adapted to detect the spectrum of reflected light received from optical coupling device 204 (e.g., light that has been reflected from transparent window 120 and focused into coupler 232 by optical coupling device 204).

[0058] In various embodiments, the controller 109 includes or is coupled to the processing device 130, and includes or is coupled to the memory 134 or other computer storage. The controller 109 can be communicatively coupled to the light source 201, the spectrometer 225, and the processing chamber 103. The controller 109 can instruct the light source 201 to flash, and then receive a first measurement (e.g., a spectrum) from the spectrometer 225. The controller 109 can also keep the light source off and receive a second measurement from the spectrometer 225 when the light source 201 is off. The second measurement can represent the OES of the plasma or another light emitting chemical process in the processing chamber. The processing device 130 can subtract the second measurement from the first measurement to determine a reflectance measurement signal at a certain moment in time. Subsequently, the processing device 130 can mathematically fit the reflectance measurement signal to one or more thin film models to determine one or more optical film properties of the process film layer that has been deposited on the transparent film of the transparent crystal 120.

[0059] In some embodiments, the one or more optical film properties include the deposited film thickness (e.g., optical thickness) and refractive index (n) and extinction coefficient (k) values. The refractive index is the ratio of the speed of light in a vacuum to the speed of light in the process film layer. The extinction coefficient is a measure of how much light is absorbed in the process film layer. In an embodiment, the processing device 130 can use the n and k values ​​to determine the composition of the process film layer. The processing device 130 can be further configured to analyze the data of the one or more optical film properties and trigger a new process (e.g., a cleaning or maintenance process) or update the current process (e.g., a deposition process, a cleaning process, an etching process, etc.) within the processing chamber 103 based on the analysis. The processing device 130 can determine a dimensionless value based on the analyzed data, and the trigger or update can be based on this. Such an update can include an alarm and reference Figure 7 Discuss in more detail.

[0060] Figure 3 According to one embodiment, Figure 2 2 is an enlarged portion of a side view of a system of FIG. 304, illustrating a transparent window 120 embedded within a wall 222 and a liner 124 of a processing chamber 303. The transparent window 120 can have a proximal end and a distal end, wherein the proximal end has a proximal surface 120A that receives light from the optical coupling device 204, and the distal end has a distal surface 120B that reflects light back to the optical coupling device 204 via the proximal surface 120A. In some embodiments, the proximal surface 120A can also reflect light. The distal surface 120B of the transparent window 120 can be proximal to the processing chamber 303, and in one embodiment, the distal surface 120B of the transparent window 120 can be substantially flush with the inner surface of the wall 222.

[0061] In various embodiments, the transparent window 120 can be shaped like a plug (e.g., a plug shape) and include a shaft 301 and a flange 305 integrally formed with the shaft 301 or attached to the shaft. For example, the plug can be a sapphire plug, a diamond plug, a quartz plug, or a silicon carbide plug. The shaft 301 and / or the flange 305 can be square, rectangular, cylindrical, or other shapes. To simplify the illustration, the shaft 301 and the flange 305 are both illustrated as cylindrical. As described above, the transparent window 120 can be made of a transparent crystalline material, such as sapphire, diamond, quartz, silicon carbide, or similar materials, to provide a hard material that is resistant to change and can reflect light. The shaft 301 can be embedded in the wall 222 and the liner 124 so that the remote surface is exposed to the interior of the processing chamber 303.

[0062] The flange 305 can be adjacent to the outer surface of the wall 222. In one embodiment, the seal 313 is positioned between the flange and the outer surface of the wall 222. The seal 313 can be, for example, an O-ring seal, a rectangular seal or a gasket seal, a bulb seal, and the like. The material of the seal 313 can be propylene diene monomer, a fluoroelastomer, or the like. In some embodiments, the seal can be placed in different ways, such as along the shaft 301, and within the scope of the present disclosure. The flange 305 can have an inner surface in physical contact with the seal 313, wherein the inner surface includes a surface roughness (Ra) of less than 20 nanometers or a scratch dig value of at least 80 / 50, which is an optical specification for surface roughness. The highly smooth inner surface of the flange 305 can help provide a tight seal between the ambient atmosphere outside the processing system 100, 200 and the highly clean, filtered air of the processing chamber 303, which may be in a vacuum state.

[0063] In various embodiments, the transparent film 307 can be deposited, for example, via atomic layer deposition (ALD) onto the remote surface 120B of the shaft 301 of the transparent window 120. In some cases, the transparent film 307 is deposited on the transparent window 120 before the transparent window 120 is embedded in the wall 222 and the liner 124. The transparent film 307 can be exposed to the interior of the processing chamber 303. In some embodiments, the transparent film 307 is substantially flush with the liner 124 and is made to have a chemical property that substantially matches the liner, as described above. In various embodiments, the transparent film 307 includes yttrium oxide, aluminum oxide, zirconium oxide, or a combination of the above to make a similar transparent alloy. In one embodiment, the transparent film 307 includes yttrium oxide, a thickness between 10 nanometers and 1 micrometer (μm), for example 270 nanometers thick, and is deposited to a known thickness, which helps analyze the reflected light.

[0064] In various embodiments, the deposition process (e.g., ALD) of the transparent film 307 is carefully controlled to ensure that it mimics the material of the liner 124, but in a different physical state, such as smoother and flatter, to facilitate reflection of light from the surface of the material of the liner 124. In this manner, the transparent film 307 reflects light in a manner that provides data that can be used to generate an accurate assessment of chamber conditions. In some embodiments, the thinness and smoothness of the transparent film 307 increases the signal-to-noise ratio (SNR) of the measurement of light passing through the remote surface, thereby facilitating the ability to test thinner process film layers 309 within the processing chamber 303. In embodiments, the transparent film 307 can also passivate and protect the material of the transparent window 120, thereby enabling the expensive transparent window 120 to survive longer in the corrosive processing environment of the processing chamber 303.

[0065] In some embodiments, during processing, a process film layer 309 (e.g., a process film layer) is deposited within the processing chamber 303. The process film layer 309 may include a single film or a stack of films to be measured by the spectrometer 225. The optical coupling device 204 may collimate the light provided by the light source 201 and direct the collimated light 304A through the proximal surface 120A of the transparent window 120 (e.g., the outer surface of the flange 305). Subsequently, this incident light is reflected from the distal surface 120B of the transparent window (e.g., the shaft 301), from the surface of the transparent film 307, and from the surface of the process film layer 309. These three reflected light sources come back together (shown as reflected light 304B) and return to the optical coupling device 204. The optical coupling device 204 may focus the reflected light into the coupler 232 for entry into the spectrometer for measurement (see Figure 2 ). In some embodiments, additional reflections may be collected and / or considered in the analysis, e.g., reflections from proximal surface 120A, reflections from the interface of flange 305 and shaft 301, etc. In some embodiments, one or more reflections may be mitigated or eliminated. For example, reflections from surface 120A may be mitigated by applying an anti-reflective coating, the design and / or construction of transparent window 120 may eliminate an optical interface between flange 305 and shaft 301, etc.

[0066] The return light signals at each interface (e.g., the above-mentioned surfaces that can reflect light) can be optically combined, resulting in constructive and destructive interference. The exact nature of this interference depends on the relative thickness of each layer (e.g., transparent window 120, transparent film 307, and process film layer 309 (or film stack)). The final result measured by the spectrometer 225 (and after normalization to the initial reference value) may include a spectrum with interference fringes. These fringes can be fitted with a mathematical optical thin film model, which can then determine one or more thin film optical parameters of the process film layer 309, such as thickness (e.g., optical thickness), n and k values. The n and k values ​​can be used to determine the composition of the process film layer 309 (e.g., the composition of the material deposited on the wall of the processing chamber). Knowing the exact initial thickness and optical properties of the transparent film 307 helps to fit the spectrum to the thin film model. The thickness of the process film layer 309 can be incorporated into the calculation of the dimensionless value, and in view of this, the process operation (e.g., cleaning operation, deposition operation, etching operation, etc.) can be updated.

[0067] FIG4 is a block diagram of an exemplary system architecture in which embodiments of the present disclosure may operate. As shown in FIG4, system architecture 400 includes manufacturing system 402, metering system 410, client device 450, data storage 440, server 420, and modeling system 470. Modeling system 470 may be part of server 420. In some embodiments, one or more components of modeling system 470 may be fully or partially integrated into client device 450. Manufacturing system 402, metering system 410, client device 450, data storage 440, server 420, and modeling system 470 may each be hosted by one or more computing devices, including server computers, desktop computers, laptop computers, tablet computers, notebook computers, personal digital assistants (PDAs), mobile communication devices, cell phones, handheld computers, or similar computing devices.

[0068] The manufacturing system 402, the metering system 410, the client device 450, the data storage 440, the server 420, and the modeling system 470 can be coupled to each other via a network 460 (e.g., for performing the methodology described herein). In some embodiments, the network 460 is a private network that enables each element of the system architecture 400 to access each other and other privately available computing devices. The network 460 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, and / or combinations thereof. In some embodiments, processing and / or data storage operations may be performed by cloud-based remote devices, virtual devices, etc., and the network 460 may enable access and communication to these devices. Alternatively or additionally, any element of the system architecture 400 may be integrated together or otherwise coupled without using the network 460.

[0069] The client device 450 may be or include any personal computer (PC), laptop, mobile phone, tablet, network computer, network-connected television (“Smart TV”), network-connected media player (e.g., Blu-ray player), set-top box, over-the-top (OOT) streaming device, operator box, etc. The client device 450 may include a browser 452, an application 454, and / or other tools described and executed by other systems of the system architecture 400. In some embodiments, as described herein, during various stages of processing in the system architecture 400, the client device 450 is able to access the manufacturing system 402, the metrology system 410, the data storage 440, the server 420 and / or the modeling system 470 and communicate (e.g., transmit and / or receive) indications of sensor data, processed data, data classifications, process result data, process update data and / or inputs and outputs of various processing tools (e.g., the metrology tool 414, the data preparation tool 416, the chamber matching tool 422, the sensor matching tool 424, the recipe development tool 426, the process control tool 428 and / or the chamber value component 494).

[0070] As shown in FIG4 , the manufacturing system 402 includes a processing tool 404, a process program 406, and a process controller 408. The process controller 408 can coordinate the operation of the processing tool 404 to perform on one or more process programs 406. For example, the various processing tools 404 can include specialized chambers, such as etching chambers, deposition chambers (including chambers for atomic layer deposition, chemical vapor deposition, or plasma enhanced versions of the above), annealing chambers, and / or the like. In another example, the machine can incorporate a sample transport system (e.g., a Selective Compliance Assembly Robot Arm (SCARA) robot, a transfer chamber, a front opening cassette (FOUP), a side storage cassette (SSP), and / or the like) to transport samples between the machine and process steps.

[0071] The process program 406, or sometimes referred to as a process recipe, process operation, or process step, can include various specifications for implementing operations by the processing tool 404. For example, the process program 406 can include process specifications such as the start-up duration of the process operation, the processing tool used for the operation, the temperature, flow rate, pressure, etc. of the machine (e.g., chamber), the deposition sequence, and similar process specifications. In another example, the process program can include transport instructions for transporting a sample to another process step, or for measurement by the metrology system 410. One or more process programs 406 include preventive maintenance procedures, cleaning procedures, conditioning procedures, aging procedures, etc.

[0072] The process controller 408 may include equipment designed to manage and / or coordinate the actions of the processing tool 404. In some embodiments, the process controller 408 is associated with a process recipe or a series of process program instructions that, when applied in a designed manner, produce a desired process result for a substrate process. For example, a process recipe may be associated with processing a substrate to produce a target process result (e.g., critical dimension, thickness, uniformity standard, etc.). The process controller 408 may coordinate the execution of preventive maintenance procedures (e.g., chamber cleaning), chamber recovery procedures (e.g., chamber aging), etc. to return the chamber to a production state, and / or updates to process operations associated with the processing chamber. In addition, the process controller 408 may also cause updates to the calibration of sensors associated with the processing tool 404 or the metrology system 410.

[0073] As shown in FIG4 , the metrology system 410 includes a metrology tool 414 and a data preparation tool 416. The metrology tool 414 may include various sensors to measure process results (e.g., critical dimensions, thickness, uniformity, etc.) and detect parameters within the manufacturing system 402. In some embodiments, the metrology tool 414 includes one or more optical sensors, such as Figure 2-3 In some embodiments, the metrology tool 414 includes a sensor for measuring the thickness of the material on the inner surface of the processing chamber. The metrology tool 414 may include a tool for performing online metrology, and may include an OWPS. The metrology tool 414 may include a plurality of sensors to detect the environmental conditions of the processing chamber. For example, sensors disposed in one or more processing chambers may be used to measure conditions in the processing chamber, such as temperature, pressure, vacuum conditions, deposition layer thickness, and similar conditions. In another example, sensors disposed in the processing chamber may be used to measure the thickness of process products deposited along the surface of the processing chamber wall. The metrology tool 414 may also include a device for measuring the process results of a substrate processed using the manufacturing system. For example, process results such as critical dimensions, thickness measurements (e.g., film layers from etching, deposition, etc.) of a substrate processed according to a processing recipe and / or actions performed by the process controller 408 may be evaluated. In another example, various sensors may measure the conditions of the chamber in a process operation.

[0074] The data preparation tool 416 may include a process methodology for extracting features and / or generating synthetic / engineered data associated with the data measured by the metrology tool 414. In some embodiments, the data preparation tool 416 may identify correlations, patterns, and / or anomalies in metrology or process performance data. For example, the data preparation tool 416 may perform feature extraction, wherein the data preparation tool 416 uses a combination of measured data to determine whether a criterion is met. For example, the data preparation tool 416 may analyze multiple data points of associated parameters (e.g., temperature, pressure, deposition layer thickness, vacuum conditions, plasma conditions, process product thickness, etc.) to determine whether rapid changes occur during iterations of process operations. In some embodiments, the data preparation tool 416 performs normalization between various sensor data associated with various processing chamber conditions. Normalization may include processing incoming sensor data so that the incoming sensor data appears similar between the various chambers and sensors used to acquire the data. The data preparation tool 416 may generate a dimensionless value based on the data collected by the metrology tool 414. Specifically, the data preparation tool 416 may determine a dimensionless value corresponding to a characteristic (eg, thickness) of a process product film layer deposited on a surface of a processing chamber wall.

[0075] In some embodiments, the data preparation tool 416 can perform one or more of a process control analysis, a univariate limit violation analysis, or a multivariate limit violation analysis on the metrology data (e.g., obtained by the metrology tool 414). For example, the data preparation tool 416 can perform statistical process control (SPC) by employing statistical-based methodologies to monitor and control the process controller 408. For example, SPC can promote efficiency and accuracy of process operations (e.g., by identifying data points that fall within and / or outside of control limits).

[0076] In some embodiments, the state of the processing chamber can be measured throughout the process operation. For example, during the process operation (e.g., during the process recipe operation, during the cleaning operation, etc.), the thickness of the process product deposited on the inner wall of the processing chamber can be measured. In some embodiments, an increased amount of metrology data is intercepted during a predetermined processing operation. For example, during or immediately after the wafer is processed, additional sensors can be activated, and / or the currently activated sensors can intercept additional data. In some embodiments, the process controller 408 can be triggered to measure by the metrology tool 414 based on the operation to be performed by the processing tool 404. For example, the process controller 408 can trigger the activation of one or more sensors (e.g., sensors of the metrology tool 414) in response to an update to one or more parameters associated with the process. In another example, the process controller 408 can trigger the activation of WOPS to measure the optical properties of the process film layer, which can indicate the thickness of the process film layer on the inner surface of the processing chamber wall.

[0077] In some embodiments, the extracted features, generated synthetic / engineering data, and statistical analysis can be used in conjunction with modeling system 470 (e.g., to train, validate, and / or test model 490). Additionally and / or alternatively, data preparation tool 416 can output data to server 420 for use by any of chamber matching tool 422, sensor matching tool 424, recipe development tool 426, and / or process control tool 428.

[0078] The data storage 440 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 storage 440 may store one or more historical data 442, including historical sensor data 444, historical chamber value data 446, and / or historical chamber update data 447 of previous chamber conditions, and process results of substrates processed under associated chamber conditions. In some embodiments, the historical chamber value data 446 includes historical normalized chamber value data. In some embodiments, the historical sensor data 444 includes historical data from an OWPS. The historical sensor data 444 may include optical spectroscopy data indicating the thickness of a process film layer deposited along the inner surface of a processing chamber wall. In some embodiments, the historical chamber update data 447 may include historical updates to one or more process operations associated with the processing chamber. For example, the historical chamber update data 447 may include historical updates (e.g., cleaning recipe updates, etching recipe updates, deposition recipe updates, etc.) made to one or more process operations associated with the processing chamber, which are made based on the historical chamber value data 446. In some embodiments, historical chamber update data 447 may be mapped to historical chamber value data 446. In some embodiments, historical data 442 may be used to train, validate, and / or test a model 490 of a modeling system 470 (see, e.g., Figure 8 illustrative methodology).

[0079] The server 420 may include one or more computing devices, such as a rack server, a router computer, a server computer, a personal computer, a mainframe computer, a laptop computer, a tablet computer, a desktop computer, etc. The server 420 may include a chamber matching tool 422 , a sensor matching tool 424 , a recipe development tool 426 , and a process control tool 428 .

[0080] Each of the chamber match tool 422, the sensor match tool 424, the recipe development tool 426, and the process control tool 428 can receive data from the data preparation tool 416. Each of the chamber match tool 422, the sensor match tool 424, the recipe development tool 426, and the process control tool 428 can receive chamber value data from the data preparation tool 416. The chamber value data can be associated with the conditions of the processing chamber. In some embodiments, the chamber value data indicates the thickness of the process product deposited on the inner surface of the processing chamber. The chamber value data can correspond to the amount or characteristics of the process product deposited along the inner surface of the processing chamber wall. The chamber value data can indicate the net increase or decrease in the film thickness of the process product added to or removed from the inner surface of the processing chamber during a process operation (e.g., a cleaning operation, a deposition operation, an etching operation, etc.).

[0081] In some embodiments, each of the chamber match tool 422, the sensor match tool 424, the recipe development tool 426, and the process control tool 428 receives chamber data from the processing tool 404 and / or receives processed data output from the data preparation tool 416. In some embodiments, the data preparation tool 416 receives raw sensor data from the metrology tool 414. The raw sensor data can be combined with engineered synthetic data from the data preparation tool 416. Each of the chamber match tool 422, the sensor match tool 424, the recipe development tool 426, and the process control tool 428 can generate updates to one or more process recipes 406, and / or can generate reports / notifications indicating chamber matching, chamber drift, anomalies, etc. The updates and / or reports or notifications can be based on the chamber value data.

[0082] In one example, the chamber matching tool 422 can determine an update to a process operation of a first processing chamber based on the chamber value data to cause the process operation performed on the first processing chamber to more closely match the corresponding process operation performed on the second processing chamber. Additional details of the chamber matching tool 422 will be referred to in Figure 6C The process 600C is discussed.

[0083] In another example, the sensor matching tool 424 can determine the calibration of one or more sensors associated with the processing chamber based on the chamber value data. In addition, the sensor matching tool 424 can also determine the sensor calibration that takes into account sensor drift based on the chamber value data. Further details of the sensor matching tool 424 will be referred to in Figure 6B The process 600B is discussed.

[0084] In another example, the recipe development tool 426 can determine updates to a process recipe based on the chamber value data. Specifically, the recipe development tool 426 can determine updates to one or more process parameters (e.g., run time, cleaning time, process knobs, etc.) based on the chamber value data to optimize the process recipe (e.g., process operation). In one embodiment, the recipe development tool 426 determines updates to a cleaning recipe based on the chamber value data, the cleaning recipe being periodically executed to clean the process chamber. In one example, further details of the recipe development tool 426 are described in detail in the accompanying drawings. Fig. 6A The process 600A is discussed.

[0085] As another example, the process control tool 428 can determine an update to an endpoint of a process operation (eg, an endpoint of a chamber cleaning operation) based on the chamber value data. Fig.6D The process 600D is discussed.

[0086] The update to the process program may change the manner in which the process tool 404 implements the process program 406. For example, the update may include increasing or decreasing the process time of one or more process operations of the process program 406. For example, the update may increase or decrease the duration of a chamber cleaning operation. In some embodiments, each of the chamber matching tool 422, the sensor matching tool 424, the recipe development tool 426, and the process control tool 428 may use a model (e.g., a model of the modeling system 470) that receives input data, such as chamber data from the process tool 404, and / or processed data output from the data preparation tool 416, and outputs one or more updates to the process program 406. As will be discussed later, the model may be a machine learning model and may include models such as a bootstrap aggregation model, a random forest tree decision tree model, a partial least squares regression (PLS) model, a least absolute shrinkage and selection operator (LASSO) regression model, and / or a ridge regression model. The model may include ensemble modeling that includes multiple models (e.g., stacked modeling) and uses a higher confidence model to perform a final classification of the received data. In addition, the models may include linear models, dynamic models, numerical models, mechanical models, statistical models, physics-based models, digital twin models, etc.

[0087] As previously described, in some embodiments, each of the chamber matching tool 422, the sensor matching tool 424, the recipe development tool 426, and / or the process control tool 428 can use a machine learning model to perform the methodology described herein. In some embodiments, each of the chamber matching tool 422, the sensor matching tool 424, the recipe development tool 426, and / or the process control tool 428 can use a traditional model and / or a physics-based model (e.g., a digital twin) to perform the methodology described herein. The associated model can be generated (e.g., trained, verified, and / or tested) using the modeling system 470. The following exemplary description of the modeling system 470 will be described in the context of using the modeling system 470 to generate a model 490 associated with the chamber matching tool 422. However, it should be noted that this description is purely exemplary. Similar processing hierarchies and methodologies may be used to generate and execute individual and / or combined models associated with each of the chamber match tool 422, sensor match tool 424, recipe development tool 426, and / or process control tool 428, as will be further discussed in conjunction with other embodiments.

[0088] The modeling system 470 may include one or more computing devices, such as a rack server, a router computer, a server computer, a personal computer, a mainframe computer, a laptop computer, a tablet computer, a desktop computer, etc. The modeling system 470 may include a chamber value component 494. In some embodiments, the chamber value component 494 may use the historical data 442 to determine updates (e.g., updates to one or more process parameters of a process program (e.g., process program 406)) based on the chamber value data. In some embodiments, the chamber value component 494 may use the model 490 to determine updates. In certain embodiments, the model 490 is a trained machine learning model. The model 490 may use the historical data to determine updates.

[0089] In some embodiments, model 490 is stored on server machine 492 and / or server 420. Server machine 480 and / or server machine 492 may store one or more recently trained machine learning models (e.g., to perform model performance management). For example, one or more models trained over a period of time (e.g., the past 6 months) may be stored in server machine 492, and one or more final models used to predict chamber conditions may be stored on server 420 (e.g., which may be operated in conjunction with chamber matching tool 422).

[0090] In some embodiments, modeling system 470 further includes server machine 472 and server machine 480. Server machines 472 and 480 may be one or more computing devices (e.g., rack servers, router computers, server computers, personal computers, mainframe computers, laptop computers, tablet computers, desktop computers, etc.), data storage (e.g., hard disks, memory databases), networks, software components, or hardware components.

[0091] The server machine 472 may include a data set generator 474 that can generate a data set (e.g., a set of data inputs and a set of target output sets) to train, validate, or test a model. The data set generator 474 may divide the historical data 442 into a training set (e.g., sixty percent of the historical data, or any other portion of the historical data), a validation set (e.g., twenty percent of the historical data, or some other portion of the historical data), and a test set (e.g., twenty percent of the historical data). In some embodiments, the chamber value component 494 generates multiple sets of training data. For example, one or more sets of training data may include each of the data sets (e.g., a training set, a validation set, and a test set).

[0092] The server machine 480 includes a training engine 482, a validation engine 484, and a testing engine 486. The training engine 482 can train a model 490 using one or more of the historical sensor data 444, the historical chamber value data 446, and / or the historical chamber update data 447 of the historical data 442 (of the data storage 440). In some embodiments, the model 490 can be trained using one or more outputs of the data preparation tool 416, the chamber matching tool 422, the sensor matching tool 424, the recipe development tool 426, and / or the process control tool 428. For example, the model 490 can be a hybrid machine learning model that uses sensor data and / or mechanistic features, such as feature extraction, mechanistic modeling, and / or statistical modeling (e.g., using the data preparation tool 416). The training engine 482 can generate a plurality of trained models 490, where each trained model 490 corresponds to a different set of features for each training set.

[0093] The validation engine 484 can determine the accuracy of each of the trained models 490 based on a corresponding set of features for each training set. The validation engine 484 can discard trained models 490 whose accuracy does not meet a threshold accuracy. The testing engine 486 can determine the trained model 490 with the highest accuracy among all trained models based on the test set (and optionally the validation set).

[0094] Model 490 may refer to a model created by training engine 482 using a training set that includes data inputs and corresponding target outputs (historical results for a processing chamber under parameters associated with the target inputs). Patterns in the data set that map the data inputs to the target outputs may be found (e.g., identifying a connection between a portion of sensor data and / or chamber value data and a chamber process update), and model 490 is provided that captures the mapping of these patterns. Model 490 may use one or more of logistic regression, syntax analysis, decision trees, or support vector machines (SVMs). In some embodiments, machine learning may consist of a single layer of linear or nonlinear operations (e.g., SVMs) and / or may be or include a neural network.

[0095] The chamber values ​​component 494 can provide current data (e.g., current sensor data indicating material thickness on an interior surface of a processing chamber) as input to the model 490 and can run the model 490 on the input to obtain one or more outputs including a set of values ​​indicating updates to one or more process recipes 406.

[0096] For purposes of illustration and not limitation, aspects of the present disclosure describe training of models and use of models using information related to historical data 442. In other embodiments, heuristic models or rule-based models may be used to determine process operation updates.

[0097] In some embodiments, the functionality of client device 450, server 420, data storage 440, and modeling system 470 may be provided by a fewer number of machines than shown in FIG4. For example, in some embodiments, server machines 472 and 480 may be integrated into a single machine, while in some other embodiments, server machines 472, 480, and 492 may be integrated into a single machine. In some embodiments, modeling system 470 may be provided in whole or in part by server 420.

[0098] In general, the functions described in one embodiment as being performed by client device 450, data storage 440, metrology system 410, manufacturing system 402, and modeling system 470 may also be performed on server 420 as appropriate in other embodiments. In addition, the functionality attributed to a particular component may also be performed by a different component or multiple components operating together.

[0099] In some embodiments, a "user" may be represented as a single individual. However, other embodiments of the present disclosure include a "user" that is an entity controlled by multiple users and / or automated sources. For example, a group of individual users united as an administrator group may be considered a "user."

[0100] 5A-E are flow diagrams of methods 500A-E associated with using an optical sensor for a process chamber. FIG. 5A is a flow diagram of method 500A for determining a process operation update based on a chamber wall value, in accordance with some embodiments. Methods 500A-E may be performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing device, etc.), software (e.g., instructions running on a processing device, a general purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. In some embodiments, methods 500A-E may be performed in part by modeling system 470 of FIG. 4 (e.g., server machine 472, server machine 480, and server machine 492 of FIG. 4). In accordance with embodiments of the present disclosure, modeling system 470 may use method 500A to generate a data set for at least one of training, validating, or testing a machine learning model. Methods 500A-E may be performed by server machine 480 (e.g., training engine 482, etc.). Methods 500A-E may be performed by server machine 492 (e.g., chamber value component 494). In some embodiments, methods 500A-E may be performed by server 420 (e.g., one or more of chamber match tool 422, sensor match tool 424, recipe development tool 426, and / or process control tool 428). In some embodiments, a non-transitory storage medium stores instructions that, when executed by a processing device (e.g., a processing device of modeling system 470, a processing device of server machine 480, a processing device of server machine 492, a processing device of server 420, etc.), cause the processing device to perform methods 500A-E.

[0101] For ease of explanation, methods 500A-E are depicted and described as a series of operations. However, operations according to the present disclosure may occur in various orders and / or in parallel, and with other operations not presented and described herein. In addition, methods 500A-E according to the disclosed subject matter can be implemented without performing all of the operations shown. In addition, those skilled in the art will understand and appreciate that methods 500A-E may alternatively be represented as a series of interrelated states via state diagrams or events.

[0102] Referring to FIG. 5A , at block 502 of method 500A, a processing device may receive first sensor data indicating a state of a wall corresponding to a first processing chamber. The wall may be an inner wall of the first processing chamber. In some embodiments, the wall is a liner of the first processing chamber. In some embodiments, the first sensor data includes optical spectral data indicating a state of the wall of the first processing chamber. The optical spectral data may be from an OWPS. The optical spectral data may include thin film interference data indicating a thickness of a thin film deposited on the wall of the first processing chamber and / or other characteristics of the thin film. In some embodiments, the sensor data indicates a thickness of a process film layer deposited on an inner wall (e.g., a liner) of the processing chamber. The process film layer may be a deposition byproduct of a first process operation (e.g., a byproduct of a deposition operation, an etching operation, etc.). The sensor data may indicate a condition (e.g., a cleanliness condition) inside the processing chamber. For example, sensor data indicating a thickness of a process film layer on a first processing chamber wall may also indicate a thickness of a process film layer deposited on one or more other inner surfaces of the first processing chamber.

[0103] At block 504, the processing device may determine a first value based on the first sensor data. The first value may be a dimensionless value indicating the thickness of the process film layer. In some embodiments, the first value is a first index indicating the thickness of the process film layer deposited on the inner wall of the first processing chamber at the first time. In some embodiments, the first value indicates the thickness of the process film layer. For example, the first value may correspond to the thickness of the process film layer deposited on the wall of the first processing chamber before the first process operation. In some embodiments, the first value may correspond to a first amount of a product (e.g., a process film layer) deposited along the wall surface. In some embodiments, the first value indicates a thickness trend of the process film layer deposited on one or more inner surfaces of the first processing chamber. As an example, a larger first value may indicate a thicker process film layer, while a smaller first value may indicate a thinner process film layer. The first value may further indicate the state of the first processing chamber.

[0104] At block 506, the processing device may determine a first update to a first process operation associated with the first processing chamber based on the first value. In some embodiments, the first process operation is a process procedure (e.g., process procedure 406 of FIG. 4 ). In some embodiments, the first update may be an update to a duration of one or more of a cleaning recipe, a deposition recipe, or an etching recipe, etc. In some embodiments, the first update may be an update to a deposition rate or an etching rate associated with the deposition recipe or the etching recipe, respectively. In some embodiments, the first update may be an update to a process knob of the first process operation. In some embodiments, the first update is performed to more closely match a target process procedure (e.g., process procedure 406 of FIG. 4 ). In some embodiments, the first update is a maintenance procedure (e.g., a preventive maintenance procedure, a reactive maintenance procedure, etc.). In some embodiments, the first update is implemented by a user (e.g., replacing a chamber component).

[0105] In some embodiments, the state of the chamber can be tracked using optical sensor data. In some embodiments, an index indicating the state of the coating (e.g., CWI) can indicate the state of the chamber. In some embodiments, an index can be generated from data including optical sensor data, such as a chamber state index. The evolution of the index over time can indicate drift or aging of components, drift of chamber performance, etc.

[0106] In some embodiments, receiving and storing optical data (and / or data based on the optical data) for a fleet of processing chambers can enable fleet-wide outlier detection, tool matching, consistency assessment, and the like. In some embodiments, the target processing conditions can be the same in a fleet of chambers. The optical data can be used to homogenize the processing conditions in different chambers. In some embodiments, the optical data from a fleet of chambers can be provided to a trained machine learning model configured to detect outliers. Such a model can have common features with a machine learning model configured to detect anomalies, as described above. In some embodiments, detecting an outlier chamber can include comparing the optical data received from the chamber with historical optical data. In some embodiments, an outlier chamber may produce optical data that is consistent with other chambers, but other data may be different (for example, a cleaning program can be run longer to achieve the same optical results). Corrective actions can be initiated, such as scheduling maintenance, providing alerts to users, and the like.

[0107] In some embodiments, multivariate methods may be used, such as for anomaly detection, flagging outliers, and the like. In some embodiments, spectral data may be combined with other data (e.g., other sensor data, such as pressure data, temperature data, and the like, and / or wafer data, such as online or in-situ metrology data, and the like). In some embodiments, spectral data, one or more indices based on spectral data, metrology data, and / or sensor data may be supplied to a machine learning model. The machine learning model may be configured to receive data indicating processing conditions, processing quality, chamber performance, and the like as input, and to generate an indication of whether the chamber's performance is similar to other chambers in the fleet (e.g., the performance is not abnormal) as an output. In some embodiments, data based on spectral data may include operations performed on the spectral data, such as the derivative of an index based on the spectral data, the slope of the spectrum at a particular wavelength, the mean or median, and the like.

[0108] At block 508, the processing device may prepare a notification indicating the first update for presentation on the GUI. In some embodiments, the notification is prepared for review by a user (e.g., an engineer, operator, technician, etc.) via the GUI. At block 508, the processing device may cause a first process operation to be performed in accordance with the first update. For example, the processing device may cause the first process operation to be run in accordance with the first update. As a specific example, the processing device may cause the processing chamber to perform a cleaning operation in accordance with the updated duration.

[0109] Optionally, at block 510, the processing device may determine an update to at least one of a sensor gain value or a sensor offset value corresponding to the OWPS based on the first value. The processing device may determine a calibration value corresponding to the OWPS based on the first value. For example, the processing device may determine that sensor data from the OWPS has drifted over time based on the first value, and may determine an update to the OWPS gain value and / or offset value based on the first value. In some embodiments, the processing device may determine an update to at least one of a sensor gain value or a sensor offset value corresponding to one or more sensors of the first processing chamber based on the first value. The processing device may determine one or more calibration values ​​for the one or more sensors of the first processing chamber based on the first value.

[0110] In some embodiments, a baseline procedure may be performed to identify and / or correct for optical sensor drift. The baseline procedure may be performed periodically, such as at intervals of a fixed amount of time, a fixed number of processed substrates, etc. In some embodiments, the baseline procedure may be performed at a target frequency. In some embodiments, the properties of the optical sensor arrangement may change over time. For example, a transparent window of an OWPS sensor may be corroded, undergo composition changes, etc. Baselining may be performed to account for changes in the behavior of the sensor assembly. Baselining may be performed by generating spectral data in a clean, idle chamber (e.g., when processing operations are not occurring). The baseline spectrum may be used to zero the spectrum collected by the OWPS system during processing of a substrate (e.g., by subtracting the baseline spectrum from the collected spectrum). The frequency of performing the baseline operation may be adjusted based on the intended use of the chamber, the quality or performance of the chamber, the service life of the chamber (e.g., the time since the chamber was installed, the number of hours of operation since the chamber was installed, etc.), or the like. In some embodiments, the baseline measurement may be used to determine a baseline response of the optical sensor (e.g., a response in the absence of processing material). The baseline data may be used to correct the data received from the optical sensor.

[0111] Optionally, at block 512, the processing device may determine a second value based on the first sensor data. The second value may correspond to a second amount of the product disposed along the surface of the first processing chamber wall at a second time. For example, the second value may correspond to a thickness of a process film layer deposited on the first processing chamber wall after the first process operation. In a more specific example, the second value may correspond to a thickness of the process film layer after the cleaning operation. The difference between the first value and the second value may correspond to a net change in the thickness of the process film layer during the first process operation.

[0112] Optionally, at block 514, the processing device may determine a process rate associated with the first process operation. The processing device may determine the process rate using the first value and the second value. The processing device may determine the process rate using the difference between the first value and the second value divided by the duration of the first process operation. The first update determined by the processing device may be further determined based on the process rate. As an example, the cleaning operation may be updated based on a cleaning rate (e.g., a process rate) determined based on a difference between a starting thickness of a process film layer and an ending thickness of the process film layer divided by the duration of the cleaning operation. The first update may correspond to extending or shortening the duration of the cleaning operation based on the cleaning rate.

[0113] 5B is a flow chart of a method 500B for using optical sensor data to initiate execution of a corrective action in accordance with some embodiments. At block 520, processing logic receives first data from an optical sensor of a processing chamber. The optical sensor may be an OWPS sensor. The first data may include spectral data. The first data may indicate a coating on a surface of an OWPS sensor assembly (e.g., a transparent window of an OSPS sensor assembly). In some embodiments, the first data may be generated by the optical sensor during a processing operation. In some embodiments, the first data may be generated between processing operations.

[0114] At block 522, processing logic processes the first data to obtain second data. The second data includes an indication of a condition of a coating on an inner surface of a process in a chamber. In some embodiments, the condition of the coating includes a thickness of the coating. In some embodiments, the second data includes a dimensionless parameter related to the condition of the coating, such as a CWI index. In some embodiments, the first data may be processed to take into account optical effects, such as baseline drift, plasma effects within the chamber, and the like. In some embodiments, a baseline response of an optical sensor assembly (e.g., an OWPS) may be obtained. In some embodiments, baselining may include collecting optical data when the process chamber is in an idle mode. In some embodiments, the effects of an emissive plasma may be taken into account. In some embodiments, optical sensor data may be collected when a light source coupled to the optical sensor is turned off and when plasma is being generated in the process chamber. In some embodiments, the plasma signal and the baseline signal may be used to correct data received from the optical sensor, such as by baseline subtraction.

[0115] In some embodiments, data can be generated when the processing chamber is in a reference state (e.g., conditions different from general processing conditions). Sensor data can be collected when the chamber is in the reference state to determine a reference sensor response. The reference sensor response can be used to correct or update the first data and / or the second data.

[0116] At block 524, the processing logic generates an indication of the performance of the processing operation of the processing chamber based on the second data. In some embodiments, the indication of performance can be based on the value of a dimensionless index (e.g., a CWI index). In some embodiments, the processing operation can be an operation of depositing a coating or removing material from a surface in the processing chamber, such as an etching operation, an atomic deposition operation, etc. In some embodiments, the processing operation can be a chamber cleaning operation. In some embodiments, generating an indication of the performance of the processing operation can include providing data (e.g., spectral data, CWI data, etc.) to a trained machine learning model. The machine learning model can be trained using historical optical sensor data (e.g., as training input). The machine learning model may have been trained using supervised learning (e.g., provided with labeled training data, such as training data indicating quality classification, anomaly classification, historical anomaly label data, etc.). The machine learning model may have been trained using unsupervised learning (e.g., configured to perform clustering, anomaly detection, etc.).

[0117] At block 526, processing logic causes execution of a corrective action based on the indication of the performance of the processing chamber. In some embodiments, the corrective action may include one or more of: providing an alert to a user (e.g., on a graphical user interface), updating a processing operation recipe (e.g., updating one or more steps or operations of an etch recipe, a cleaning recipe, or a deposition recipe), scheduling preventive maintenance, scheduling corrective maintenance, etc. In some embodiments, updating the recipe may include updating an endpoint of a processing operation, e.g., if insufficient time is determined, the operation may be extended, the operation may be shortened to increase chamber throughput, etc.

[0118] 5C is a flow chart of a method 500C for maintaining consistency of a processing chamber according to some embodiments. At block 530, processing logic receives first data from an optical sensor of the processing chamber. The first data is associated with a first operation of the processing chamber. In some embodiments, the first operation is an operation that changes the condition of a coating on an inner surface of the chamber, such as a deposition or etching operation. In some embodiments, the operation is a cleaning operation, such as an in-situ chamber cleaning operation. In some embodiments, processing logic receives a plurality of historical data, such as the first data is one of many optical data associated with many operations. In some embodiments, the first data includes spectral data. In some embodiments, the first data includes an index, such as CWI. In some embodiments, the first data indicates the evolution of spectral data throughout the processing operation, such as a time trace of a CWI value, a time trace spectral value, etc.

[0119] At block 532, processing logic receives second data from an optical sensor of the processing chamber. The second data is associated with a second operation of the processing chamber. In some embodiments, the second data is associated with a processing operation that corresponds to the processing operation of the first data, such as the same operation of the same recipe performed in different runs, on different wafers, etc. The second data may include similar features as the first data, such as spectral data, evolution of data over time, etc.

[0120] At block 534, the processing logic generates a first indication and a second indication of performance of the processing chamber based on the first data and the second data. The first indication of performance can be associated with the performance of the first processing operation. The second indication of performance can be associated with the performance of the second processing operation. In some embodiments, the indication of performance can be generated by a machine learning model, for example, the second data and / or the first data can be provided to a trained machine learning model, and the trained machine learning model generates the indication of performance. In some embodiments, the machine learning model can be configured to detect outliers, abnormal values, etc. based on the optical sensor data.

[0121] At block 536, processing logic determines one or more differences in performance between the first operation and the second operation. In some embodiments, the difference in performance can indicate a drift in performance over time. In some embodiments, multiple historical performance data are considered. In some embodiments, the evolution of performance over time can indicate a trend in performance, such as a performance trend in possible historical instances. In some embodiments, multivariate analysis can be performed. The optical sensor data can be used with other sensor data, manufacturing parameters, hardware data, metrology data, etc. to determine the performance of the processing operation, the performance of the chamber, the performance drift over time, the performance drift with the number of chamber activity hours, etc. At block 538, processing logic causes corrective actions to be performed based on the one or more differences in performance data. Corrective actions can include updating a processing recipe, such as updating an endpoint condition (e.g., endpoint time, target endpoint sensor reading, etc.) of a processing recipe. Corrective actions can include scheduling maintenance, such as scheduling replacement of a component, scheduling preventive maintenance, scheduling further investigation of a subsystem, and the like. Corrective actions can include providing an alert to a user, such as via a user interface.

[0122] 5D is a flow chart of a method 500D for detecting and correcting anomalies and / or inconsistencies in a team of processing chambers, according to some embodiments. At block 540, processing logic receives first data from an optical sensor of a first processing chamber. The first data may include spectral data. The first data may include processed data, such as an index, such as a CWI index. The first data may indicate a characteristic of a coating on an inner surface of the processing chamber, such as a thickness of the coating, a composition of the coating, etc. At block 542, processing logic receives second data from an optical sensor of a second processing chamber. The second data may share features with the first data.

[0123] At block 544, the processing logic generates a first indication of the performance of the first chamber based on the first data, and generates a second indication of the performance of the second chamber based on the second data. The indication of performance may include an index representing the state of the chamber, such as a dimensionless index, such as CWI. The indication of performance may incorporate additional data, such as the generation may include multivariate analysis. In some embodiments, data indicating the optical conditions of the chamber (e.g., OWPS sensor data, data indicating the thickness and / or composition of the deposited film in the processing chamber, etc.) may be provided to a trained machine learning model to generate an indication of performance. In some embodiments, other sensor data, metrology data, processing parameter data, etc. may also be supplied to the machine learning model. The trained machine learning model may be configured to determine differences in performance, such as may be configured to detect outliers, abnormal values, etc. In some embodiments, the performance of a processing chamber may be compared with the performance of multiple chambers, such as abnormal performance in a team of chambers may be identified. The performance of a processing chamber may be compared with historical data of multiple processing chambers.

[0124] At block 546, processing logic determines one or more differences in performance (e.g., quality of performance) between the first process chamber and the second process chamber. In some embodiments, the operations of block 546 may be performed by providing data to a trained machine learning model. In some embodiments, the operations of block 546 may be performed by comparing the data of one process chamber to an average, median, range, etc. of a plurality of other chambers (e.g., a plurality of chambers including the first process chamber).

[0125] At block 548, processing logic causes a corrective action to be performed based on the one or more differences between the performance of the first processing chamber and the second processing chamber. In some embodiments, a corrective action may be selected to change the performance of the second processing chamber. In some embodiments, the corrective action may change the performance of the second processing chamber so that the changed performance of the second processing chamber is more similar to the performance of the first processing chamber than the unchanged performance of the second processing chamber. In some embodiments, the performance of the second chamber may match the first chamber. The corrective action may include updating a process recipe. In some embodiments, a process recipe associated with the optical sensor data may be updated. For example, process time, process temperature or pressure, gas flow rate, etc. may be adjusted to more closely align the performance of the second chamber with the performance of the first chamber. In some embodiments, a process other than a process associated with the optical sensor data may be updated. For example, a cleaning program recipe may be updated, and updating the cleaning program may affect the performance of a process recipe (e.g., a substrate handling program) of the second chamber.

[0126] FIG5E is a flow diagram of a method 500E for utilizing optical sensor data collected over time, according to some embodiments. At block 550, processing logic receives first sensor data. The first sensor data includes optical data, such as data from an OWPS. The first data indicates a state of a wall of a first processing chamber. At block 552, processing logic receives second sensor data. The second sensor data may have similar characteristics to the first sensor data. The operations of blocks 550 and 552 may share characteristics with the operations of block 502 of FIG5A.

[0127] At block 554, processing logic compares the first sensor data with the second sensor data. In some embodiments, the first sensor data may include historical sensor data. In some embodiments, corrective actions may be determined based on the historical data. Historical optical sensor data (e.g., spectral data generated by the OWPS) for many processing runs may be recorded and stored (e.g., as first sensor data). In some embodiments, historical data collected from the same processing chamber, from processing runs executing the same recipe, from processing runs sharing a target final product, etc. may be utilized. Second sensor data (e.g., current dimensionless values ​​derived from the OWPS sensor data) may be compared with historical optical sensor data. In some embodiments, statistical analysis may be utilized for anomaly detection—for example, it may be determined whether the dimensionless value associated with the OWPS sensor data falls outside a threshold value (e.g., three standard deviations from the mean of the historical data). In some embodiments, a machine learning model may be trained using historical optical data (e.g., spectral data, parameters generated based on the spectral data, etc.). The machine learning model may be configured to accept current data (e.g., spectral data indicating the condition of a liner deposited on a chamber wall) as input and generate an indication of whether the processing run associated with the input data is likely to be abnormal (e.g., producing a product that does not meet a performance threshold, etc.) as output. In some embodiments, the machine learning model can be trained via unsupervised learning (e.g., generated without explicit labeling to determine whether input data is similar to historical data). In some embodiments, the machine learning model can be trained using supervised learning (e.g., trained on data labeled with classification labels, such as labeling data as associated with acceptable products, labeling data with specific abnormal reasons to classify abnormal reasons, etc.).

[0128] In some embodiments, data associated with an optical sensor may be monitored over time (e.g., during processing of many substrates, over long chamber run times, over long RF hours, etc.). In some embodiments, the optical data may be utilized to maintain consistency from run to run, e.g., process parameters may be adjusted to compensate for drift, aging, etc. of processing equipment components. Run-to-run consistency operations may share features with the anomaly detection operations described above (e.g., as described in conjunction with block 506 of FIG. 5A ). Run-to-run consistency operations may include statistical calculations, providing current sensor data to a machine learning model, etc. In some embodiments, chamber conditions may be tracked over time, and process operations (e.g., process recipes) may be updated to maintain chamber consistency. In some embodiments, tracking changes in processing conditions, recipes (e.g., changes made to maintain consistent processing conditions), etc. may enable one to understand chamber performance over time, upcoming maintenance, etc. In some embodiments, monitoring data from a chamber over time may include monitoring spectral data and / or monitoring index data, such as a value designed to represent the quality of a coating, chamber, etc. based on spectral data. In some embodiments, evaluating performance over time may include evaluating data generated at the end of a process, such as a wall index value (CWI value) associated with spectral data at the end of a process. In some embodiments, evaluating performance over time may include evaluating data generated throughout a process, such as the evolution of a wall index value throughout a process. In some embodiments, OWPS data may be utilized to determine whether an update to a process recipe has had the desired effect, such as whether an update corrected a process failure. In some embodiments, one or more other values ​​may be tracked over time, such as a derivative or slope of spectral data or an index, higher order derivatives, an average value over time (e.g., an average value over a process operation), and the like.

[0129] In some embodiments, data from several processing operations (e.g., processing runs performed over a period of time) can be analyzed. In some embodiments, data can be tracked over time (e.g., an index such as CWI can be tracked in multiple processing runs). The performance of the process can be related to the optical data being tracked, for example, the etching rate can be related to the CWI value. Drifts of chambers, processes, fleets, etc. can be detected based on the optical data being tracked. Corrective actions can be taken based on drifts detected over time. For example, OWPS data can be generated during a cleaning program. OWPS data collected at the end of a cleaning program can be tracked over time (e.g., in some cleaning program runs). Drifts in OWPS data can be detected (e.g., by calculating the slope of an index associated with OWPS data plotted over time, by calculating the difference between index values ​​over time, etc.). A process can be adjusted due to drift detection, for example, the end point of a cleaning program can be adjusted. In some embodiments, the program can be adjusted to provide improved results, for example, the cleaning program can be extended for more thorough cleaning. In some embodiments, the program can be adjusted to provide improved efficiency, for example, the cleaning program can be shortened to increase throughput, reduce costs, etc.

[0130] Fig. 6A -D is a flow chart of processes 600A-D associated with determining process operations based on chamber wall value updates, according to some embodiments. The methods 600A-D may be performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing devices, etc.), software (e.g., instructions running on a processing device, a general purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. In some embodiments, processes 600A-D may be performed in part by modeling system 470 of FIG. 4 (e.g., server machine 472, server machine 480, and server machine 492 of FIG. 4). Processes 600A-D may be performed by server machine 480 (e.g., training engine 482, etc.). Processes 600A-D may be performed by server machine 492 (e.g., chamber value component 494). In some embodiments, processes 600A-D may be performed by server 420 (e.g., one or more of chamber match tool 422, sensor match tool 424, recipe development tool 426, and / or process control tool 428). In some embodiments, a non-transitory storage medium stores instructions that, when executed by a processing device (e.g., a processing device of modeling system 470, a processing device of server machine 480, a processing device of server machine 492, a processing device of server 420, etc.), cause the processing device to perform processes 600A-D.

[0131] refer to Fig. 6A, a process 600A for updating a chamber cleaning recipe is shown. The process 600A may be performed by the recipe development tool 426 of FIG. 4 . In some embodiments, at block 602, a processing device (e.g., of the recipe development tool 426) may receive OWPS data. The OWPS data may indicate one or more optical film properties of a process film layer deposited on an inner surface of a process chamber wall (e.g., on a process chamber liner). The OWPS data may indicate the thickness of the process film layer deposited on the inner wall (e.g., liner) of the process chamber. The processing device may determine a first characteristic of the film deposited on the wall at block 604. In some embodiments, the processing device may determine the thickness of the film. In some embodiments, the processing device may determine the chemical composition of the film. In some embodiments, the processing device may determine the chemical arrangement or packaging geometry (e.g., crystal structure, molecular or atomic stacking structure, etc.) of the film. In some embodiments, the processing device may determine another property of the film. The processing device may determine a characteristic of the process film layer before a process operation. The processing device may determine a first dimensionless value indicating a characteristic of the process film layer before a process operation. In some embodiments, the first dimensionless value may be determined before a chamber cleaning operation.

[0132] At block 608, the processing device may cause initiation of a processing operation. In some embodiments, the initiated processing operation may include a cleaning process, an etching process, a deposition process, or the like.

[0133] After the processing operation, at block 610, the processing device may determine a second characteristic of the film on the processing chamber wall, such as a characteristic of the film at a second time. The processing device may determine a second dimensionless value indicative of the characteristic of the process film layer after the processing operation. The processing device may determine the second value based on sensor data received from the OWPS after the processing operation.

[0134] At block 612, the processing equipment may determine whether a threshold process layer characteristic has been met. In some embodiments, the threshold process layer characteristic includes a critical wall thickness. In some embodiments, the threshold process layer characteristic includes a target chemical composition. The processing equipment may make a determination based on a comparison of the first dimensionless value and the second dimensionless value. As an example, the processing equipment may determine whether the processing chamber has been cleaned so that the thickness of the process film layer on the processing chamber wall is less than a predetermined value (e.g., a critical wall thickness). If the threshold thickness has not been met, the processing equipment may cause additional chamber cleaning (block 608). If the threshold thickness has been met, the processing equipment may cause the cleaning recipe to be updated at block 614 (e.g., based on any additional cleaning time, etc.). After the cleaning process has been updated, the processing equipment may cause the processing chamber to be released to a production mode. Although the operation of process 600A is described with reference to a chamber cleaning operation, it will be understood by those of ordinary skill in the art that process 600A may also be applied to any chamber process operation, such as a deposition process, an etching process, etc.

[0135] refer to Figure 6B , a process 600B for calibrating a sensor of a processing chamber is shown. Process 600B may be performed by the sensor matching tool 424 of FIG. 4 . Process 600B may be performed to take into account sensor drift that varies over time. In some embodiments, at block 602, a processing device (e.g., of the sensor matching tool 424 of FIG. 4 ) may receive OWPS data. The OWPS data may indicate one or more optical film properties of a process film layer deposited on an inner surface of a processing chamber wall (e.g., on a processing chamber liner). The OWPS data may indicate the thickness of the process film layer, the chemical composition of the process film layer, etc. At block 624, the processing device may determine a characteristic of the film deposited on the processing chamber wall. The characteristic may correspond to the amount of material deposited on the surface, the thickness of the film on the surface, the chemical composition of the film, or another characteristic of the film. In some embodiments, the thickness of the film may be determined based on data indicating the amount of material deposited near the sensor. The processing device may determine a dimensionless value indicating a characteristic of the process film layer at block 626. The processing equipment may use the dimensionless value to calibrate one or more sensors associated with the processing chamber at block 628. The one or more sensors may include the OWPS of the processing chamber, but may also include other sensors of the processing chamber, sensors associated with one or more other processing chambers, etc. Calibration of the sensors may include baseline subtraction, percentage response or another functional adjustment, parameter-dependent adjustment (e.g., wavelength-dependent adjustment), etc. Calibration of the sensors by the processing equipment may be in response to sensor drift and / or process drift caused by accumulation of process film layers on one or more interior surfaces of the processing chamber.

[0136] refer to Figure 6C, a process 600C for updating a chamber process recipe is shown. The process 600C may be performed by the chamber matching tool 422 of FIG. 4 . In some embodiments, at block 630, a processing device (e.g., of the chamber matching tool 422 of FIG. 4 ) may receive first OWPS data corresponding to a first process chamber. At block 632, the processing device may determine a first dimensionless value based on the first OWPS data. The first dimensionless value may indicate a process film layer characteristic of a film deposited on an inner wall (or other inner surface) of the first process chamber. At block 634, the processing device may receive second OWPS data corresponding to a second process chamber. The processing device may determine a second dimensionless value based on the second OWPS data at block 636. The second dimensionless value may indicate a process film layer characteristic of a film deposited on an inner wall of the second process chamber. In some embodiments, the processing device may normalize the first dimensionless value and the second dimensionless value by a data normalization operation. Normalization of the first dimensionless value and the second dimensionless value may be based on process drift caused by accumulation of process materials (eg, increased thickness of a process film layer) in the first processing chamber and / or the second processing chamber.

[0137] At block 638, the processing device may update a processing recipe associated with the first processing chamber (e.g., process program 406 of FIG. 4 ) based on the first dimensionless value and the second dimensionless value. In some embodiments, the processing device may update the processing recipe based on the normalized first dimensionless value and / or the normalized second dimensionless value. Updating the processing recipe associated with the first processing chamber based on the first dimensionless value and the second dimensionless value may match or adjust the first processing chamber to the second processing chamber. Updating the processing recipe associated with the first processing chamber based on the first dimensionless value and the second dimensionless value may be included in a chamber matching procedure for matching the first chamber to the second chamber. In some embodiments, the processing recipe set point may be adjusted, such as temperature, pressure, gas flow, etc. In some embodiments, the processing recipe operation duration may be adjusted, such as adjusting an end point or a transition point.

[0138] refer to Fig.6D , a process 600D for updating a chamber process recipe endpoint is shown. The process 600D may be performed by the process control tool 428 of FIG. 4 . In some embodiments, at block 602, a processing device (e.g., of the process control tool 428 of FIG. 4 ) may receive OWPS data indicating one or more optical film properties of a process film layer deposited on an inner surface of a process chamber wall (e.g., on a process chamber liner). The OWPS data may indicate a thickness of the process film layer, a chemical composition of the process film layer, a structure or stacking arrangement of the process film layer, etc.

[0139] At block 624, the processing device may determine the characteristics of the film deposited on the wall of the processing chamber. The determination may be based on the OWPS data. In some embodiments, the processing device determines a dimensionless value indicating the thickness of the process film layer at block 626. The dimensionless value may be a ratio, a percentage, etc. of a target value of one or more characteristics of the film. In some embodiments, the dimensionless value may indicate a percentage thickness of an allowable thickness (e.g., a critical wall thickness) of the process film layer. The processing device may update the process operation recipe based on the dimensionless value. The processing device may update the process recipe endpoint (e.g., a cleaning recipe endpoint, a deposition recipe endpoint, an etching recipe endpoint, etc.), one or more recipe processing parameters (e.g., temperature, pressure, gas flow, etc.), one or more hardware parameters (e.g., placement of internal components) based on the dimensionless value at block 648. The process update to the process recipe may be related to the target process recipe endpoint condition. For example, the update to the process recipe may be related to the end of the process recipe at a certain point where the thickness of the process film layer is at (e.g., above or below) a predetermined value (e.g., a critical wall thickness). In some embodiments, the update to the process recipe causes the process recipe to end before the process film layer characteristic reaches a critical value (eg, before the film layer reaches a critical wall thickness). In some embodiments, the update to the process recipe causes the process recipe to end after the process film layer characteristic falls below a critical value.

[0140] Figure 7 According to various aspects of the present disclosure, a transparent film (e.g. Figure 3 The transparent film 307) covers the transparent window (eg Figure 1 Flowchart of method 700 for improving processing within a processing chamber by using one or more optical film properties of a process film layer of a transparent window 120 of a process chamber. Method 700 may be performed by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (e.g., running on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. For example, method 700 may be performed by controller 109 (e.g., processing device 130), as these components are referred to herein. In some embodiments, various baselines or thresholds (e.g., thickness changes or changes in n and k) may be stored and indexed for specific compositions in memory 134 of controller 109. These values ​​may be used by comparison by various methods 700 now discussed. Although shown in a particular sequence or order, the order of these processes may be modified unless otherwise specified. Therefore, the embodiments shown should be understood only as examples, and the processes shown may be performed in different orders, and some processes may be performed in parallel. In addition, one or more processes may be omitted in various embodiments. Therefore, not all processes are required in every embodiment. Other process flows are also possible.

[0141] refer to Figure 7 The method 700 may begin at block 710, where processing logic processes the reflection measurement signal to fit the reflection measurement signal to a thin film optical model to determine the deposition on the transparent film 307 (see Figure 3 ) of the process film layer 309 on the substrate 309. When the measurement is taken during processing of the substrate in the processing chamber, the method 700 may continue, where the processing logic determines at block 715 whether the property or characteristic, such as thickness, matches a baseline measurement of the process film layer, for example, in view of the composition of the process film layer 309. The processing logic may determine a dimensionless value indicative of a characteristic (e.g., thickness) of the process film layer to match the characteristic to a baseline, standard, or golden run measurement (e.g., a measurement associated with a processing run that is considered to have sufficiently high quality, performance, etc.). If there is a match, for example, if the one or more property or characteristic values ​​of the process film layer meet a threshold, the method 700 may continue at block 705 to process the substrate in the processing chamber. If there is no match, the method 700 may continue to block 720. At block 720, the processing logic determines a change in the rate of change of the value of the one or more properties of the process film layer. The rate of change may not meet a threshold performance metric (e.g., a layer may be deposited too fast or too slow).

[0142] In various embodiments, a threshold change (e.g., a change from a baseline measurement, a standard measurement, a golden run measurement, an average or median of several measurements, a two sigma change from the average, etc.) may be sufficient to warrant a change in a processing operation (e.g., a deposition process, an etch process, a cleaning process, etc.). Any such threshold change may be application specific and may be determined empirically based on how much change is sufficient to induce a process shift. The monitored process may be an etch process, a deposition process, a cleaning process, or the like.

[0143] The method 700 may continue to block 725, where processing logic triggers a process within the process chamber to correct the rate of deposition or removal of the process film layer. This portion of the method 700 may determine whether the process of the process chamber has shifted (or drifted) over time using a differential measurement compared to a baseline measurement (e.g., a golden run measurement, an average of several measurements, etc.) in this manner. In some embodiments, a machine learning model may be utilized to identify abnormal changes in the rate of film change. In some embodiments, an unsupervised machine learning model may be configured to detect optical measurements (e.g., optical measurement derivatives, dimensionless parameters, etc.) that indicate process failure, chamber drift or aging, component failure, etc.

[0144] The method 700 may similarly continue, where processing logic determines at block 730 whether the material composition of the process film layer 309 matches a baseline composition expected to be deposited on the transparent film. If the measured composition matches the baseline composition (e.g., within an acceptable error threshold), the method 700 may continue to block 705 to continue processing the substrate within the processing chamber. If there is no match, the method 700 may continue to block 735, where processing logic sends a feedback signal to the controller 109 (e.g., other computing device having an interface with a processing system operator) to report an error caused by a process drift detected within the processing chamber. When comparing the detected composition to the expected composition, a process shift may be detected, such as determining that the composition of the process film has drifted. In some embodiments, this portion of the method 700 may be performed when the processing system is inactive.

[0145] In some embodiments, the method 700 may continue to block 740, where the processing equipment determines whether the process film layer has reached a cumulative limit (e.g., a cumulative thickness limit within specifications) during or after processing the substrate in the processing chamber. If not, the method 700 may continue to process the substrate in the processing chamber at block 705. If the process film has reached the cumulative limit, the method 700 may continue, where the processing logic triggers the processing chamber at block 745 to initiate a cleaning process in the processing chamber. The processing equipment may determine whether the process film layer has reached the cumulative limit based on the dimensionless value indicating the thickness of the process film layer. The cleaning process may be intended to clean the surface of the processing equipment and the deposited film to improve future processing results and / or restore the processing equipment to certain specifications. The cleaning process may also use a plasma process, so there may also be an OES to subtract the spectrum determined when the light source is turned on to determine the reflection measurement signal on the process film.

[0146] In various embodiments, the method 700 may continue, wherein during such cleaning process triggered by operation 745, it is determined at block 750 whether the process film layer has been removed to a predetermined threshold thickness. The determination may be based on a dimensionless value indicating the thickness of the process film layer. Such a determination may be to check whether the process film layer has been sufficiently reduced by the cleaning process. Once the process film layer has been removed to or exceeds the predetermined threshold thickness, the method 700 may continue, wherein the processing logic triggers the processing chamber at block 755 to end the cleaning process performed in the processing chamber. Once completed, the method 700 may continue to process the substrate in the processing chamber at block 705.

[0147] Conceived as additional to or similar to Figure 7The method 700 of the present invention may be described in detail. For example, during processing of a substrate in a processing chamber, the processing logic may detect when a process film layer has reached a threshold thickness level on the transparent film 307. The processing logic may further trigger the end of a deposition process for depositing a deposited film layer in the processing chamber. The processing logic may make other decisions similar to this to update the process or process state of the processing chamber to improve substrate throughput, quality, and / or reduce product changeover.

[0148] According to some embodiments, the processing chamber may run a cleaning process after each substrate (or group of substrates) is processed, after a determined number of hours of operation, after a determined span of hours that the plasma is active in the chamber, after a certain volume of processing material is consumed, or at some other periodicity. The processing logic can accurately determine the moment when the endpoint of such cleaning is reached on the remote surface 120B or 420B of the transparent film 307. It can then send a trigger signal to the processing chamber to stop the cleaning and proceed to the next step, thereby improving throughput. If the cleaning rate of the remote surface 120B or 420B is different from the rest of the chamber, this difference can be characterized in advance and compensated via a lookup table in the controller 109.

[0149] Figure 8The model training workflow and model application workflow for chamber operation update prediction and simulation according to aspects of the present disclosure are illustrated. In an embodiment, the model training workflow 805 can be executed at a server, which may or may not include a chamber operation update predictor application, and the trained model is provided to the chamber operation update predictor application, which can execute the model application workflow 817. The model training workflow 805 and the model application workflow 817 can be executed by processing logic executed by a processor of a computing device (e.g., the modeling system 470 of Figure 4). One or more of these workflows 805, 817 can be implemented by one or more machine learning models, which are implemented on other software and / or firmware executed on the processing device and / or the processing device. The workflow can be implemented by one or more physics-based models, such as optical models. The workflow can be implemented by one or more digital twin models. As used herein, a digital twin is a digital replica of a physical asset (e.g., a manufactured part). The digital twin includes the properties of the physical asset at each stage of the manufacturing process, where these properties include but are not limited to coordinate axis dimensions, weight properties, material properties (such as density, surface roughness), electrical properties (such as conductivity), optical properties (such as reflectivity), etc. In some embodiments, the physics-based models and digital twin models can be trained to, for example, take into account unknown information (such as variations within manufacturing tolerances of components), to correct poor assumptions (such as non-ideal fluid flow), and so on.

[0150] Model training workflow 805 is used to train one or more machine learning models (e.g., deep learning models), physics-based models, etc. to perform one or more of the tasks of determining, predicting, modifying, updating, etc., which are associated with a chamber operation update predictor (e.g., determining updates to process operations associated with a processing chamber) and / or associated with a chamber matching tool, a sensor matching tool, a recipe development tool, and / or a process control tool, as illustrated in FIG4. Model application workflow 817 is used to apply the one or more trained models to perform tasks such as determining and / or updating for chamber data (e.g., raw sensor data, synthetic data, indicating the state of a processing chamber). One or more machines can receive process result data (e.g., chamber operation updates).

[0151] Various model outputs are described herein. Specific numbers and arrangements of models are described and shown. However, it should be understood that the number and type of models used and the arrangement of such models can be modified to achieve the same or similar end results. Therefore, the arrangements of the described and shown models (e.g., machine learning models, physics-based models, etc.) are only examples and should not be construed as limiting.

[0152] In some embodiments, one or more models are trained to perform one or more of the previously described tasks. Each task can be performed by a separate model. Alternatively, a single model can perform each task or a subset of tasks. Additionally or alternatively, different models can be trained to perform different combinations of tasks. In one example, one or more machine learning models can be trained, wherein the trained machine learning (ML) model is a single shared neural network having multiple shared layers and multiple higher-level distinct output layers, wherein each output layer outputs a different prediction, classification, recognition, etc. The tasks that the one or more trained models can be trained to perform may include operating a chamber operation update predictor 864, a chamber quality index generator 870, a performance matching model 874, a corrective action predictor 878, etc. As previously described, various input data such as sensor data, pre-processed data, synthetic data indicating the state of a processing chamber before and / or after a chamber operation (e.g., a cleaning operation, a deposition operation, an etching operation, etc.) can be received and processed by various machine learning models, rule-based models, physics-based models, etc. associated with the model application workflow 817.

[0153] The chamber operation update predictor 864 can output various values ​​corresponding to various chamber operation updates (e.g., chamber update values, process recipe value adjustments, etc.) based on various property values ​​of the process film layer on the inner surface of the process chamber wall (e.g., on the process chamber liner). The various values ​​can correspond to an index indicating the state of the process chamber. In some embodiments, the various values ​​correspond to a ratio and / or percentage of an allowable process film layer thickness (e.g., a critical wall thickness). The various values ​​can be used by a processing device (e.g., server 420 of FIG. 4) to determine updates to chamber operations (e.g., operating parameters, control knobs, durations, etc.). The chamber operation update predictor 864 generates chamber update data 866, which can be used to change the chamber, chamber operation, etc., for example, to improve the performance of the chamber.

[0154] The chamber quality index generator 870 may generate quality index data 872. In some embodiments, the quality index data 872 may include a dimensionless index indicative of chamber quality. In some embodiments, the quality index data 872 may include one or more CWI indices. In some embodiments, chamber quality index values ​​(e.g., CWI values) may be used as input to additional analysis (e.g., input data 862 to a chamber operation update predictor 864, a performance matching model 874, a corrective action predictor 878, etc. may include one or more CWI values).

[0155] The performance matching model 874 may accept input data 862 and generate as output performance matching data 876. The performance matching model 874 may include one or more algorithms, physics-based models, machine learning models, etc., configured to accept input data (e.g., spectral data, OWPS data, CWI data) and generate as output performance matching data 876. The performance matching data 876 may include data associated with corrective actions to be performed associated with one or more matching modalities (e.g., chamber-to-chamber matching, matching of current chamber performance to historical chamber performance, etc.).

[0156] The corrective action predictor 878 may accept input data 862 and generate corrective action data 880 as output. The input data may include sensor data, process parameter data, recipe data, product data, optical data, chamber quality index data, etc. The output may include recommended component replacement, recommended subsystem investigation (e.g., pressure subsystem, plasma generation subsystem, etc.), recommended maintenance, etc. In some embodiments, Figure 8 One or more of the blocks described in may not be performed, for example, analyzing the data stream may not include performance matching operations associated with blocks 874 and 876.

[0157] One machine learning model that can be used to perform some or all of the above tasks is an artificial neural network, such as a deep neural network. An artificial neural network typically includes a feature representation component with a classifier or regression layer that maps features to a desired output space. For example, a convolutional neural network (CNN) contains multiple layers of convolutional filters. At the lower layers, pooling is performed and nonlinear problems can be solved, and multi-layer perceptrons are usually attached to these lower layers to map the top features extracted by the convolutional layer to decisions (e.g., classification outputs). Deep learning is a class of machine learning algorithms that uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. Deep neural networks can be learned in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. A deep neural network includes a hierarchy of layers, with different layers learning different representation levels corresponding to different levels of abstraction. In deep learning, each layer learns to transform its input data into a slightly more abstract and comprehensive representation. For example, in plasma process tuning, the raw input may be chamber data (e.g., raw sensor chamber measurement parameters such as temperature, pressure, material thickness, vacuum conditions, etc. within the chamber); the second layer may consist of feature data associated with parameters associated with a substrate processing procedure or process recipe; and the third layer may include historical thresholds and chamber states under similar conditions. It is noteworthy that the deep learning process can learn by itself which features to best place at which layer. The "depth" in "deep learning" refers to the number of layers through which the transformed data passes. More specifically, deep learning systems have a considerable credit assignment path (CAP) depth. A CAP is a chain of transformations from an input to an output. A CAP describes the potential cause-effect relationship between an input and an output. For a feedforward neural network, the depth of the CAP can be the depth of the network and can be the number of hidden layers plus 1. For a recurrent neural network where a signal can propagate through a layer more than once, the CAP depth can be infinite.

[0158] In one embodiment, one or more machine learning models are recurrent neural networks (RNNs). An RNN is a neural network that includes memory, which enables the neural network to capture temporal dependencies. An RNN is able to learn input-output mappings that depend on both current inputs and past inputs. The RNN will process past and future sensor and / or process result measurements and make predictions based on this continuous metrology information. The RNN can be trained using a training data set to produce a fixed number of outputs (e.g., determining a set of substrate processing rates, determining chamber conditions, and production mode requirements and condition thresholds). One type of RNN that can be used is a long short-term memory (LSTM) neural network.

[0159] Training of neural networks can be accomplished in a supervised learning fashion, which involves feeding a training dataset consisting of labeled inputs through the network, observing its output, defining an error (defined by measuring the difference between the output and the label value), and using techniques such as deep gradient descent and backpropagation to adjust the weights of all layers and nodes of the network to minimize the error. In many applications, repeating this process for many labeled inputs in the training dataset results in a network that can produce correct outputs when presented with inputs that differ from those present in the training dataset.

[0160] For the model training workflow 805, a training data set should be formed using chamber data 810 (e.g., sensor data, OWPS data, synthetic data indicating the state of the associated processing chamber). In some embodiments, the training data set may also include associated process result data 812 (e.g., measured chamber variables (e.g., critical dimensions, uniformity requirements, film thickness results, etc.)) to form a training data set, where each data point may include various labels or classifications of one or more types of useful information. Each case may include, for example, data indicating the state of one or more processing chambers undergoing a process operation and / or associated process results of a substrate evaluated during and / or after the process operation. This data may be processed to generate one or more training data sets 836 for training one or more machine learning models. The machine learning model may be trained, for example, to predict chamber process operation parameter updates.

[0161] To implement training, processing logic inputs the training data set 836 into one or more untrained machine learning models. Before the first input is input into the machine learning model, the machine learning model can be initialized. Processing logic trains the untrained machine learning model based on the training data set to generate one or more trained machine learning models that perform the various operations set forth above.

[0162] Training may be performed by inputting one or more of the chamber data 810 and the process result data 812 into the machine learning model one at a time.

[0163] Machine learning models process inputs to produce outputs. Artificial neural networks include an input layer, which consists of the values ​​in the data points. The next layer is called a hidden layer, and the nodes at the hidden layer each receive one or more input values. Each node contains parameters (such as weights) applied to the input values. Therefore, each node basically inputs the input values ​​into a multivariate function (for example, a nonlinear mathematical transformation) to produce an output value. The next layer may be another hidden layer, or an output layer. In both cases, the nodes at the next layer receive output values ​​from the nodes at the previous layer, and each node applies weights to those values ​​and then produces its own output value. This can be performed at each layer. The last layer is the output layer, where there is a node for each category, prediction, and / or output that the machine learning model can produce.

[0164] Thus, the output may include one or more predictions or inferences. For example, the output predictions or inferences may include one or more modifications to the plasma process data (e.g., modifications to one or more plasma exposure durations), updates to the cleaning recipe (e.g., updates to the duration of the cleaning recipe). The processing logic may cause the processing chamber to perform the updated process operation. The processing logic may compare the results of the updated process operation with the results of the process operation before the update. The processing logic determines an error (i.e., a chamber update error) based on the difference between the identified process operation results. The processing logic adjusts the weights of one or more nodes in the machine learning model based on the error. An error term or delta may be determined for each node in the artificial neural network. Based on this error, the artificial neural network adjusts one or more parameters of one or more of its nodes (the weights of one or more inputs of the node). The parameters may be updated in a back-propagation manner so that the nodes at the highest level are updated first, followed by the nodes at the next level, and so on. The artificial neural network includes multiple layers of "neurons", each of which receives as input values ​​from neurons at the previous layer. The parameters of each neuron include weights associated with the values ​​received from each neuron at the previous layer. Therefore, adjusting the parameters may include adjusting the weights of each input of one or more neurons assigned to one or more layers in the artificial neural network.

[0165] Once the model parameters are optimized, model validation can be performed to determine whether the model has improved and to determine the current accuracy of the deep learning model. After one or more rounds of training, the processing logic can determine whether the stopping criteria have been met. The stopping criteria can be a target level of accuracy, a target number of processed data blocks from a training data set, a target amount of change in a parameter relative to one or more previous data points, a combination of the above, and / or other criteria. In one embodiment, the stopping criteria are met when at least a minimum number of data points have been processed and at least a threshold accuracy has been reached. The threshold accuracy can be, for example, 70%, 80%, or 90% accuracy. In one embodiment, if the accuracy of the machine learning model has stopped improving, the stopping criteria are met. If the stopping criteria are not met, further training is performed. If the stopping criteria are met, the training may be completed. Once the machine learning model is trained, a retained portion of the training data set can be used to test the model.

[0166] Once one or more trained machine learning models 838 are generated, they can be stored in a model storage 845 and can be added to a manufacturing system (e.g., manufacturing system 402 of FIG. 4 ), a modeling system (e.g., modeling system 470 of FIG. 4 ), etc. Subsequently, the modeling system, manufacturing system, etc. can use the one or more trained ML models 838 and additional processing logic to implement an automatic mode in which user manual input information is minimized or even eliminated in some cases.

[0167] For the model application workflow 817, according to one embodiment, input data 862 may be input into a chamber operation update predictor 864, which may include a trained machine learning model. Based on the input data 862, the chamber operation update predictor 864 outputs updated information indicative of chamber process operations associated with the processing chamber conditions.

[0168] In some embodiments, input data 862 may be provided to a chamber quality index generator 870 and an indication of the chamber quality may be output as quality index data 872. The chamber quality index generator 870 may include one or more trained machine learning models. In some embodiments, the chamber quality index value may be used as an input to a further analysis model (e.g., a machine learning model).

[0169] In some embodiments, input data 862 can be provided to a performance matching model 874. Performance matching model 874 can include one or more machine learning models. Performance matching model 874 can be configured to generate data associated with performance matching, such as performance matching data 876. Performance matching can include matching chamber performance to another chamber's performance (e.g., changing a chamber's performance to be more similar to a second chamber), matching a chamber's performance to historical performance (e.g., changing a chamber's performance to be more similar to a performed processing run that produced a product that met a performance threshold), and the like.

[0170] In some embodiments, input data 862 may be provided to a corrective action predictor 878. The corrective action predictor 878 may include one or more trained machine learning models. The corrective action predictor 878 may include a supervised machine learning model, such as a model trained using data labeled with associated corrective actions. The corrective action predictor 878 may generate corrective action data 880 as an output, which may include a suggested corrective action associated with the input data 862.

[0171] Fig. 9 A block diagram of an example computing device operating in accordance with one or more aspects of the present disclosure is depicted. In various illustrative examples, various components of computing device 900 may represent various components of client device 450, metering system 410, server 420, data storage 140, and modeling system 470 shown in FIG.

[0172] The example computing device 900 can be connected to other computer devices in a LAN, an intranet, an extranet, and / or the Internet. The computing device 900 can operate as a server in a client-server network environment. The computing device 900 can be a personal computer (PC), a set-top box (STB), a server, a network router, a switch or a bridge, or any device capable of executing a set of instructions (in sequence or otherwise) that specify actions to be taken by the device. Further, although only a single example computing device is illustrated, the term "computer" should also be considered to include a collection of any computers that execute a set (or multiple sets) of instructions to perform any one or more methods discussed herein, either individually or collectively.

[0173] The example computing device 900 may include a processing device 902 (also referred to as a processor or CPU), a main memory 904 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous dynamic random access memory (SDRAM), etc.), a static memory 906 (e.g., flash memory, static random access memory (SRAM), etc.), and an auxiliary memory (e.g., a data storage device 918), which may communicate with each other via a bus 930.

[0174] The processing device 902 represents one or more general-purpose processing devices, such as a microprocessor, a central processing unit, or the like. In more detail, the processing device 902 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 that implements other instruction sets, or a processor that implements a combination of instruction sets. The processing device 902 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. In accordance with one or more aspects of the present disclosure, the processing device 902 may be configured to execute instructions that implement the methods 500A-D and / or shown in Figures 5A-D. Fig. 6A -D shows the process 600A-D.

[0175] The example computing device 900 may further include a network interface device 908, which may be communicatively coupled to a network 920. The example computing device 900 may further include a video display 910 (e.g., a liquid crystal display (LCD), a touch screen, or a cathode ray tube (CRT)), an alphanumeric input device 912 (e.g., a keyboard), a cursor control device 914 (e.g., a mouse), and an acoustic signal generating device 916 (e.g., a speaker).

[0176] The data storage device 918 may include a computer-readable storage medium (or more specifically, a non-transitory machine-readable storage medium) 928 on which one or more sets of executable instructions 922 are stored. According to one or more aspects of the present disclosure, the executable instructions 922 may include instructions for executing the methods 500A-D and / or Fig. 6A -D shows the executable instructions associated with the process 600A-D.

[0177] The executable instructions 922 may also reside completely or at least partially within the main memory 904 and / or within the processing device 902 during execution of the instructions by the example computing device 900, the main memory 904 and the processing device 902 also constituting computer-readable storage media. The executable instructions 922 may further be transmitted or received over a network via the network interface device 908.

[0178] Despite Fig. 9 928 is shown as a single medium, but the term "computer-readable storage medium" should also be considered to include a single medium or multiple media (such as a centralized or distributed database and / or associated caches and servers) storing the one or more sets of operating instructions. The term "computer-readable storage medium" should also be considered to include any medium capable of storing or encoding a set of instructions for execution by a machine, causing the machine to perform any one or more of the methods described herein. Therefore, the term "computer-readable storage medium" should be considered to include (but not limited to) solid-state memory and optical and magnetic media.

[0179] Some portions of the foregoing detailed description are presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the art of data processing to most effectively communicate the content of their work to other skilled in the art. Here, an algorithm is generally considered to be a self-consistent sequence of steps leading to a desired result. These steps are steps that require physical manipulation of physical quantities. Usually, although not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It proves convenient at times, primarily for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0180] It should be remembered, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless otherwise specifically stated, it will be understood from the discussion below that throughout the description, discussions utilizing terms such as "identify," "determine," "store," "adjust," "result," "return," "compare," "create," "stop," "load," "copy," "throwing," "replace," "execute," "receive," or similar terms refer to the actions and processes of a computer system or similar electronic computing device that manipulates and transforms data represented as physical (electronic) quantities within the computer system's buffers and memories into other data similarly represented as physical quantities within the computer system's memories or buffers or other such information storage, transmission, or display devices.

[0181] Examples of the present disclosure are also related to an apparatus for performing the methods described herein. This apparatus may be specially constructed for the required purpose, or it may be a general purpose computer system selectively programmed by a computer program stored in the computer system. Such a computer program may be stored in a computer readable storage medium, such as but not limited to any type of disk (including optical disk, CD-ROM and magneto-optical disk), read-only memory (ROM), random access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disk storage media, optical storage media, flash memory devices, other types of machine-accessible storage media, or any type of media suitable for storing electronic instructions, each of which is coupled to a computer system bus.

[0182] The foregoing description has been described many specific details of examples such as specific systems, components, methods, etc., so that people can have a good understanding of several embodiments of the present disclosure. However, it will be appreciated by those skilled in the art that at least some embodiments of the present disclosure can be implemented without these specific details. In other cases, well-known components or methods are not described in detail, or the components or methods are presented in a simple block diagram format to avoid unnecessary concealment of the present disclosure. Therefore, the specific details described are only exemplary. Specific implementations may be different from these exemplary details and are still considered to be within the scope of the present disclosure.

[0183] References throughout this specification to "one / a embodiment" mean that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, the phrase "in one / a embodiment" appearing in various places throughout this specification does not necessarily refer 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, the term is intended to mean that the accuracy of the nominal value presented is within ±10%.

[0184] Although the operations of the methods herein are shown and described in a particular order, the order of operations of each method may be changed so that certain operations may be performed in reverse order so that certain operations may be performed at least partially in parallel with other operations. In another embodiment, instructions or sub-operations of different operations may be performed in an intermittent and / or alternating manner.

[0185] It should be understood that the above description is intended to be illustrative, rather than restrictive. After reading and understanding the above description, those skilled in the art will find many other embodiments. Therefore, the scope of this disclosure will be determined with reference to the entire scope of the equivalents given by the accompanying claims and such claims.

Claims

1. A method, include: receiving, by a processing device, first sensor data indicative of a condition of a wall corresponding to a first processing chamber, the first sensor data comprising optical spectroscopy data; determining, by the processing device, a first value based on the first sensor data, the first value corresponding to a first amount of product disposed along a surface of the wall at a first time; determining, by the processing tool, a first update to a first process operation associated with the first processing chamber based on the first value; as well as By the processing device, one or more of (i) preparing a notification indicating the first update for presentation on a graphical user interface (GUI), or (ii) causing the first process operation to be performed in accordance with the first update.

2. The method of claim 1, wherein the first sensor data is received from an optical wall process sensor (OWPS), the method further comprising: include: An update to at least one of a sensor gain value or a sensor offset value corresponding to the OWPS is determined, by the processing device, based on the first value.

3. The method of claim 1, wherein the first value corresponds to the first amount of the product disposed along the surface at a first time, the method further comprising: include: by the processing device, based on the first sensor data a second value, wherein the second value corresponds to a second amount of the product disposed along the surface of the wall at a second time; as well as A process rate associated with the first process operation is determined, by the processing tool, using the first value and the second value, wherein the first update is further determined using the process rate. 4 . The method of claim 3 , wherein the first update corresponds to at least one of a duration of the first process operation or an endpoint condition of the first process operation.

5. The method of claim 3, wherein at least one of the process rate or the endpoint condition is associated with at least one of a cleaning rate, an etching rate, or a deposition rate corresponding to the first processing chamber.

6. The method of claim 1, wherein the first value is indicative of a thickness of the product disposed along the surface of the wall.

7. The method of claim 1, further comprising: include: receiving, by the processing device, second sensor data indicating a second state of a second wall corresponding to a second processing chamber, wherein the second sensor data includes second optical spectral data; and determining, by the processing device, a second value based on the second sensor data, wherein the second value corresponds to a second amount of the product disposed along the second surface of the second wall at a second time, Wherein the first update is further determined using the second value.

8. The method of claim 7, further comprising: include: Calibration of one or more sensors associated with the first process chamber is performed based on a comparison between the first value and the second value, caused by the processing device.

9. The method of claim 1, further comprising: include: Data normalization is performed, by the processing device, using the first sensor data to determine the first value, wherein the first value comprises a normalized dimensionless value.

10. The method of claim 1, further comprising: include: using, by the processing device, the first sensor data as input to a machine learning (ML) model; as well as One or more outputs of the ML model are received, by the processing device, the one or more outputs indicating the first update.

11. A system, include: a processing chamber comprising a first wall having an inner surface positioned within the processing chamber; an optical sensor coupled to the processing chamber, the optical sensor configured to acquire spectral data associated with the interior surface; Memory; as well as A processing device is coupled to the memory, and the processing device is configured to: receiving first sensor data indicative of a condition of a wall corresponding to a first processing chamber, the first sensor data comprising optical spectroscopy data; determining a first value based on the first sensor data, the first value corresponding to a first amount of product disposed along a surface of the wall at a first time; determining a first update to a first process operation associated with the first processing chamber based on the first value; as well as Performing one or more of (i) preparing a notification indicating the first update for presentation on a graphical user interface (GUI), or (ii) causing the first process operation to be performed in accordance with the first update.

12. The system of claim 11, wherein the first sensor data is received from an optical wall process sensor (OWPS), and wherein the processing device is further configured to: An update to at least one of a sensor gain value or a sensor offset value corresponding to the OWPS is determined based on the first value.

13. The system of claim 11, wherein the first value corresponds to the first amount of the product disposed along the surface at a first time, and wherein the processing device is further configured to: determining a second value based on the first sensor data, wherein the second value corresponds to a second amount of the product disposed along the surface of the wall at a second time; and A process rate associated with the first process operation is determined using the first value and the second value, wherein the first update is further determined using the process rate.

14. The system of claim 11, wherein the first value indicates a thickness of the product disposed along the surface of the wall.

15. The system of claim 11, wherein the processing device is further configured to: receiving second sensor data indicative of a second state of a second wall corresponding to a second processing chamber, wherein the second sensor data comprises second optical spectral data; and determining a second value based on the second sensor data, wherein the second value corresponds to a second amount of the product disposed along the second surface of the second wall, Wherein the first update is further determined using the second value.

16. A non-transitory machine-readable storage medium storing instructions which, when executed, cause a processing device to: receiving first sensor data indicative of a condition of a wall corresponding to a first processing chamber, the first sensor data comprising optical spectroscopy data; determining a first value based on the first sensor data, the first value corresponding to a first amount of product disposed along a surface of the wall at a first time; determining a first update to a first process operation associated with the first processing chamber based on the first value; as well as Performing one or more of (i) preparing a notification indicating the first update for presentation on a graphical user interface (GUI), or (ii) causing the first process operation to be performed in accordance with the first update.

17. The non-transitory machine-readable storage medium of claim 16, wherein the first sensor data is received from an optical wall process sensor (OWPS), and wherein the processing device is further configured to: An update to at least one of a sensor gain value or a sensor offset value corresponding to the OWPS is determined based on the first value.

18. The non-transitory machine-readable storage medium of claim 17, wherein the first value corresponds to the first amount of the product disposed along the surface at a first time, and wherein the processing device is further configured to: determining a second value based on the first sensor data, wherein the second value corresponds to a second amount of the product disposed along the surface of the wall at a second time; and A process rate associated with the first process operation is determined using the first value and the second value, wherein the first update is further determined using the process rate.

19. The non-transitory machine-readable storage medium of claim 16, wherein the processing device is further configured to: receiving second sensor data indicative of a second state of a second wall corresponding to a second processing chamber, wherein the second sensor data comprises second optical spectral data; and determining a second value based on the second sensor data, wherein the second value corresponds to a second amount of the product disposed along the second surface of the second wall, Wherein the first update is further determined using the second value.

20. The non-transitory machine-readable storage medium of claim 16, wherein the processing device is further configured to: using the first sensor data as input to an ML model; and One or more outputs of the ML model are received, the one or more outputs indicating the first update.