Characterizing substrate supports to build a digital twin

Abstract

The method includes receiving data indicative of a characteristic of the substrate support from one or more sensors of a removable sensor assembly disposed proximate to the substrate support. The method further includes providing data based on the data indicative of the characteristic of the substrate support to a physics-based model of the substrate support. The method further includes receiving predicted performance data of the substrate support from the physics-based model.

Description

[Technical Field]

[0001] This specification relates to characterizing a substrate support for substrate processing, and more particularly, to characterizing a substrate support for building a physics-based digital twin model of the substrate support. [Background technology]

[0002] Chambers are used in many types of processing systems. Examples of chambers include etch chambers, deposition chambers, anneal chambers, etc. Typically, a substrate, such as a semiconductor wafer, is placed on a substrate support within the chamber, and conditions within the chamber are set and maintained to process the substrate. The characteristics of the substrate support affect the properties of the finished substrate. Summary of the Invention

[0003] The following is a simplified summary of the disclosure to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is not intended to identify key or critical elements of the disclosure, nor is it intended to delineate the scope or claims of particular implementations of the disclosure. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.

[0004] In one aspect of the present disclosure, a method includes receiving data indicative of a characteristic of a substrate support from one or more sensors of a removable sensor assembly positioned proximate to the substrate support. The method further includes providing data based on the data indicative of the characteristic of the substrate support to a physics-based model of the substrate support. The method further includes receiving predicted performance data of the substrate support from the physics-based model.

[0005] In another aspect of the present disclosure, a method includes receiving predicted performance data for a substrate support from a physics-based model of the substrate support. The method further includes determining predicted characteristics of a substrate to be processed using the substrate support taking into account the predicted performance data of the substrate support. The method further includes performing corrective action taking into account the predicted characteristics of the substrate.

[0006] In another aspect of the present disclosure, a sensor assembly system is disclosed. The sensor assembly system includes a stand configured to be positioned proximate to a substrate support. The sensor assembly system further includes a movable support coupled to the stand, the movable support being movable on at least two axes. The sensor assembly further includes a plurality of sensors disposed on the movable support. The movable support of the sensor assembly system is configured to move the sensors to a series of positions relative to the substrate support. The sensors generate one or more measurements at each of the positions relative to the substrate support.

[0007] The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a block diagram illustrating an example system (example system architecture), according to some embodiments. [Figure 2] FIG. 10 is a block diagram of an exemplary dataset generator used to create a dataset for a model, according to some embodiments. [Figure 3] FIG. 2 is a block diagram illustrating a system for generating output data (e.g., prediction data 168 of FIG. 1) according to some embodiments. [Figure 4A] 1 is a flow diagram of a method associated with characterizing a substrate support and triggering corrective action, according to some embodiments. [Figure 4B]1 is a flow diagram of a method associated with characterizing a substrate support and triggering corrective action, according to some embodiments. [Figure 4C] 1 is a flow diagram of a method associated with characterizing a substrate support and triggering corrective action, according to some embodiments. [Figure 4D] 1 is a flow diagram of a method associated with characterizing a substrate support and triggering corrective action, according to some embodiments. [Figure 4E] 1 is a flow diagram of a method associated with characterizing a substrate support and triggering corrective action, according to some embodiments. [Figure 5] 1 illustrates a cross-sectional view of a manufacturing chamber (e.g., a semiconductor wafer manufacturing chamber) according to some embodiments. [Figure 6] FIG. 1 illustrates a removable sensor assembly system, according to some embodiments. [Figure 7] FIG. 1 is a block diagram illustrating a computer system, according to some embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0009] Described herein are techniques directed to characterizing properties of a substrate support (e.g., a chuck assembly associated with a substrate manufacturing apparatus) using a sensor assembly system to build a physics-based digital twin model. The manufacturing apparatus is used to produce substrates, such as semiconductor wafers. These substrate properties are determined by the conditions under which the substrates are processed. Accurate knowledge of property values ​​within an operating manufacturing chamber, particularly in the immediate vicinity of the substrate, allows for prediction of final product properties, consistent production of substrates with the same properties, and adjustment of process parameters to optimize substrate production.

[0010] In some cases, wafers are placed on a substrate support for processing. The substrate support may include some processing architecture, e.g., an electrostatic chucking mechanism, a heater, etc. Substrate support assemblies are typically manufactured within tolerance levels. In some cases, a manufacturing process using a particular substrate support consistently produces high-quality substrates. In some cases, even if the substrate support is within manufacturing tolerances, a manufacturing process using a particular substrate support may not produce high-quality substrates. Often, determining the quality of substrates produced using a particular substrate support (and, by extension, selecting an appropriate substrate support) is done empirically. Such an approach involves experimentation and can be costly in terms of time spent, materials used, energy and gas used, etc. Additionally, as manufacturing chambers are used repeatedly, their characteristics tend to drift due to material accumulation, aging of parts, temperature and gas cycling, changes during cleaning procedures, etc. Substrate support characteristics may drift. Any knowledge gained to map input process parameters to output substrate characteristics degrades as chamber and substrate support characteristics drift. A further problem with this conventional approach is that the substrate support system has a finite lifespan. Ultimately, the process of replacing substrate support components, producing wafers, performing measurements on the wafers, and selecting substrate supports based on the measurements is repeated. The problem is exacerbated when attempting to produce new types of wafers using process parameters outside the range of process parameters understood by conventional methods.

[0011] In one or more embodiments, methods and devices of the present disclosure address at least these shortcomings of conventional approaches. According to the present disclosure, a sensor assembly can be used to characterize a property of a substrate support. Property values ​​associated with the support, measured by sensors in the sensor assembly, are then provided to inform a physics-based digital twin model of the substrate support. In this manner, characterization of the substrate support can be performed directly, avoiding costs associated with producing, characterizing, and disposing of the substrate. In some embodiments, the methods of the present disclosure include the use of a removable sensor assembly. The sensor assembly includes a support device for the sensor and a means for securing the assembly in proximity to the substrate support. In some embodiments, the substrate support may be located within a fabrication chamber. In some embodiments, measurements of the properties of the substrate support may be performed outside of the fabrication chamber (e.g., before the substrate support is installed in the fabrication chamber). Data collected by the sensor assembly is correlated to substrate support performance. For example, the substrate support may include electrodes beneath the surface of the support. The electrodes may include chuck electrodes (e.g., in an electrostatic chuck), heater electrodes, RF electrodes, etc. The sensors of the sensor assembly can measure properties of the substrate support that are indicative of the performance of these electrodes, such as measuring the distance below the surface of the substrate support at which the electrodes are located, measuring properties related to the interaction between the substrate support and the substrate (e.g., surface profilometry that characterizes heat transfer to the wafer), etc.

[0012] In some embodiments, the disclosed methods and devices further include providing data generated by considering the measurement data (e.g., processed measurement data) from the sensor assembly to a physics-based digital twin model. The physics-based digital twin model is configured to provide a solution to an equation (e.g., a heat transfer equation) related to the performance of the substrate support. Various parameters involved in the physics-based digital twin model may be calculated based on the measurements provided by the sensor assembly. In some embodiments, the disclosed methods further include receiving predicted performance data associated with the substrate support from the physics-based model.

[0013] In some embodiments, the disclosed method further includes using the predicted performance data of the substrate support (output by the physics-based model) to generate predicted characteristics of a substrate produced using the substrate support. The characteristics of a substrate are strongly correlated with conditions proximate to the substrate. The performance of the substrate support affects these conditions. For example, a non-uniform temperature profile at the surface of the substrate support may result from a non-uniform depth of at least one heating electrode below the surface of the substrate support. The non-uniform temperature profile at the surface of the substrate support may cause non-uniform processing of the substrate, e.g., non-uniform deposition, etching, etc. The method may further include performing corrective action taking into account the predicted substrate characteristics (e.g., updating a process recipe to produce a substrate having characteristics different from those predicted).

[0014] The present disclosure further enables, in some embodiments, a removable sensor assembly system. The sensor assembly system can be used to measure properties of a substrate support. In some embodiments, the sensor assembly system can be used inside a fabrication chamber (e.g., to measure properties of a substrate support installed in the fabrication chamber). In some embodiments, the sensor assembly system can be used outside the fabrication chamber. The sensor assembly can include a stand configured to secure a sensor of the sensor assembly proximate to the substrate support. The sensor assembly system can further include a movable support coupled to the stand. The movable support can be movable on two or more axes. The movable support causes relative motion between the sensor assembly and the substrate support. The measurement area of ​​the sensor can be smaller than the surface of the substrate support, and the movable support can be used to perform a scan of the entire surface of the support. In some embodiments, the substrate support can be coupled to the movable support, and the sensor assembly can remain stationary.

[0015] Aspects of the present disclosure provide technical advantages over conventional methods. Characterization of a substrate support can be achieved by taking measurements and feeding data indicative of those measurements into a physics-based digital twin model. This characterization avoids the traditional, expensive method of testing a substrate support by using the substrate support to fabricate wafers, performing measurements on the wafers, and inferring the quality of the substrate support from the measurements. Aspects of the present disclosure also enable an improved selection process for a substrate support to use in a manufacturing process when multiple substrate supports are available. Some substrate supports may be provided with mechanisms for spatially localizing control of some characteristics (e.g., multiple heating electrodes positioned in different regions below the surface of the substrate support). Characterizing the substrate support enables adjustment of substrate control to create more favorable conditions for substrate manufacturing (e.g., supplying different voltages to each heater electrode to create a target temperature profile at the surface of the substrate support). Additionally, the substrate support can be recharacterized (e.g., during a planned maintenance event) to account for any drift in components, and if a problem is suspected in the manufacturing chamber, the substrate support can be recharacterized to confirm or rule out that component as the cause.

[0016] 1 is a block diagram illustrating an example system 100 (example system architecture) according to some embodiments. System 100 includes client devices 120, manufacturing equipment 124, sensors 126, metrology equipment 128, a prediction server 112, a data store 140, and a sensor assembly 195. Prediction server 112 may be part of a prediction system 110. Prediction system 110 may further include server machines 170 and 180.

[0017] In some embodiments, the manufacturing tool 124 (e.g., a cluster tool) is part of a substrate processing system (e.g., an integrated processing system). The manufacturing tool 124 includes one or more of a controller, an enclosure system (e.g., a substrate carrier, a front-opening unified pod (FOUP), an autoteach FOUP, a process kit enclosure system, a substrate enclosure system, a cassette, etc.), a side storage pod (SSP), an aligner device (e.g., an aligner chamber), a factory interface (e.g., an equipment front-end module (EFEM)), a load lock, a transfer chamber, one or more processing chambers, a robot arm (e.g., disposed in the transfer chamber, disposed in the front interface, etc.), etc. The enclosure system, the SSP, and the load lock are attached to the factory interface, and the robot arm disposed in the factory interface is for transferring contents (e.g., substrates, process kit rings, carriers, verification wafers, etc.) between the enclosure system, the SSP, the load lock, and the factory interface. The aligner device is disposed in the factory interface and aligns the contents. The load locks and processing chambers are mounted in a transfer chamber, and a robotic arm located in the transfer chamber is for transferring contents (e.g., substrates, process kit rings, carriers, validation wafers, etc.) between the load locks, processing chambers, and the transfer chamber. In some embodiments, the manufacturing tool 124 includes components of a substrate processing system. In some embodiments, the manufacturing tool 124 is used to produce one or more products (e.g., substrates, semiconductors, wafers, etc.). In some embodiments, the manufacturing tool 124 is used to produce one or more components used in a substrate processing system. In some embodiments, the manufacturing tool 124 is used to produce and / or includes a bonded metal plate structure (e.g., a showerhead used in a processing chamber of a substrate processing system).

[0018] The sensors 126 can provide sensor data 142 associated with the manufacturing equipment 124 (e.g., associated with the manufacturing of a corresponding product, such as a wafer, by the manufacturing equipment 124). The sensor data 142 can be used, for example, for equipment health and / or product health (e.g., product quality). The manufacturing equipment 124 can operate according to a recipe or over a period of time to produce a product. In some embodiments, the sensor data 142 can include one or more values ​​of temperature (e.g., heater temperature), spacing (SP), pressure, high frequency radio frequency (HFRF), electrostatic chuck (ESC) voltage, current, flow rate (e.g., of one or more gases), power, voltage, etc. The sensor data 142 can include historical sensor data and current sensor data. The manufacturing equipment 124 can be configured according to manufacturing parameters 150. The manufacturing parameters 150 can be associated with or indicative of parameters such as hardware parameters (e.g., settings or components (e.g., size, type, etc.) of the manufacturing equipment 124) and / or process parameters of the manufacturing equipment. The manufacturing parameters 150 may include historical manufacturing data and / or current manufacturing data. The manufacturing parameters 150 may indicate input settings to the manufacturing devices (e.g., heater power, gas flow, etc.). The sensor data 142 and / or manufacturing parameters 150 may be provided while the manufacturing equipment 124 is performing the manufacturing process (e.g., equipment readings as the product is processed). The sensor data 142 may vary from product to product (e.g., from wafer to wafer).

[0019] In some embodiments, the sensor data 142, the metrology data 160, and / or the manufacturing parameters 150 may be processed (e.g., by the client device 120 and / or by the prediction server 112). Processing the sensor data 142 may include generating features. In some embodiments, the features are patterns (e.g., slope, width, height, peaks, etc.) of the sensor data 142, the metrology data 160, and / or the manufacturing parameters 150, or combinations of values ​​(e.g., power derived from voltage and current, etc.) from the sensor data 142, the metrology data 160, and / or the manufacturing parameters 150. The sensor data 142 may include features, which may be used by the prediction component 114 to perform signal processing and / or to obtain prediction data 168, and possibly to perform corrective actions. The prediction data 168 may be any data associated with the prediction system 110, such as predicted performance data of a substrate support.

[0020] Each instance (e.g., set) of sensor data 142 may correspond to a product (e.g., wafer), a set of manufacturing equipment, a type of substrate produced by the manufacturing equipment, combinations thereof, etc. Each instance of metrology data 160 and manufacturing parameters 150 may similarly correspond to a product, a set of manufacturing equipment, a type of substrate produced by the manufacturing equipment, combinations thereof, etc. The data store may further store information associating sets of different data types, e.g., information indicating that a set of sensor data, a set of metrology data, and / or a set of manufacturing data are all associated with the same product, manufacturing equipment, type of substrate, etc.

[0021] In some embodiments, the predictive system 110 may use machine learning to generate predictive data 168 (e.g., target outputs including data indicative of manufacturing failures provided by the predictive system 110), as discussed in more detail below.

[0022] Client devices 120, manufacturing tools 124, sensors 126, metrology tools 128, prediction server 112, data store 140, server machine 170, server machine 180, and sensor assembly 195 may be coupled together via network 130 to generate sensor assembly data 152, substrate support data 154, and / or prediction data 168. This data may be used in performing corrective actions.

[0023] In some embodiments, network 130 is a public network that provides client device 120 with access to prediction server 112, data store 140, and / or other publicly available computing devices. In some embodiments, network 130 is a private network that provides client device 120 with access to manufacturing equipment 124, sensors 126, metrology equipment 128, data store 140, sensor assembly 195, and / or other privately available computing devices. Network 130 may include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.

[0024] The client device 120 may include computing devices such as a personal computer (PC), a laptop, a mobile phone, a smartphone, a tablet computer, a netbook computer, a network-connected television ("smart TV"), a network-connected media player (e.g., a Blu-ray player), a set-top box, an over-the-top (OTT) streaming device, an operator box, etc. The client device 120 may include a corrective action component 122. The corrective action component 122 may receive user input of instructions associated with the manufacturing equipment 124 (e.g., via a graphical user interface (GUI) displayed via the client device 120). In some embodiments, the corrective action component 122 sends instructions to the prediction system 110, receives output (e.g., prediction data 168) from the prediction system 110, determines corrective actions based on the output, and causes the corrective actions to be implemented.

[0025] In some embodiments, the prediction system 110 can further include a prediction component 114. The prediction component 114 can generate prediction data 168 using data obtained from the model 190. In some embodiments, the prediction component 114 provides the prediction data 168 to the client device 120, which takes the prediction data 168 into account and triggers a corrective action via the corrective action component 122. In some embodiments, the corrective action component 122 obtains sensor assembly data 152 associated with a substrate support of the manufacturing tool 124 (e.g., from the data store 140, etc.) and provides the sensor assembly data 152 associated with the substrate support of the manufacturing tool 124 to the prediction system 110.

[0026] In some embodiments, the corrective action component 122 stores the sensor assembly data 152 in the data store 140, and the prediction server 112 retrieves the sensor assembly data 152 from the data store 140. In some embodiments, the prediction server 112 can store the output of the trained model 190 (e.g., the prediction data 168, the substrate support data 154) in the data store 140, and the client device 120 can retrieve the output from the data store 140. In some embodiments, the corrective action component 122 receives corrective action instructions from the prediction system 110 and causes the corrective action to be implemented. Each client device 120 can include an operating system that enables a user to one or more of create, view, or edit data (e.g., instructions associated with the manufacturing equipment 124, corrective actions associated with the manufacturing equipment 124, etc.).

[0027] The sensor assembly data 152 includes data collected by a sensor assembly 195. The sensor assembly 195 is configured to collect data describing characteristics of a substrate support (e.g., an electrostatic chuck, a vacuum chuck, etc.). The sensor assembly data 152 may be provided to a physics-based digital twin model (e.g., model 190), which may output predicted performance data (e.g., substrate support data 154) for the substrate support. The substrate support data 154 may be taken as input by a prediction system 110 to generate predicted property data (e.g., predicted data 168) for substrates produced using the substrate support as output. The prediction system 110 may input the substrate support data 154 into a model (e.g., model 190), which may include a trained machine learning model, to generate output predicted property data for the substrate. The predicted property data and corresponding substrate metrology data 160 may be compared by the prediction system 110 to output predictions of manufacturing faults, chamber component drift, etc., and / or corrective actions. The prediction system 110 can input the predicted property data and metrology data 160 into a model (e.g., model 190) to generate data indicative of corrective action. The model may, in some embodiments, be a trained machine learning model. In some embodiments, the corrective action may be an update to the manufacturing parameters 150 for future substrates produced by the manufacturing tool 124.

[0028] In some embodiments, metrology data 160 corresponds to historical characteristic data of a product (e.g., produced using historical sensor data and manufacturing parameters associated with historical manufacturing parameters), and predictive data 168 is associated with predicted characteristic data (e.g., of a product that will be or has been produced under conditions recorded by current sensor data and / or current manufacturing parameters). In some embodiments, predictive data 168 is predicted metrology data (e.g., virtual metrology data) of a product that will be or has been produced according to conditions recorded as current sensor data and / or current manufacturing parameters. In some embodiments, predictive data 168 is or includes an indication of an anomaly (e.g., an abnormal product, an abnormal component, an abnormal manufacturing equipment, an abnormal energy usage, etc.) and / or one or more causes of the anomaly. In some embodiments, predictive data 168 includes an indication of aging or drift in some components, such as manufacturing equipment 124, sensors 126, or measurement equipment 128. In some embodiments, predictive data 168 includes an indication of end-of-life for a component, such as manufacturing equipment 124, sensors 126, or measurement equipment 128.

[0029] Performing a manufacturing process that results in a defective product can be costly in terms of time, energy, product, components, manufacturing equipment 124, costs of identifying and discarding the defective product, etc. By inputting sensor assembly data 152 (e.g., measurements of properties of the substrate support) into a physics-based digital twin model (e.g., model 190), receiving output of substrate support data 154, and performing corrective actions based on the substrate support data 154, system 100 can have the technical advantage of avoiding the costs of producing, identifying, and discarding defective product.

[0030] Running a manufacturing process that results in a failure of a component of manufacturing equipment 124 can be costly in terms of downtime, product damage, equipment damage, rush orders for replacement components, etc. By inputting sensor assembly data 152 (e.g., measurements of properties of the substrate support) into a physics-based digital twin model (e.g., model 190), receiving the output of substrate support data 154, and comparing substrate support datasets from the same substrate support over time to diagnose drifting or failing components (recorded as predictive data 168), and performing corrective actions (e.g., predicted operational maintenance such as component replacement, processing, cleaning, etc.) based on the predictive data 168, system 100 can have the technical advantage of avoiding the costs of one or more of unexpected component failures, unscheduled downtime, lost production, unexpected equipment failures, product scrap, etc. Monitoring the performance of components, e.g., manufacturing equipment 124, sensors 126, metrology devices 128, etc., over time can indicate signs of component degradation. By monitoring the performance of the substrate support over time, for example, if measurements indicate that the substrate support may still perform well (e.g., above a threshold) for some time (e.g., until the next planned maintenance event) after a standard replacement interval has elapsed, the operational life of the substrate support can be extended.

[0031] The manufacturing parameters may not be optimal for producing the product, which can result in costly consequences such as increased consumption of resources (e.g., energy, coolant, gas, etc.), increased time to produce the product, increased component failures, increased quantity of defective products, etc. By inputting the sensor assembly data 152 into a trained physics-based model (e.g., model 190), receiving the output of the substrate support data 154, and performing corrective action (e.g., based on the substrate support data 154) to update the manufacturing parameters (e.g., set optimal manufacturing parameters), the system 100 may have the technical advantage of using optimal manufacturing parameters (e.g., hardware parameters, process parameters, optimal design) to avoid the costly consequences of non-optimal manufacturing parameters.

[0032] The corrective action may be associated with one or more of computational process control (CPC), statistical process control (SPC) (e.g., SPC of electronic components to determine the process under control, SPC to predict the useful life of components, SPC to compare with 3 sigma graphs, etc.), advanced process control (APC), model-based process control, preventative operational maintenance, design optimization, manufacturing parameter updates, manufacturing recipe updates, feedback control, machine learning corrections, etc.

[0033] In some embodiments, the corrective action includes providing a warning (e.g., an alarm that stops or prevents a manufacturing process from running if the predictive data 168 indicates a predicted anomaly, such as an anomaly in a product, component, or manufacturing equipment 124). In some embodiments, the corrective action includes providing feedback control (e.g., modifying a manufacturing parameter in response to the predictive data 168 indicating an anomaly). In some embodiments, the corrective action includes providing machine learning (e.g., modifying one or more manufacturing parameters based on the predictive data 168). In some embodiments, performing the corrective action includes causing an update of one or more manufacturing parameters. In some embodiments, performing the corrective action includes causing one or more calibration tables and / or equipment constants to be updated (e.g., a set point provided to a component may be adjusted by a value across several process recipes; e.g., the voltage applied to a heater may be increased by 3% for all processes that use the heater).

[0034] The manufacturing parameters may include hardware parameters (e.g., replacing a component, using a specific component, replacing a processing chip, updating firmware, etc.) and / or process parameters (e.g., temperature, pressure, flow rate, speed, current, voltage, gas flow, lift speed, etc.). In some embodiments, the corrective action includes triggering preventative operational maintenance (e.g., replacing, treating, cleaning, etc., of a manufacturing tool 124 component). In some embodiments, the corrective action includes triggering design optimization (e.g., updating manufacturing parameters, manufacturing process, manufacturing tool 124, etc., for an optimized product). In some embodiments, the corrective action includes updating a strategy (e.g., placing the manufacturing tool 124 in idle mode, sleep mode, warm-up mode, etc.).

[0035] Prediction server 112, server machine 170, and server machine 180 may each include one or more computing devices such as a rack-mounted server, a router computer, a server computer, a personal computer, a mainframe computer, a laptop computer, a tablet computer, a desktop computer, a graphics processing unit (GPU), an accelerator application-specific integrated circuit (ASIC) (e.g., a tensor processing unit (TPU)), etc.

[0036] The prediction server 112 may include a prediction component 114. The prediction component 114 may be used to produce prediction data 168. In some embodiments, the prediction component 114 may receive sensor assembly data 152 and / or manufacturing parameters 150 (e.g., received from the client device 120 and retrieved from the data store 140) and generate outputs for performing corrective actions associated with the manufacturing equipment 124 based on the current data. In some embodiments, the prediction component 114 may use one or more models 190 to determine outputs for performing corrective actions based on the current data. The models 190 may be a single model, an ensemble model, or a collection of models used to process data. The models 190 may include one or more physics-based digital twin models, supervised machine learning models, unsupervised machine learning models, semi-supervised machine learning models, statistical models, etc.

[0037] In some embodiments, data indicative of characteristics of substrates produced using the substrate support (e.g., predicted data 168 based on information in substrate support data 154) is provided to a trained machine learning model (e.g., model 190). The machine learning model is trained to output data indicative of corrective actions to produce substrates having different characteristics. In some embodiments, data indicative of predicted characteristics of substrates produced using the substrate support and metrology data for substrates produced with the substrate support are provided as input to the trained machine learning model (e.g., model 190). The trained machine learning model predicts root causes of differences between the predicted data and the measured data (e.g., manufacturing faults, component aging or drift, etc.).

[0038] Historical sensor data and / or historical sensor assembly data may be used in combination with current sensor data and current sensor assembly data to detect drift, changes, aging, etc., in components of the manufacturing equipment 124. Sensor assembly data 152 monitored over time can generate information indicative of changes in the substrate support. Sensor assembly data 152 can also provide information about other components of the manufacturing equipment 124; for example, uneven wear on the surface of the substrate support may indicate a problem with a robot handler that places substrates on the substrate support. The prediction component 114 can use a combination and comparison of these data types to generate prediction data 168. In some embodiments, prediction data 168 includes data that predicts the lifespan of components such as the manufacturing equipment 124, the sensors 126, etc.

[0039] Typically, conventional systems perform little or no characterization of the substrate supports. Classifying substrate supports as satisfactory or unsatisfactory may be done in response to measurements of produced substrates to determine whether the product is satisfactory or unsatisfactory. Characterizing support-to-support variations (e.g., differences in response to supply voltage) may also be done empirically based on measurements taken from produced substrates. Characterizing the substrate supports allows the process of substrate support selection and tuning of substrate support usage (e.g., optimal voltages to apply to various electrodes) to be done without the waste associated with producing and measuring substrates.

[0040] In some embodiments, the prediction component 114 can receive data, such as sensor assembly data 152, metrology data 160, and substrate support data 154, and perform preprocessing, such as extracting patterns in the data or combining the data into new composite data. The prediction component 114 can then provide the data as input to the model 190. The model 190 can include a physics-based digital twin model that accepts the sensor assembly data as input. The model may include a trained machine learning model, a statistical model, or the like, configured to further process data associated with characteristics of the substrate support. The prediction component 114 can receive predictive data from the model 190 indicative of substrate support performance, predicted substrate characteristics, manufacturing faults, component drift, or the like. The prediction component 114 can then cause corrective action to be taken. The corrective action can include sending an alert to the client device 120. The corrective action can also include updating manufacturing parameters of the manufacturing tool 124. The corrective action can also include generating predictive data 168 indicative of chamber or instrument drift, aging, or failure.

[0041] The data store 140 may be a memory (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or another type of component or device capable of storing data. The data store 140 may include multiple storage components (e.g., multiple drives or multiple databases) that may span multiple computing devices (e.g., multiple server computers). The data store 140 may store sensor data 142, manufacturing parameters 150, sensor assembly data 152, substrate support data 154, metrology data 160, and prediction data 168. The sensor data may include sensor data time traces over the duration of a manufacturing process, associations between physical sensors and data, preprocessed data such as averages or composite data, and data indicative of sensor performance over time (i.e., over many manufacturing processes). The manufacturing parameters 150 and metrology data 160 may include similar characteristics. The sensor assembly data 152 includes measurements taken by the sensor assembly 195 used to characterize the substrate support. Sensor assembly data 152 may include data from many types of sensors associated with many characteristics, such as capacitive sensors (e.g., for determining the depth of electrodes below the surface of the substrate support), temperature sensors (e.g., for measuring the temperature of the surface of the substrate support), surface profilometers, etc. Substrate support data 154 includes data generated by a physics-based digital twin model taking into account measurement data from sensor assembly 195 (e.g., sensor assembly data 152). Substrate support data 154 provides a characterization of the performance of the substrate support. Sensor assembly data, substrate support data, and metrology data may include historical data (e.g., at least a portion for training various models represented in FIG. 1 by model 190). Metrology data 160 may be metrology data of produced substrates, as well as sensor data, manufacturing data, and model data corresponding to those products. Metrology data 160 may be utilized to design processes for fabricating additional substrates.The predictive data 168 may include predictions of metrology data resulting from operation of the substrate support, predictions of component drift, aging, or failure, predictions of component life, etc. The predictive data 168 may also include data indicative of aging and failure of components of the system 100 over time.

[0042] In some embodiments, the prediction system 110 further includes a server machine 170 and a server machine 180. The server machine 170 includes a dataset generator 172 that can generate datasets (e.g., a set of data inputs and a set of target outputs) for training, validating, and / or testing the model 190. Some operations of the dataset generator 172 are described in more detail below with respect to FIGS. 2 and 4A . In some embodiments, the dataset generator 172 can divide historical data (e.g., historical sensor assembly data, historical metrology data, etc.) and physical model data (e.g., substrate support data 162) into a training set (e.g., 60 percent of the data), a validation set (e.g., 20 percent of the data), and a test set (e.g., 20 percent of the data). In some embodiments, the prediction system 110 generates multiple feature sets (e.g., via the prediction component 114). For example, a first feature set may correspond to a first type of sensor dataset (e.g., from the first sensor set, a first combination of values ​​from the first sensor set, a first pattern of values ​​from the first sensor set) corresponding to each dataset (e.g., training set, validation set, and test set), and a second feature set may correspond to a second type of sensor dataset (e.g., from a second sensor set different from the first sensor set, a second combination of values ​​different from the first combination, a second pattern different from the first pattern) corresponding to each dataset.

[0043] Server machine 180 includes a training engine 182, a validation engine 184, a selection engine 185, and / or a test engine 186. The engines (e.g., training engine 182, validation engine 184, selection engine 185, and test engine 186) may refer to hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing device, etc.), software (e.g., instructions executing on a processing device, general-purpose computer system, or dedicated machine), firmware, microcode, or a combination thereof. Training engine 182 may be capable of training a model 190 using one or more feature sets associated with a training set from dataset generator 172. Training engine 182 may generate multiple trained models 190, each trained model 190 corresponding to a different feature set of the training set (e.g., sensor data from a different sensor set). For example, a first trained machine learning model may have been trained using all features (e.g., X1-X5), a second trained machine learning model may have been trained using a first subset of features (e.g., X1, X2, X4), and a third trained machine learning model may have been trained using a second subset of features (e.g., X1, X3, X4, and X5) that may partially overlap with the first subset of features. Dataset generator 172 may receive the output of the trained models (e.g., 190), compile the data into training, validation, and test datasets, and use the datasets to train a second model. Some or all of the operations of server machine 180 may be used to train various types of models, including physics-based models, supervised machine learning models, unsupervised machine learning models, etc.

[0044] The validation engine 184 may be able to validate the trained models 190 using the corresponding feature sets of the validation set from the dataset generator 172. For example, a first trained model 190 trained using a first feature set of the training set may be validated using the first feature set of the validation set. The validation engine 184 may determine the accuracy of each of the trained models 190 based on the corresponding feature sets of the validation set. The validation engine 184 may discard trained models 190 whose accuracy does not meet a threshold accuracy. In some embodiments, the selection engine 185 may be able to select one or more trained models 190 whose accuracy meets a threshold accuracy. In some embodiments, the selection engine 185 may be able to select the trained model 190 with the highest accuracy among the trained models 190.

[0045] The testing engine 186 may be able to test the trained models 190 using the corresponding feature set of the test set from the dataset generator 172. For example, a first trained model 190 trained using a first feature set of the training set may be tested using the first feature set of the test set. The testing engine 186 may determine the most accurate trained model 190 of all the trained models based on the test set.

[0046] Model 190 may refer to a physics-based digital twin model of the substrate support. The physics-based model is configured to solve equations that describe the flow of heat, energy, etc. in and around the substrate support. Measurements performed by sensor assembly 195 inform the parameters of the physics-based model. The physics-based model may be improved by further training, for example, by applying various voltages to the heater electrodes to measure the steady-state temperature of the surface of the substrate support and feeding that data into the physics-based model to further refine the operation of the physics-based model.

[0047] Model 190 may refer to a machine learning model, which may be a model artifact created by training engine 182 using a training set that includes data inputs and corresponding target outputs (ground truths for each training input). Patterns that map data inputs to target outputs (ground truths) may be found in the data set, and machine learning model 190 is provided with a mapping that captures these patterns. In some embodiments, machine learning model 190 may predict substrate properties. In some embodiments, machine learning model 190 may predict failure modes of fabrication chamber components.

[0048] The prediction component 114 can provide input data to a trained machine learning model 190 and can run the trained machine learning model 190 on the input to obtain one or more outputs. The prediction component 114 may be able to determine (e.g., extract) prediction data 168 from the output of the trained machine learning model 190 and can determine (e.g., extract) confidence data from the output indicating a confidence level that the prediction data 168 is an accurate predictor of a produced product or process associated with the input data for the product to be produced, or an accurate predictor of a component of the manufacturing equipment 124. The prediction component 114 may be able to determine the prediction data 168 based on the output of the model 190, including predictions regarding finished substrate properties and predictions of the useful life of components of the manufacturing equipment 124, sensors 126, or metrology equipment 128. The prediction component 114 or the corrective action component 122 can use the confidence data to determine whether to trigger a corrective action associated with the manufacturing equipment 124 based on the prediction data 168.

[0049] The confidence data may include or indicate a confidence level. As an example, the predicted data 168 may indicate characteristics of a finished wafer given a set of manufacturing inputs, including the use of a substrate support described using the substrate support data 154. The confidence data may indicate that the predicted data 168 is an accurate prediction of a product associated with at least a portion of the input data. In one example, the confidence level is a real number between 0 and 1, inclusive, where 0 indicates no confidence that the predicted data 168 is an accurate prediction of a product processed according to the input data and 1 indicates absolute confidence that the predicted data 168 will accurately predict the characteristics of a product processed according to the input data. In response to confidence data indicating a confidence level below a threshold level for a given number of instances (e.g., a percentage of instances, a frequency of instances, a total number of instances, etc.), the prediction component 116 may retrain the trained machine learning model 190 (e.g., based on current sensor data 146, current manufacturing parameters 150, etc.).

[0050] For purposes of illustration and not limitation, embodiments of the present disclosure describe using historical data to train one or more models 190 and inputting current data into the one or more trained models 190 to determine predicted data 168. Other implementations use heuristic or rule-based models (e.g., without using trained machine learning models) to determine predicted data. The prediction component 114 can monitor historical data and measurement data 160. Any of the information described with respect to data input 210 in FIG. 2 can be monitored or otherwise used in the heuristic or rule-based models.

[0051] In some embodiments, the functionality of client device 120, prediction server 112, server machine 170, and server machine 180 may be provided by fewer machines. For example, in some embodiments, server machines 170 and 180 may be combined into a single machine, while in some other embodiments, server machine 170, server machine 180, and prediction server 112 may be combined into a single machine. In some embodiments, client device 120 and prediction server 112 may be combined into a single machine.

[0052] In general, functions described in one embodiment as being performed by client device 120, prediction server 112, server machine 170, and server machine 180 may also be performed on prediction server 112 in other embodiments, where appropriate. In addition, functions attributed to particular components may be performed by different components or multiple components operating together. For example, in some embodiments, prediction server 112 may determine corrective actions based on prediction data 168. In another example, client device 120 may determine prediction data 168 based on output from a trained machine learning model or a physics-based digital twin model.

[0053] Additionally, the functionality of a particular component may be performed by different components or multiple components working together. One or more of prediction server 112, server machine 170, or server machine 180 may be accessed as a service offered to other systems or devices via an appropriate application programming interface (API).

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

[0055] Embodiments of the present disclosure may be applied to data quality assessment, feature enhancement, model evaluation, virtual metrology (VM), predictive maintenance (PdM), marginal optimization, and the like.

[0056] Although embodiments of the present disclosure are discussed in terms of generating predictive data 168 to perform corrective actions in a manufacturing facility (e.g., a semiconductor manufacturing facility), the embodiments may also be applied generally to improving data processing by utilizing physics-based digital twin models and sensor assemblies to characterize substrate supports.

[0057] FIG. 2 is a block diagram of an example dataset generator 272 (e.g., dataset generator 172 of FIG. 1 ) used to create a dataset for a model (e.g., model 190 of FIG. 1 ), according to some embodiments. The dataset generator 272 may be part of the server machine 170 of FIG. 1 . In some embodiments, the system 100 of FIG. 1 includes multiple models. In such cases, each model may have a separate dataset generator, or the models may share a dataset generator. FIG. 2 illustrates a dataset generator associated with a machine learning model configured to take as input predicted performance data for substrates produced using a substrate support and measured performance data (e.g., metrology data) for substrates produced using the same substrate support, and to provide output information indicative of corrective actions associated with manufacturing faults, component failures, component drift, etc. that cause differences between the predicted and measured performance data.

[0058] 2 , a system 200 including a dataset generator 272 (e.g., dataset generator 172 of FIG. 1 ) creates a dataset for a machine learning model (e.g., model 190 of FIG. 1 ). The dataset generator 272 may create the dataset using data extracted as output from a physics-based digital twin model. In some embodiments, the dataset generator 272 creates training inputs from predicted performance data of substrates produced using the substrate support and measured performance data of substrates produced using the substrate support. The dataset generator 272 also generates target outputs 220 for training the machine learning model. The target outputs include data indicative of root causes of differences between the predicted and measured performance data, e.g., corrective actions to address manufacturing failures. The training input data 210 and the target output data 220 are provided to the machine learning model.

[0059] It is within the scope of this disclosure to express the training inputs 210 and target outputs 220 in a variety of different ways: a two-dimensional map of substrate properties, a function that replicates the map, or other data indicative of substrate performance data can be used as the training inputs 210, with similar flexibility for the target outputs 220.

[0060] In some embodiments, the dataset generator 272 generates a dataset (e.g., a training set, a validation set, a test set) that includes one or more data inputs 210 (e.g., training inputs, validation inputs, test inputs) and may include one or more target outputs 220 corresponding to the data inputs 210. The dataset may also include mapping data that maps the data inputs 210 to the target outputs 220. The data inputs 210 may also be referred to as “features,” “attributes,” or “information.” In some embodiments, the dataset generator 272 may provide a dataset to the training engine 182, the validation engine 184, or the test engine 186 of FIG. 1, where the dataset is used to train, validate, or test the machine learning model 190 of FIG. 1. Some embodiments of generating a training set may be further described with respect to FIG. 4A .

[0061] In some embodiments, the dataset generator 272 may generate a first data input corresponding to the first predicted performance dataset 244A and the first measured performance dataset 252A for training, validating, or testing a first machine learning model, and the dataset generator 272 may generate a second data input corresponding to the second predicted performance dataset 244A and the second measured performance dataset 252B for training, validating, or testing a second machine learning model.

[0062] In some embodiments, the dataset generator 272 can perform operations on one or more of the data inputs 210 and the target outputs 220. The dataset generator 272 can extract patterns from the data (slope, curvature, etc.), combine the data (average, feature generation, etc.), or separate the data into groups (e.g., train a model on a subset of the predicted performance data) and use the groups to train separate models.

[0063] The data inputs 210 and target outputs 220 for training, validating, or testing a machine learning model may include information about a particular substrate support (e.g., a particular substrate chuck assembly). The data inputs 210 and target outputs 220 may include information about a particular substrate support design (e.g., to be used for all substrate supports of that design). The data inputs 210 and target outputs 220 may include information about a particular type of process, target substrate design, target substrate characteristics, or may be grouped together in another manner.

[0064] In some embodiments, the dataset generator 272 can generate a set of target outputs 220 that include instructions for corrective actions 230. The target outputs 220 may be divided into sets that correspond to sets of input data. Different sets of target outputs 220 may be used in conjunction with similarly defined sets of data inputs 210, including training different models, using different sets for training, validation, and testing, etc.

[0065] The target output 220 may be generated by correlating trends in performance data to appropriate corrective actions using methods other than machine learning. A user may indicate that taking a particular corrective action eliminated a difference between past predicted and measured performance, a manufacturing fault may be intentionally introduced to generate data useful for training, etc. In some embodiments, a model may be trained without a target output 220 (e.g., an unsupervised model or a semi-supervised model). A model trained without being provided with a target output may, for example, be trained to recognize significant differences (e.g., outside an error threshold) between predicted and measured performance data.

[0066] In some embodiments, the information used to train the machine learning model may be from a particular type of manufacturing equipment (e.g., manufacturing equipment 124 in FIG. 1 ) in a manufacturing facility having particular characteristics, allowing the trained machine learning model to determine an outcome for a particular group of manufacturing equipment 124 based on input of predicted and measured performance data associated with one or more components that share the characteristics of the particular group. In some embodiments, the information used to train the machine learning model may be for components from more than one manufacturing facility, allowing the trained machine learning model to determine an outcome for a component based on input from one manufacturing facility.

[0067] In some embodiments, after generating a dataset and using the dataset to train, validate, or test a machine learning model, the machine learning model may be further trained, validated, or tested, or tuned.

[0068] FIG. 3 is a block diagram illustrating a system 300 for generating output data (e.g., predicted data 168 of FIG. 1 ), according to some embodiments. System 300 can be used to analyze differences between predicted and measured performance data for a substrate and provide predictive data that accounts for the differences and indicates corrective actions. Systems similar to system 300 may also be used for other models, such as machine learning models that correlate substrate support characterizations with strategy inputs to predict substrate properties. Some or all of the operations of system 300 can be used to generate data indicative of substrate support characterization via a physics-based digital twin model. In these cases, other data may be used as inputs by system 300 and, if desired, generated as outputs.

[0069] 3, in block 310, system 300 (e.g., a component of prediction system 110 of FIG. 1) performs data partitioning (e.g., via dataset generator 172 of server machine 170 of FIG. 1) of historical data 364 (e.g., historical predicted and measured performance data for substrates, and past corrective actions taken) to generate training set 302, validation set 304, and test set 306. For example, the training set may be 60% of the performance data, the validation set may be 20% of the performance data, and the test set may be 20% of the performance data.

[0070] At block 312, the system 300A performs model training using the training set 302 (e.g., via the training engine 182 of FIG. 1 ). The system 300 may train one model or multiple models using multiple feature sets of the training set 302 (e.g., a first feature set including a subset of the performance data of the training set 302, a second feature set including a different subset of the performance data of the training set 302, etc.). For example, the system 300 may train machine learning models to generate a first trained machine learning model using the first feature set of the training set and a second trained machine learning model using the second feature set of the training set (e.g., data different from the data used to train the first machine learning model). In some embodiments, the first trained machine learning model and the second trained machine learning model may be combined to generate a third trained machine learning model (e.g., which may be a better predictor than either the first or second trained machine learning model alone). In some embodiments, the sets of features used to compare models may overlap (e.g., one model may be trained with performance data indicative of film thickness, another model may be trained with performance data indicative of both film thickness and film stress, different models may be trained with data from different locations on the substrate, etc.) In some embodiments, hundreds of models may be generated, including models with various feature permutations and model combinations.

[0071] At block 314, the system 300 performs model validation using the validation set 304 (e.g., via the validation engine 184 of FIG. 1 ). The system 300 may validate each of the trained models using the corresponding feature set of the validation set 304. For example, the validation set 304 may use the same subset of performance data used in the training set 302, but with different input conditions. In some embodiments, the system 300A may validate hundreds of models (e.g., models with various permutations of features, combinations of models, etc.) generated at block 312. At block 314, the system 300 may determine the accuracy of each of the one or more trained models (e.g., via model validation) and may determine whether one or more of the trained models have an accuracy that meets a threshold accuracy. In response to determining that none of the trained models have an accuracy that meets the threshold accuracy, flow returns to block 312, and the system 300 performs model training using a different feature set from the training set. In response to determining that one or more of the trained models have an accuracy that meets the threshold accuracy, flow proceeds to block 316. The system 300 may discard trained machine learning models that have an accuracy below the threshold accuracy (e.g., based on a validation set).

[0072] In block 316, the system 300 may perform model selection (e.g., via the selection engine 185 of FIG. 1 ) to determine which of the one or more trained models that meet the threshold accuracy has the highest accuracy (e.g., the selected model 308 based on the validation of block 314). If only a single model is trained, the operations of block 316 may be skipped. In response to determining that two or more of the trained models that meet the threshold accuracy have the same accuracy, flow may return to block 312, where the system 300 performs model training using a further refined training set corresponding to the further refined feature set to determine the trained model with the highest accuracy.

[0073] In block 318, the system 300 performs model testing (e.g., via the test engine 186 of FIG. 1 ) using the test set 306 to test the selected model 308. The system 300 may test the first trained machine learning model using the first feature set of the test set and determine that the first trained machine learning model meets a threshold accuracy (e.g., based on the first feature set of the test set 306). In response to the accuracy of the selected model 308 not meeting the threshold accuracy (e.g., the selected model 308 overfits the training set 302 and / or the validation set 304 and cannot be applied to other datasets, such as the test set 306), the flow proceeds to block 312, where the system 30A performs model training (e.g., retraining) using a different training set, possibly corresponding to a different feature set, or a reorganization of the board divided into the training set, validation set, and test set. In response to determining that the selected model 308 has an accuracy that meets the threshold accuracy based on the test set 306, the flow proceeds to block 320. At least in block 312, the model can learn patterns in the simulated sensor data to make predictions, and in block 318, the system 300 can apply the model to the remaining data (e.g., test set 306) to test the predictions.

[0074] In block 320, the system 300 receives current data 354 (e.g., predicted and measured performance data for substrates not included in historical data 364) using the trained model (e.g., selected model 308), determines (e.g., extracts) predicted data 368 (e.g., predicted data 168 of FIG. 1) from the output of the trained model, and performs an action (e.g., performs corrective action associated with manufacturing equipment 124 of FIG. 1, provides and alerts client device 120 of FIG. 1, etc.).

[0075] In some embodiments, retraining of a machine learning model is performed by supplying additional data to further train the model. Current data 354 may be provided in block 312. Additional corrective action data 346 may be provided as well. This data may differ from the data originally used to train the model by incorporating input parameter combinations that were not part of the original training, input parameters outside the parameter space covered by the original training, or may be updated to reflect chamber-specific knowledge (e.g., variations from an ideal chamber due to manufacturing tolerance ranges, aging components, etc.). The selected model 308 may be retrained based on this data.

[0076] In some embodiments, one or more of acts 310-320 may be performed in various orders and / or with other acts not presented and described herein. In some embodiments, one or more of acts 310-320 may not be performed. For example, in some embodiments, one or more of data partitioning of block 310, model validation of block 314, model selection of block 316, or model testing of block 318 may not be performed. For example, a subset of these operations may be performed in training a physics-based digital twin model to take measurements of the substrate support as input and generate predicted performance data for the substrate support as output.

[0077] 4A-4E are flowcharts of methods 400A-E associated with characterizing a substrate support and triggering corrective action according to certain embodiments. Methods 400A-E may be performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, a 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 400A-E may be performed in part by prediction system 110. Method 400A may be performed in part by prediction system 110 (e.g., server machine 170 and dataset generator 172 in FIG. 1 , dataset generator 272 in FIG. 2 ). Prediction system 110 may use method 400A to generate a dataset for at least one of training, validating, or testing a model according to embodiments of the present disclosure. The model may be a physics-based digital twin model (e.g., for generating predicted performance data for a substrate support), a machine learning model (e.g., for generating predicted performance data for a wafer, for generating data indicative of corrective actions associated with a component of manufacturing equipment, etc.), a statistical model, or another model trained to receive inputs related to characterization of a substrate support and generate outputs. Methods 400B-C may be performed by prediction server 112 (e.g., prediction component 114, etc.). Method 400D may be performed by server machine 180 (e.g., training engine 182). Method 400E may be performed by prediction server 112 (e.g., prediction component 114), client device 120 (e.g., corrective action component 122), etc. In some embodiments, a non-transitory storage medium stores instructions that, when executed by a processing device (e.g., prediction system 110, server machine 180, prediction server 112, etc.), cause the processing device to perform one or more of methods 400A-E.

[0078] For ease of explanation, methods 400A-E are shown and described as a series of operations. However, operations in accordance with the present disclosure may occur in various orders and / or simultaneously, as well as with other operations not shown and described herein. Moreover, not all illustrated operations may be performed to implement methods 400A-E in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that methods 400A-E may alternatively be represented as a series of interrelated states via a state diagram or events.

[0079] FIG. 4A is a flow diagram of a method 400A for generating a dataset for a machine learning model for generating predicted data (eg, predicted data 168 of FIG. 1) according to a particular embodiment.

[0080] Referring to FIG. 4A, in some embodiments, at block 401, processing logic performing method 400A initializes a training set T to an empty set.

[0081] At block 402, processing logic generates first data inputs (e.g., first training inputs, first validation inputs) that may include sensor assembly data, substrate support performance data, measured substrate performance data, substrate metrology data (e.g., film properties such as thickness, material composition, optical properties, roughness, etc.), etc. In some embodiments, the first data inputs may include a first set of features related to the type of data, and the second data inputs may include a second set of features related to the type of data (e.g., as described with respect to FIG. 3 ).

[0082] At block 403, processing logic generates a first target output for one or more of the data inputs (e.g., a first data input). In some embodiments, the first target output is performance data for the substrate. In some embodiments, the first target output is data indicative of a corrective action. In some embodiments, no target output is generated (e.g., for training an unsupervised machine learning model).

[0083] At block 404, processing logic optionally generates mapping data indicating an input-output mapping. An input-output mapping (or mapping data) may refer to data inputs (e.g., one or more of the data inputs described herein), target outputs for the data inputs, and associations between the data inputs and the target outputs. In some embodiments (e.g., without target output data), these operations may not be performed.

[0084] At block 405, processing logic, in some embodiments, adds the mapping data generated at block 404 to dataset T.

[0085] At block 406, processing logic branches based on whether dataset T is sufficient for at least one of training, validation, and / or testing of model 190 of FIG. 1. If so, execution proceeds to block 407; if not, execution continues back to block 402. Note that while in some embodiments the sufficiency of dataset T may be determined simply based on the number of inputs in the dataset, which in some embodiments are mapped to outputs, in some other implementations the sufficiency of dataset T may be determined based on one or more other criteria in addition to or instead of the number of inputs (e.g., a measure of diversity of the data examples, accuracy, etc.).

[0086] At block 407, processing logic provides dataset T (e.g., to server machine 180 of FIG. 1 ) to train, validate, and / or test model 190. In some embodiments, dataset T is a training set and is provided to training engine 182 of server machine 180 to perform training. In some embodiments, dataset T is a validation set and is provided to validation engine 184 of server machine 180 to perform validation. In some embodiments, dataset T is a test set and is provided to test engine 186 of server machine 180 to perform testing.

[0087] 4B illustrates a method 400B for characterizing a substrate support, according to some embodiments. In block 410 of method 400B, measurements of the substrate support are performed. A processing device different from the processing device (or devices) that performs the other operations of method 400B may perform the operation of block 410. In preparation for performing the operation of block 410, a sensor assembly including one or more sensors is positioned proximate to the substrate support to be characterized. In some embodiments, the area of ​​the substrate support may be larger than the area within the field of view of the sensors of the sensor assembly. The sensor assembly and substrate support may be positioned on a movable support such that the sensor assembly and substrate support are proximate and relative motion between them can be induced (e.g., the assembly and / or the substrate support are mounted on a stage that is movable in two or three spatial dimensions). The sensor assembly is then positioned at several different positions relative to the substrate support. One or more measurements are then generated by the sensors of the sensor assembly at each position. By inducing relative motion in at least two dimensions, the entire surface of the substrate support can be measured. In some embodiments, smaller areas of the surface of the substrate support (e.g., suspected problem areas) may be measured. The method can be used to generate a two-dimensional map of measurements of the surface of the substrate support. The measurements may include electrode depth below the surface of the substrate, surface profile, surface temperature, etc. The measurements can indicate properties of the substrate support.

[0088] At block 412, processing logic receives data indicative of a characteristic of the substrate support from a sensor of the removable sensor assembly. The received data may be measurement data paired with position data, a two-dimensional map of the characteristic, or may be represented in another form. The data may include data from different types of sensors including capacitance sensors, temperature sensors, surface profilometers, etc. The data may be processed in some manner, such as by removing outliers, statistical processing (e.g., boxcar averaging, smoothing), etc., as corresponding to this or other operations of method 400B.

[0089] At block 414, processing logic provides data based on the data received at block 412 (e.g., data indicative of properties of the substrate support) to a physics-based digital twin model of the substrate support. The physics-based model may be configured to receive data indicative of measurements of properties of the substrate support. The physics-based model may be configured to provide solutions to equations, such as equations describing heat transfer, energy balance, etc. The physics-based model may be configured to output predicted performance data for the substrate support.

[0090] At block 416, processing logic receives predicted performance data for the substrate support from a physics-based digital twin model of the substrate support. The physics-based digital twin may have been previously trained using historical data. The predicted performance data may predict conditions at a location of a substrate processed using the substrate support. In some embodiments, the predicted performance data for the substrate support includes a prediction of a temperature profile at the location of the substrate. The predicted performance data for the substrate support may include a prediction of a temperature profile (e.g., a two-dimensional temperature map) at the location of the substrate for a particular set of inputs (e.g., power supplied to various heater electrodes included in the substrate support). In some embodiments, the predicted performance data for the substrate support may include predictions of properties other than temperature, such as chucking force, RF power, heat transfer, etc. These properties may also depend on input values ​​(e.g., power supplied to the chuck electrodes, etc.). The predicted performance data for the substrate support may be further used to obtain predicted performance data (e.g., predicted metrology data) for a substrate processed using the substrate support, as described in more detail in connection with FIG. 4C . The predicted performance data for the substrate may be provided by a trained machine learning model.

[0091] At block 418, processing logic may receive second predicted performance data for the substrate support. The second predicted performance data is based on second sensor assembly data collected at a different time (e.g., a later time) than the first sensor assembly data received at block 412. The second sensor assembly data may be generated after processing several substrates using the substrate support; for example, the second sensor assembly data may be used to quantify the impact of processing several substrates on the substrate support. Processing logic may perform many of the operations included in method 400B to generate the second predicted performance data for the substrate support, as described above in connection with generating the first predicted performance data for the substrate support. Processing logic may receive second data indicative of predicted performance of the substrate support, generated by the trained physics-based model taking into account the second sensor assembly data.

[0092] In block 419, processing logic may cause corrective action to be taken in consideration of second predicted performance data of the substrate support. Processing logic may determine at least one difference between first data indicative of predicted performance data of the substrate support and second data indicative of predicted performance data of the substrate support. In some embodiments, processing logic may compare first sensor assembly data indicative of a characteristic of the substrate support with second sensor assembly data indicative of a characteristic of the substrate support. Corrective action may be taken in consideration of the at least one difference, such as scheduling replacement or maintenance of a deteriorating component of the fabrication chamber. Measurements of the substrate support taken over time may indicate drift, failure, lifetime estimates, etc. of the substrate support. Measurements of the substrate support taken over time may also provide information about other components of the fabrication chamber. Uneven wear on the surface of the substrate support may indicate, for example, a problem with a robot handle that places and removes substrates from the substrate support. The operations of blocks 410, 418, and 419 of method 400B may be separated from the operations of the other blocks of method 400B. For example, the operations of blocks 410, 418, and / or 419 may be performed at different times, may be performed by different processing devices, may not be included in method 400B, etc., as indicated by the dashed boxes in FIG. 4B.

[0093] 4C is a flow diagram of a method 400C for performing corrective action, according to some embodiments. Method 400C may be performed by processing logic after performance of a method for characterizing a substrate support, such as method 400B. The processing logic may receive data indicative of properties of the substrate support from sensors of a sensor assembly. The processing logic may provide the data indicative of properties of the substrate support to a physics-based digital twin model of the substrate support.

[0094] At block 420, processing logic receives predicted performance data for the substrate support from the physics-based model.

[0095] At block 422, processing logic determines predicted properties of substrates produced using the substrate support given the predicted performance data of the substrate support. In some embodiments, the predicted properties of the substrate are generated as output from a trained machine learning model. In some embodiments, the predicted properties of the substrate take the form of a two-dimensional map of the properties of the substrate. In some embodiments, the two-dimensional map may be a map of film thickness, etch rate, film stress, or other property of the substrate.

[0096] At block 424, the processing logic causes corrective actions to be performed taking into account the predicted characteristics of the substrate to be processed using the substrate support. The determination of the corrective actions to be performed may take into account the output of the trained machine learning model. The corrective actions may include one or more of sending an alert to a user, scheduling corrective maintenance, scheduling preventive maintenance, updating a process recipe, updating calibration tables or equipment constants (e.g., parameters associated with one or more process recipes), correcting chamber drift, etc. Chamber drift includes gradual changes in the processing equipment. Such changes may include gradual changes in the surfaces of chamber components caused by repeated processing of products, changes in sensors and / or sensor calibration, changes in electronics including control electronics, etc. Chamber drift may include the addition of material to a surface (e.g., via deposition), the removal of material from a surface (e.g., via etching), etc. Changes in the surfaces of components of the manufacturing equipment may affect processing conditions, e.g., processing kinetics.

[0097] 4D is a method 400D for training a machine learning model to identify corrective actions, according to some embodiments. The trained machine learning model can be used in connection with the present disclosure in many ways. Methods 400D-E describe methods associated with a machine learning model that receives metrology data for a substrate and predicted metrology data for the substrate as input, determines a difference between the two data, and provides data indicative of a corrective action as output. Other machine learning models are also within the scope of the present disclosure, such as a model that takes predicted performance data for a substrate support as input and produces predicted substrate properties as output. The machine learning model may also be used in different ways; for example, a model may take metrology data for a substrate and predicted metrology data for the substrate as input (similar to the input for the machine learning model of methods 400D-E) and, as an unsupervised machine learning model, generate an output that indicates a significant difference (e.g., greater than a certain threshold) between the measured data and the predicted data.

[0098] At block 430, processing logic receives metrology data for substrates produced using a manufacturing equipment. The manufacturing equipment includes a substrate support that has been characterized (e.g., characterized using a method such as method 400B). At block 432, processing logic receives data indicative of predicted properties of the substrates produced using the manufacturing data. The data indicative of the predicted properties may have been received as output of a trained machine learning model. The data indicative of the predicted properties of the substrate may have been generated taking into account the characterization of the substrate support output by the trained physics-based digital twin model.

[0099] At block 434, processing logic receives data indicative of corrective actions associated with manufacturing equipment used to produce the substrate. The data indicative of corrective actions may be generated by a corrective action implementer identifying and providing to processing logic corrective actions for improving manufacturing equipment associated with the substrate. The data indicative of corrective actions may be generated by a user intentionally producing substrates in a non-optimal manner to obtain predicted and measured characteristics of the substrates associated with specific corrective actions associated with how the user configured the manufacturing equipment.

[0100] At block 436, processing logic trains the machine learning model by providing data indicative of metrology data for the substrate and data indicative of predicted properties of the substrate as training inputs, and providing data indicative of corrective actions associated with the manufacturing equipment used to produce the substrate as target outputs.

[0101] 4E is a flow diagram of a method 400E for determining corrective actions using a trained machine learning model, according to some embodiments. At block 440, processing logic provides predicted properties of the substrate as input to the trained machine learning model. The predicted properties of the substrate may have been generated by another trained machine learning model. The predicted properties of the substrate (e.g., film thickness, surface roughness, optical or chemical properties, etc.) may have been generated by considering predicted performance characteristics of the substrate support (e.g., a two-dimensional map of predicted temperatures at the substrate for given input conditions) output by a trained physics-based digital twin model of the substrate support.

[0102] At block 442, processing logic provides measured metrology data of the substrate as input to the trained machine learning model. The measured metrology data is associated with substrates processed using the same equipment, processing parameters, etc. as input to the physics-based model used to generate the predicted properties of the substrate. The metrology data can include substrate film thickness, surface roughness, optical properties, chemical composition, etc.

[0103] At block 444, processing logic receives data from the trained machine learning model. This data indicates a corrective action associated with the manufacturing equipment used to produce the substrate. The corrective action may be any of the corrective actions discussed in connection with this disclosure, for example, in connection with FIG. 4D . In some embodiments, processing logic then causes the corrective action to be executed. The trained machine learning model of method 4E may include one or more of a neural network (e.g., an artificial neural network), a support vector machine, a radial basis function, clustering, a k-nearest neighbor algorithm, a random forest, etc. The corrective action may indicate a fault with the substrate support; for example, metrology data indicating a hot spot (e.g., metrology indicating an area where the temperature increased during processing using the substrate support) may indicate a problem with the lift pins of the substrate support, metrology data indicating a temperature gradient may indicate a problem with the robot handler not properly positioning the substrate on the substrate support, etc.

[0104] FIG. 5 illustrates a cross-sectional view of a fabrication chamber 500 (e.g., a semiconductor wafer fabrication chamber) according to some embodiments of the present disclosure. The fabrication chamber 500 may be one or more of an etch chamber, a deposition chamber (including atomic layer deposition, chemical vapor deposition, physical vapor deposition, or plasma-enhanced versions thereof), an anneal chamber, etc. For example, the fabrication chamber 500 may be a chamber for a plasma etching apparatus, a plasma cleaning apparatus, etc. Examples of chamber components may include a substrate support 504, a chuck (e.g., an electrostatic chuck, a vacuum chuck, etc.), a ring (e.g., a process kit ring), chamber walls, a base, a showerhead 506, a gas distribution plate, a liner, a liner kit, a shield, a plasma screen, a flow equalizer, a cooling base, a chamber viewport, a chamber lid, a nozzle, etc.

[0105] In one embodiment, the fabrication chamber 500 includes a chamber body 508 and a showerhead 506 that enclose an interior volume 510. In some chambers, the showerhead 506 may be replaced by a lid and nozzle. The chamber body 508 may be constructed of aluminum, stainless steel, or other suitable material. The chamber body 508 generally includes sidewalls 512 and a bottom 514.

[0106] An exhaust port 516 may be defined in the chamber body 508 and may couple the interior volume 510 to a pumping system 518. The pumping system 518 may include one or more pumps and valves utilized to evacuate and regulate the pressure of the interior volume 510 of the fabrication chamber 500.

[0107] The showerhead 506 may be supported on a sidewall 512 of the chamber body 508 or on top of the chamber body. The showerhead 506 (or, in some embodiments, a lid) can open to provide access to the interior volume 510 of the fabrication chamber 500 and, when closed, provide a seal for the fabrication chamber 500. A gas panel 520 can be coupled to the fabrication chamber 500 to supply process or cleaning gases to the interior volume 510 through the showerhead 506 (or lid and nozzle). The showerhead 506 can include multiple gas supply holes throughout. Examples of process gases that can be used to process substrates in the fabrication chamber 500 include halogen-containing gases such as C2F6, SF6, SiCl4, HBr, NF3, CF4, CHF3, F2, Cl2, CCl4, BCl3, and SiF4, among others, as well as other gases such as O2 or N2O. Examples of carrier gases include N2, He, Ar, and other gases that are inert to the process gases (e.g., non-reactive gases).

[0108] A substrate support 504 is disposed in an interior volume 510 of the fabrication chamber 500 below a showerhead 506. In some embodiments, the substrate support 504 includes a susceptor 522 and a shaft 524. The substrate support 504 supports the substrate 502 during processing. In some embodiments, one or more heaters 526 and reflectors 528 are also disposed within the fabrication chamber 500.

[0109] The substrate support 504 may include electronics that provide power and control to electrodes disposed within the substrate support 504. The substrate support 504 shown in FIG. 5 includes an electrostatic chuck assembly, although substrate supports including other components are possible and within the scope of this disclosure. The substrate support 504 may include one or more clamping electrodes (not shown). The clamping electrodes may be controlled by a chuck power supply 530. The chuck power supply 530 may include individual outputs to each clamping electrode so that the clamping electrodes can be individually controlled. For clarity, only one output of the chuck power supply 530 (and other similarly disposed components) is shown in FIG. 5.

[0110] The substrate support 504 can include one or more heating elements 536 disposed within the support (e.g., the substrate support 504 can include an electrostatic chuck heater assembly). The embedded heating elements can be regulated by a heater power supply 532. The heater power supply 532 can have many of the same characteristics as the chuck power supply 530. As shown in FIG. 5 , the heating elements 536 can be disposed at different depths within the substrate support (e.g., different distances from the surface of the substrate support). Characterizing the depth of the heater elements below the surface of the substrate support informs a physics-based model of the substrate support. Other elements (e.g., chuck electrodes, RF elements, etc.) can also be disposed at various depths below the surface of the substrate support. In some embodiments, the heating elements 536 can be individually controlled. The physics-based digital twin model can provide predicted performance data for the substrate support 504 as output, which can inform the setpoints of the various electrodes housed in the substrate support 504 to reach one or more target property value profiles (e.g., a target temperature profile on the substrate) during substrate processing. The substrate support 504 may further include one or more radio frequency (RF) elements controlled by an RF power generator 534. The RF power generator 534 and RF elements may have some features in common with the chucking and heating system of the substrate support 504, as described above. Other types of electrodes may also be present in the substrate support and are within the scope of this disclosure.

[0111] In some embodiments, a removable sensor assembly can be used to measure properties of the substrate support 504. The properties can include characterizing the performance of electrodes within the support. Electrode performance (e.g., heat transfer to the surface of the substrate support 504, chucking force, etc.) can depend on the distance below the surface of the substrate support at which the electrode is located. The sensor assembly can be equipped with a capacitive sensor. For example, by connecting the electrode to ground, the capacitive sensor can measure the depth of an electrode or heating element below the surface of the substrate support assembly (e.g., below the top surface of an electrostatic chuck, below the top surface of a heater, etc.). This measurement can be repeated across the surface of the substrate support to generate a map of electrode / heating element depths, or for each electrode / heating element of interest included in the substrate support. The depth data can be provided to train a physics-based digital twin model of the substrate support. Parameters of the physics-based model (e.g., heat transfer from a heating electrode to the surface of the substrate support) can be a function of the electrode's depth below the surface.

[0112] When multiple electrodes of the same type are embedded within the substrate support, individual control of each electrode may be informed by measurements from the sensor assembly. For example, additional power may be supplied to heating electrodes located deeper below the surface (e.g., further from the substrate being processed) to compensate for reduced heat transfer through the material of the substrate support.

[0113] 6 is a diagram of a removable sensor assembly system 600, according to some embodiments. The geometries, sensor placement, component locations, etc. shown in FIG. 6 are intended to illustrate elements of embodiments of the present disclosure and are not intended to limit the scope of the present disclosure beyond the claims.

[0114] The sensor assembly system 600 includes a removable sensor assembly 620 (e.g., sensor assembly 195 in FIG. 1 ) including a sensor 602 disposed on a stand 604 configured to mount the sensor. The stand 604 may be configured to accommodate sensors of different ranges (as shown in FIG. 1 as two arms of the stand 604 disposed at different distances from a substrate support 608). The stand may be coupled to a movable support 606 (coupling is not shown in FIG. 6 but may include, for example, fixing both the stand 604 and the movable support 606 to the same surface). In some embodiments, the movable support 606 is coupled to a substrate support 608 (e.g., substrate support 504 in FIG. 5 ), the properties of which are measured by the sensor 602. The movable support 606 may be configured to cause relative motion between the substrate support 608 and the sensor 602 in at least two dimensions (e.g., two axes). For example, the movable support may be movable along an x-axis and a y-axis. In another example, the movable support 606 may be movable along the x-axis and rotatable about the z-axis. In some embodiments, the movable support 606 may be configured to induce relative motion in three dimensions, for example, to account for various ranges of the sensor 602. The movable support 606 may be configured to induce relative motion such that the entire working surface of the substrate support 608 passes through the field of view of the sensor 602. The processing device may receive data indicative of the position of the movable support 606 along with data indicative of sensor measurements of the sensor 602. The processing device may be configured to generate a two-dimensional map of the measured property of the substrate support 608. The processing device may also control aspects of the sensor assembly system 600, such as the sensor 602, the movable support 606, etc. In some embodiments, instead of the substrate support 608 being coupled to the movable support 606, the stand 604 is coupled to the movable support 606.

[0115] In some embodiments, the sensor 602 may include a capacitive displacement sensor 610, an infrared temperature sensor 612, and / or a surface profilometer 614 (e.g., a laser profilometer). The capacitive displacement sensor 610 may be configured to measure the depth of an electrode (e.g., a heating electrode, a clamp / chuck electrode, an RF electrode, etc.) below the surface of the substrate support 608. The infrared temperature sensor 612 may be configured to measure the temperature of a working surface of the substrate support 608 (e.g., when a heating element is turned on, when the substrate support 608 reaches a steady-state surface temperature, etc.). The surface profilometer 614 may be configured to measure a surface characteristic of the substrate support 608 (e.g., engineered surface roughness).

[0116] In some embodiments, the sensors 602 are selected to provide data to a physics-based digital twin model of the substrate support 608. Measured properties of the substrate support can include properties indicative of the performance of the heater assembly of the substrate support. For example, heater electrode power, heater electrode depth, surface profile, and surface temperature while the heater electrode is generating heat can all contribute to an accurate prediction of the temperature of a substrate being processed using the substrate support 608. By measuring the electrode depth, surface profile, and surface temperature and working with a known set point for heater electrode power, the temperature of the substrate can be predicted. Other combinations of sensors or generating predictions of other properties (e.g., measuring the depth of a clamping electrode to predict chucking force) are within the scope of this disclosure.

[0117] In some embodiments, the substrate support 608 may be measured by the sensor 602 outside of the fabrication chamber (e.g., before the substrate support 608 is installed in the fabrication chamber). In some embodiments, the sensor assembly 620 may be located in the fabrication chamber (e.g., during a planned maintenance event), and the substrate support 608 may be measured for characterization while the substrate support 608 is installed in the fabrication chamber. In some embodiments, the sensor assembly 620 may be located in the fabrication chamber during a planned or unplanned maintenance event to characterize or recharacterize the substrate support of the fabrication chamber. Execution of corrective action may be based on the characterization or recharacterization of the substrate support, as described in connection with FIGS. 4B, 4C, and 4E.

[0118] 7 is a block diagram illustrating a computer system 700 according to a particular embodiment. In some embodiments, computer system 700 may be connected to other computer systems (e.g., via a network such as a local area network (LAN), an intranet, an extranet, or the Internet). Computer system 700 may operate in the capacity of a server or a client computer in a client-server environment, or as a peer computer in a peer-to-peer or distributed network environment. Computer system 700 may be provided by a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a mobile phone, a web appliance, a server, a network router, switch, or bridge, or any device capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that device. Furthermore, the term "computer" is intended to include any collection of computers that individually or jointly execute a set (or sets) of instructions to perform any one or more of the methodologies described herein.

[0119] In a further aspect, computer system 700 may include a processing device 702, a volatile memory 704 (e.g., random access memory (RAM)), a non-volatile memory 706 (e.g., read-only memory (ROM) or electrically erasable programmable ROM (EEPROM)), and a data storage device 718, which may communicate with each other via a bus 708.

[0120] The processing device 702 may be provided by one or more processors, such as a general-purpose processor (e.g., a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a microprocessor implementing other types of instruction sets, or a microprocessor implementing a combination of multiple types of instruction sets) or a special-purpose processor (e.g., an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), or a network processor).

[0121] The computer system 700 may further include a network interface device 722 (e.g., coupled to a network 774). The computer system 700 may also include a video display unit 710 (e.g., an LCD), an alphanumeric input device 712 (e.g., a keyboard), a cursor control device 714 (e.g., a mouse), and a signal generating device 720.

[0122] In some implementations, the data storage device 718 may include a non-transitory computer-readable storage medium 724 (e.g., a non-transitory machine-readable storage medium) capable of storing instructions 726 encoding any one or more of the methods or functions described herein, including instructions for encoding the components of FIG. 1 (e.g., the prediction component 114, the model 190, etc.) and performing the methods described herein.

[0123] The instructions 726 may also reside, completely or partially, within the volatile memory 704 and / or within the processing device 702 during execution thereof by the computer system 700; thus, the volatile memory 704 and the processing device 702 may also constitute machine-readable storage media.

[0124] Although the computer-readable storage medium 724 is shown as a single medium in the illustrative example, the term "computer-readable storage medium" is intended to include a single medium or multiple media (e.g., a centralized or distributed database and / or associated caches and servers) that store one or more sets of executable instructions. The term "computer-readable storage medium" is also intended to include any tangible medium that is capable of storing or encoding a set of instructions for execution by a computer that cause the computer to perform any one or more of the methodologies described herein. The term "computer-readable storage medium" is intended to include, but is not limited to, solid-state memory, optical media, and magnetic media.

[0125] The methods, components, and features described herein may be implemented by discrete hardware components or integrated into the functionality of other hardware components, such as an ASIC, FPGA, DSP, or similar device. Furthermore, the methods, components, and features may be implemented by firmware modules or functional circuits within a hardware device. Furthermore, the methods, components, and features may be implemented in any combination of hardware devices and computer program components, or in a computer program.

[0126] Unless otherwise specified, terms such as "receive," "execute," "provide," "obtain," "cause," "access," "determine," "add," "use," "train," "generate," and the like may refer to actions and processes performed or implemented by a computer system that manipulate and transform data represented as physical (electronic) quantities in computer system registers and memory into other data similarly represented as physical quantities in computer system memory or registers or other such information storage, transmission, or display devices. Also, terms such as "first," "second," "third," "fourth," and the like, as used herein, are intended as labels to distinguish different elements and may not have any significance in terms of their numerical designation as to their ordering.

[0127] The examples described herein also relate to apparatus for performing the methods described herein. The apparatus may be specially constructed to perform the methods described herein, or may comprise a general-purpose computer system selectively programmed by a computer program stored on the computer system. Such a computer program may be stored on a computer-readable tangible storage medium.

[0128] The methods and illustrative embodiments described herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may be used in accordance with the teachings described herein, or it may prove convenient to construct more specialized apparatus to perform the methods described herein and / or each of their individual functions, routines, subroutines, or operations. Examples of configurations for a variety of these systems are set forth in the description above.

[0129] The above description is intended to be illustrative, not limiting. While the present disclosure has been described with reference to particular illustrative examples and implementations, it will be recognized that the present disclosure is not limited to the described examples and implementations. The scope of the present disclosure should be determined with reference to the following claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. receiving first data indicative of a characteristic of the substrate support from one or more sensors of a removable sensor assembly disposed proximate to the substrate support and including a capacitive displacement sensor, the first data including capacitive displacement sensor data; determining a depth of one or more electrodes below a surface of the substrate support based on the first data including the capacitive displacement sensor data; providing the depths of the one or more electrodes to a model of the substrate support; receiving, from the model, predicted performance data for the substrate support based on the depth of the one or more electrodes; obtaining measured performance data of the substrate support; taking corrective action in consideration of one or more differences between the measured performance data and the predicted performance data; and A method comprising:

2. the one or more sensors associated with the removable sensor assembly; Capacitive sensors, Surface profile measuring device, and Temperature Sensor The method of claim 1 , comprising:

3. The method of claim 1 , wherein the characteristic of the substrate support comprises a characteristic indicative of a performance of a heater assembly of the substrate support.

4. The method of claim 1 , wherein the substrate support comprises an electrostatic chuck heater assembly.

5. moving the one or more sensors to a plurality of positions relative to the substrate support; generating one or more measurements at the plurality of locations relative to the substrate support, the data indicative of the characteristic of the substrate support comprising the one or more measurements; The method of claim 1 further comprising:

6. the first data indicative of the characteristic of the substrate support includes measurements from a capacitive sensor; the method further comprising determining a depth of an electrode below the surface of the substrate support at one or more locations based on the measurements from the capacitive sensor. The method of claim 1.

7. generating the model of the substrate support, the model comprising a physics-based model of the substrate support, the physics-based model comprising a digital twin model, generating the physics-based model comprising: selecting values ​​for parameters describing heat transfer in the substrate support given the data characteristic of the substrate support; and solving a heat transfer equation describing heat transfer in the substrate support given the values ​​of the parameters; The method of claim 1 further comprising:

8. the data indicative of a characteristic of the substrate support includes a two-dimensional map of at least one of a heater electrode depth, a surface profile, or a steady-state surface temperature; The method of claim 1.

9. receiving second data indicative of a characteristic of the substrate support from the one or more sensors of the removable sensor assembly, the second data being generated taking into account measurements made by the one or more sensors at a different time than measurements associated with the first data indicative of the characteristic of the substrate support; determining at least one difference between the first data indicative of a property of the substrate support and the second data indicative of a property of the substrate support; performing a corrective action based on the at least one difference; and The method of claim 1 further comprising:

10. Acquiring capacitive displacement sensor data of a substrate support; providing the capacitive displacement sensor data to a model of the substrate support; receiving, from the model of the substrate support, predicted performance data for the substrate support based on a depth of one or more electrodes below a surface of the substrate support; determining predicted characteristics of a substrate processed using the substrate support in view of the predicted performance data; obtaining a measured property of the substrate processed using the substrate support; performing corrective action taking into account the predicted characteristics of the substrate and the measured characteristics of the substrate; A method comprising:

11. receiving data indicative of a characteristic of the substrate support from one or more sensors of a removable sensor assembly positioned proximate to the substrate support; providing data based on the data characteristic of the substrate support, including depth data for one or more electrodes, to a model of the substrate support, the model configured to generate the predicted performance data for the substrate support; The method of claim 10 further comprising:

12. The corrective measures are: Sending an alert to the user; scheduling preventative maintenance; Scheduling corrective maintenance; updating process strategies; Updating calibration tables or device constants, or Correcting chamber drift; The method of claim 10 , comprising at least one of:

13. receiving metrology data for the substrate; receiving data indicative of corrective actions associated with manufacturing equipment used to produce the substrate; training a machine learning model by providing metrology data of the substrate and data indicative of predicted properties of the substrate as training inputs, and providing data indicative of the corrective actions associated with manufacturing equipment used to produce the substrate as target outputs; The method of claim 10 further comprising:

14. providing the predicted properties of the substrate as input to a trained machine learning model; receiving data from the trained machine learning model indicative of corrective actions associated with manufacturing equipment; The method of claim 10 further comprising:

15. The method of claim 10 , wherein the predicted properties of the substrate include a two-dimensional map of at least one of a film thickness, an etch rate, or a stress of the substrate.

16. a stand configured to be positioned proximate to a substrate support; a movable support coupled to the stand and movable on at least two axes; a plurality of sensors disposed on the stand, the sensors including a capacitive displacement sensor configured to determine a depth of one or more electrodes below a surface of the substrate support, a temperature sensor, and a surface profilometer; Equipped with the movable support is configured to move the plurality of sensors to a plurality of positions relative to the substrate support, the plurality of sensors generating one or more measurements at the plurality of positions relative to the substrate support; Sensor assembly system.

17. 1. A processing device, comprising: receiving predicted performance data for the substrate support from a model of the substrate support, the predicted performance data being generated considering data collected by the plurality of sensors of the sensor assembly system; and performing corrective action considering the predicted performance data; a processing device configured to: The sensor assembly system of claim 16 further comprising:

18. 17. The sensor assembly system of claim 16, wherein the one or more measurements at the multiple positions relative to the substrate support are used by a processing device to generate a two-dimensional map of values ​​of at least one property of the substrate support.

19. The sensor assembly system of claim 16 , wherein the plurality of sensors comprises capacitive sensors, and the sensor assembly system is configured to determine a depth of an electrode below the surface of the substrate support.

Images