Heat transfer management in substrate support system

Abstract

A method includes identifying property data associated with a substrate support system; identifying target performance data associated with the substrate support system; based on the property data and the target performance data, determining region configuration data associated with the substrate supporting system; and causing the substrate support system to be configured based on the region configuration data.

Description

Technical Field

[0001] This disclosure relates to management in manufacturing systems, such as substrate support systems, and in particular to heat transfer management in manufacturing systems. Background Technology

[0002] Products are manufactured by performing one or more manufacturing processes using manufacturing equipment. For example, substrate processing equipment is used to process substrates by performing processes on substrates in a processing chamber. Summary of the Invention

[0003] The following is a brief overview of this disclosure in order to provide a basic understanding of some aspects of this disclosure. This overview is not an exhaustive summary of this disclosure. It is not intended to identify any important or key elements of this disclosure, nor is it intended to depict any category of particular embodiments of this disclosure or any category of the claims. Its sole purpose is to present some concepts of this disclosure in a concise form as a prelude to the more detailed description that follows.

[0004] In one aspect of this disclosure, a method includes: identifying property data associated with a substrate support system; identifying target performance data associated with the substrate support system; and causing thermal transfer management of the substrate support system based on the property data and the target performance data.

[0005] In another aspect of this disclosure, a non-transitory machine-readable storage medium stores instructions that, when executed, cause a processing device to perform operations including: identifying property data associated with a substrate support system; identifying target performance data associated with the substrate support system; and, based on the property data and the target performance data, causing one or more material operations to be performed on the substrate support system.

[0006] In another aspect of this disclosure, a system includes memory and processing means coupled to the memory. The processing means is configured to: identify property data associated with a substrate support system; identify target performance data associated with the substrate support system; and, based on the property data and the target performance data, cause one or more material operations to be performed on the substrate support system.

[0007] In one aspect of this disclosure, a method includes: identifying property data associated with a substrate support system; identifying target performance data associated with the substrate support system; determining region configuration data associated with the substrate support system based on the property data and the target performance data; and causing the substrate support system to be configured based on the region configuration data.

[0008] In another aspect of this disclosure, a non-transitory machine-readable storage medium stores instructions that, when executed, cause a processing device to perform operations including: identifying property data associated with a substrate support system; identifying target performance data associated with the substrate support system; determining area configuration data associated with the substrate support system based on the property data and the target performance data; and causing the substrate support system to be configured based on the area configuration data.

[0009] In another aspect of this disclosure, a system includes memory and processing means coupled to the memory. The processing means is configured to: identify property data associated with a substrate support system; identify target performance data associated with the substrate support system; determine region configuration data associated with the substrate support system based on the property data and the target performance data; and cause the substrate support system to be configured based on the region configuration data. Attached Figure Description

[0010] This disclosure is illustrated by way of example rather than limitation in the accompanying drawings.

[0011] Figure 1 This is a block diagram illustrating an exemplary system architecture according to certain implementations.

[0012] Figure 2 A dataset generator for building machine learning models is shown according to some implementations.

[0013] Figure 3 This is a block diagram illustrating the determination of prediction data according to certain implementation methods.

[0014] Figures 4A to 4F This is a flowchart of a method associated with heat transfer management according to certain implementations.

[0015] Figures 5A to 5U A portion of a substrate support system according to certain embodiments is shown.

[0016] Figure 6 This is a block diagram illustrating a computer system according to certain embodiments. Detailed Implementation

[0017] This article describes techniques involving heat transfer management in substrate support systems (e.g., wafer support systems).

[0018] Products are manufactured by performing one or more manufacturing processes using manufacturing equipment. For example, substrate processing equipment is used to process substrates by performing processes on substrates in a processing chamber. The processes performed on the substrate may include one or more of the following: plasma-operated chemical vapor deposition (CVD), atomic layer deposition (ALD), physical vapor deposition (PVD), plasma-enhanced chemical vapor deposition (PECVD), sputtering, electron beam processing, thermal evaporation, heating, cooling, etc.

[0019] Conventionally, during the processing of a substrate, heat transfer with the substrate is not precisely controlled. This can be caused by insufficient heat transfer coefficient of the coolant supplied to the cooling base through conventional cooling channels and / or insufficient compensation of azimuth non-uniformity by conventional heaters. Such inaccurate heat transfer management (e.g., insufficient wafer thermal uniformity) can lead to poor substrate quality, reduced yield, and the use of expensive equipment to attempt more precise heat transfer management, etc.

[0020] The apparatus, system, and method disclosed herein provide solutions to these and other drawbacks of conventional systems.

[0021] The processing device identifies property data associated with the substrate support system. In some embodiments, the substrate support system is an electrostatic chuck (ESC) stack, which includes layers such as ceramic disks, cooling plates, and adhesive materials for bonding the ceramic disks to the cooling plates. The ceramic disks may include heaters, clamping electrodes, gas channels, etc. The cooling plates may include cooling channels, gas channels, etc. The property data may correspond to measurement data of the components of the substrate support system (e.g., thickness, width, spacing, size, etc.).

[0022] The processing apparatus further identifies target performance data associated with the substrate support system. Target performance data may be a target thermal map, etch depth map, or deposition thickness map of the upper surface of the substrate support system. In some embodiments, the target thermal map is a substantially uniform thermal map (e.g., substantially the same temperature at all points on the upper surface of the substrate support system). In some embodiments, the target thermal map is an intentionally non-uniform thermal map (e.g., a first portion of the upper surface is at a first temperature and a second portion of the upper surface is at a second temperature different from the first temperature). Target performance data may be one or more target temperatures of the upper surface of the substrate support system. Target performance data may also be one or more target temperatures of the substrate disposed on the upper surface of the substrate support system.

[0023] The processing apparatus further enables thermal transfer management of the substrate support system based on property data and target performance data. In some embodiments, the processing apparatus uses a trained machine learning model, which is trained using historical property data and historical target performance data as inputs and historical thermal transfer management data as target outputs. The processing apparatus provides the property data and target performance data as inputs to the trained machine learning model and determines thermal transfer management data based on prediction data associated with the output of the trained machine learning model. This enables thermal transfer management to be based on the thermal transfer management data.

[0024] In some embodiments, to enable heat transfer management, the processing apparatus causes one or more material operations to be performed on the substrate support system. Material operations may include one or more of the following: removing material, adding material, processing surfaces, etc. For example, material operations may include adding material to a first portion of the upper surface of the substrate support system, removing material from a second portion of the upper surface of the substrate support system, and / or roughening (e.g., corrugating) the surface of the substrate support system (e.g., channel surface, outer surface).

[0025] In some embodiments, to enable heat transfer management, the processing apparatus determines region configuration data associated with the substrate support system based on property data and target performance data, and causes the substrate support system to be configured based on the region configuration data. In some embodiments, the regions may be annular (e.g., a ring having an outer circumference and an inner circumference), disk-shaped (e.g., substantially circular), or segmented (e.g., formed by different segments, multiple segments forming a ring or disk, etc.). In some embodiments, causing the substrate support system to be configured based on the region configuration data includes causing the substrate support system to be manufactured based on the region configuration data. This may include causing the substrate support system to have one or more continuous heaters and one or more pixelated heaters. In some embodiments, causing the substrate support system to be configured based on the region configuration data includes causing the substrate support system to be controlled based on the region configuration data. This may include causing one or more continuous heaters and / or one or more pixelated heaters to be controlled.

[0026] The aspects of this disclosure offer technical advantages. Compared to conventional systems, this disclosure allows for more precise thermal management of the substrate during the processes performed on the substrate (e.g., more precise thermal transfer with the substrate). Compared to conventional systems, this disclosure allows for more accurate substrate thermal uniformity or more accurate thermal mapping. This results in better substrate quality, increased yield, and reduced (or eliminated) use of expensive equipment for performing thermal management, etc.

[0027] Figure 1 This is a block diagram illustrating an exemplary system 100 (exemplary system architecture) according to certain embodiments. System 100 includes a client device 120, manufacturing equipment 124, a sensor 126, a metering device 128, a prediction server 112, and a data storage 140. In some embodiments, the prediction server 112 is part of the prediction system 110. In some embodiments, the prediction system 110 further includes server machines 170 and 180.

[0028] In some embodiments, one or more of the following components—client device 120, manufacturing equipment 124, sensor 126, metering device 128, prediction server 112, data storage 140, server machine 170, and / or server machine 180—are coupled to each other via network 130 to generate prediction data 168 for performing heat transfer management. In some embodiments, network 130 is a public network that provides client device 120 with access to prediction server 112, data storage 140, and 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, sensor 126, metering device 128, data storage 140, and other privately available computing devices. In some implementations, network 130 includes 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.

[0029] In some embodiments, the client device 120 and / or the substrate support system 125 includes a heat transfer management component 122 and / or a data storage device 140. The data storage device 140 may be a data storage chip, a printed circuit board (PCB) having memory such as random-access memory (RAM), and / or the like. The heat transfer management component 122 and / or the data storage device 140 may be used to: control heaters (e.g., pixelated heaters and / or continuous heaters), perform predictive calculations, and / or change the setpoint of heaters (e.g., multi-zone heaters).

[0030] In some embodiments, client device 120 includes a computing device, such as a personal computer (PC), portable computer, mobile phone, smartphone, tablet computer, laptop computer, etc. In some embodiments, client device 120 includes a heat transfer management component 122. In some embodiments, heat transfer management component 122 may also be included in prediction system 110 (e.g., machine learning processing system). In some embodiments, heat transfer management component 122 is alternatively included in prediction system 110 (e.g., not included in client device 120). Client device 120 includes an operating system that allows a user to perform one or more of the following: merging, generating, viewing, or editing data, providing guidance to prediction system 110 (e.g., machine learning processing system), etc.

[0031] In some embodiments, the heat transfer management component 122 receives one or more user inputs (e.g., a graphical user interface (GUI) displayed via client device 120), receives property data 142, receives target performance data 152, receives heat transfer management data 162, etc. In some embodiments, the heat transfer management component 122 sends data (e.g., user input, property data 142, target performance data 152, heat transfer management data 162, etc.) to the prediction system 110, receives prediction data 168 from the prediction system 110, and causes heat transfer management of the substrate support system 125 of the manufacturing equipment 124 based on the prediction data 168. In some embodiments, the heat transfer management component 122 stores data (e.g., user input, property data 142, target performance data 152, heat transfer management data 162, etc.) in a data memory 140, and the prediction server 112 retrieves data from the data memory 140. In some implementations, prediction server 112 stores the output of trained machine learning model 190 (e.g., prediction data 168) in data memory 140 and client device 120 retrieves the output from data memory 140. In some implementations, heat transfer management component 122 receives instructions for heat transfer management (e.g., based on prediction data 168) from prediction system 110 and causes heat transfer management of substrate support system 125.

[0032] In some implementations, the prediction data 168 is associated with heat transfer management (e.g., heat transfer management data 162). In some implementations, heat transfer management is associated with one or more of the execution of one or more material operations (e.g., removing material, adding material, treating surfaces, etc.) on the substrate support system, thereby resulting in the configuration (e.g., manufacturing, controlling, etc.) of the substrate support system and / or the like.

[0033] In some embodiments, corrective actions are performed based on thermal transfer management data 162. In some embodiments, corrective actions include providing an alarm (e.g., a warning not to use substrate processing equipment parts or manufacturing equipment 124 if predicted data 168 indicates that the predicted performance data does not meet expectations). In some embodiments, corrective actions include providing feedback control (e.g., cleaning, repairing, adding material to, removing material from, processing, and / or replacing substrate processing equipment parts in response to predicted data 168 indicating that the prediction is unfavorable). In some embodiments, corrective actions include providing machine learning (e.g., determining thermal transfer management data 162 based on predicted data 168).

[0034] In some implementations, prediction server 112, server machine 170, and server machine 180 each include one or more computing devices, such as rack servers, router computers, server computers, personal computers, mainframe computers, portable computers, tablet computers, desktop computers, graphics processing units (GPUs), accelerator application-specific integrated circuits (ASICs) (e.g., tensor processing units (TPUs)), etc.

[0035] The prediction server 112 includes a prediction component 114. In some embodiments, the prediction component 114 receives property data 142 and target performance data 152 (e.g., received from client device 120, retrieved from data storage 140) and generates prediction data 168 associated with heat transfer management data (e.g., heat transfer management). In some embodiments, the prediction component 114 uses one or more trained machine learning models 190 to determine the prediction data 168 for heat transfer management. In some embodiments, the trained machine learning model 190 is trained using historical property data 144, historical target performance data 154, and historical heat transfer management data 164.

[0036] In some implementations, prediction system 110 (e.g., prediction server 112, prediction component 114) uses supervised machine learning (e.g., a supervised dataset, historical property data 144 labeled with historical heat transfer management data 164, and historical target performance data 154, etc.) to generate prediction data 168. In some implementations, prediction system 110 uses semi-supervised learning (e.g., a semi-supervised dataset, heat transfer management data 162 being a predicted percentage, etc.) to generate prediction data 168. In some implementations, prediction system 110 uses unsupervised machine learning (e.g., an unsupervised dataset, a cluster, a cluster based on historical property data 144 and historical target performance data 154, etc.) to generate prediction data 168.

[0037] In some embodiments, manufacturing equipment 124 (e.g., clustering tool) is part of a substrate processing system (e.g., an integrated processing system). Manufacturing equipment 124 includes one or more of the following: a substrate support system 125, a controller, a housing system (e.g., a substrate carrier, a front-opening unified pod (FOUP), an automated teaching FOUP, a processing kit housing system, a substrate housing system, a die, etc.), a side storage pod (SSP), an alignment device (e.g., an alignment chamber), a factory interface (e.g., an equipment front end module (EFEM)), a load lock, a transfer chamber, one or more processing chambers, a robotic arm (e.g., disposed in the transfer chamber, disposed in the front interface, etc.), and / or the like. The housing system, SSP, and load lock mounted to the factory interface, and the robotic arm disposed in the factory interface, are used to transfer contents (e.g., substrates, processing kit rings, carriers, verification wafers, etc.) between the housing system, SSP, load lock, and factory interface. The alignment device is disposed in the factory interface to align the contents. A load locking and processing chamber installed in the transfer chamber, and a robotic arm disposed in the transfer chamber for transferring contents (e.g., substrate, processing sleeve ring, carrier, verification wafer, etc.) between the load locking, processing chamber, and transfer chamber.

[0038] In some embodiments, sensor 126 provides sensor data (e.g., sensor values, such as historical and current sensor values) associated with manufacturing equipment 124. In some embodiments, sensor 126 includes one or more of the following: imaging sensors (e.g., cameras, image capture devices, etc.), pressure sensors, temperature sensors, flow rate sensors, spectral sensors, and / or the like. In some embodiments, sensor data is used for equipment health and / or product health (e.g., product quality). In some embodiments, sensor data is received over a period of time. In some embodiments, sensor 126 provides sensor data such as image data, leakage rate, temperature, pressure, flow rate (e.g., gas flow rate), pumping efficiency, spacing (SP), high frequency radio frequency (HFRF), current, power, voltage, and / or the like. In some embodiments, property data 142 and target performance data 152 and / or thermal transfer management data 162 include sensor data from one or more sensors 126.

[0039] In some embodiments, property data 142 and / or target performance data 152 (e.g., historical property data 144 and historical target performance data 154, current property data 146 and current target performance data 156, etc.) are processed via client device 120 and / or via prediction server 112. In some embodiments, processing of property data 142 and target performance data 152 includes generating features. In some embodiments, features are a portion of property data 142 and / or target performance data 152 (e.g., dimensions, heatmaps, etc.), patterns in property data 142 and target performance data 152 (e.g., repetition of dimensions, heatmaps, etc.), or combinations of values ​​from property data 142 and target performance data 152 (e.g., size ratios, etc.). In some embodiments, property data 142 and target performance data 152 include features used by prediction component 114 to obtain prediction data 168.

[0040] In some embodiments, metrology equipment 128 (e.g., imaging equipment, spectroscopic equipment, ellipsometer equipment, etc.) is used to determine metrological data (e.g., inspection data, image data, spectral data, ellipticity data, material composition, optical, or structural data, etc.) corresponding to a substrate produced by manufacturing equipment 124 (e.g., substrate processing equipment). In some instances, after the substrate is processed by manufacturing equipment 124, metrology equipment 128 is used to inspect multiple portions (e.g., multiple layers) of the substrate. In some embodiments, metrology equipment 128 performs scanning acoustic microscopy (SAM), ultrasound inspection, X-ray inspection, and / or computed tomography (CT) inspection. In some instances, after manufacturing equipment 124 deposits one or more layers on the substrate, metrology equipment 128 is used to determine the quality of the processed substrate (e.g., layer thickness, layer uniformity, interlayer spacing, and / or the like). In some embodiments, metrology equipment 128 includes image capture devices (e.g., SAM equipment, ultrasound equipment, X-ray equipment, CT equipment, and / or the like).

[0041] In some embodiments, data storage 140 is memory (e.g., random access memory), a drive (e.g., a hard disk, USB flash drive), a database system, or another type of component or device capable of storing data. In some embodiments, data storage 140 includes multiple storage components (e.g., multiple drives or multiple databases) spanning multiple computing devices (e.g., multiple server computers). In some embodiments, data storage 140 stores one or more of the following: property data 142, target performance data 152, heat transfer management data 162, and / or prediction data 168.

[0042] Property data 142 includes historical property data 144 and historical target performance data 154, as well as current property data 146 and current target performance data 156. In some embodiments, property data 142 may include one or more of the thickness, width, spacing, size, etc., of the substrate support system. The substrate support system may include ceramic disks (e.g., electrostatic disks) and cooling plates bonded together. The ceramic disks may include clamping electrodes, heaters, gas channels, etc. The cooling plate may form cooling channels and gas channels.

[0043] The target performance data 152 includes historical target performance data 154 and current target performance data 156. The target performance data 152 may include a thermal map of the upper surface of the substrate support system (e.g., the upper surface of the ceramic disk, on which the substrate will be disposed).

[0044] Heat transfer management data 162 includes historical heat transfer management data 164 and current heat transfer management data 166. Heat transfer management data 162 may include one or more material operations (e.g., material removal, material addition, material handling), area configuration data, etc., of the substrate support system. In some embodiments, the area configuration data is associated with the fabrication of areas of the substrate support system (e.g., 3D printing of components forming channels, green sheet fabrication of components forming channels, etc.). In some embodiments, the area configuration data is associated with controlling the areas of the substrate support system (e.g., the duty cycle and / or voltage of each area, tuning the heater azimuth angle) to achieve a desired setpoint temperature (e.g., within a predetermined or minimum time amount).

[0045] In some implementations, historical data includes one or more of the following: historical property data 144, historical target performance data 154, and / or historical heat transfer management data 164 (e.g., at least a portion used to train machine learning model 190). Current data includes one or more of the following: current property data 146, current target performance data 156, and / or current heat transfer management data 166 (e.g., at least a portion to be input into the trained machine learning model 190 after training model 190 using historical data). In some implementations, current data is used to retrain the trained machine learning model 190.

[0046] In some implementations, the predicted data 168 will be used for heat transfer management (e.g., predicting heat transfer management data, resulting in processing the substrate based on the heat transfer management data).

[0047] In some implementations, the prediction system 110 further includes server machines 170 and 180. Server machine 170 includes a dataset generator 172 capable of generating datasets (e.g., a set of data inputs and a set of target outputs) to train, validate, and / or test the machine learning model 190. The dataset generator 172 has functions for data collection, compilation, simplification, and / or partitioning to put the data into a form suitable for machine learning. In some implementations (e.g., for small datasets), partitioning for post-training validation (e.g., explicit partitioning) is not used. Repeated cross-validation (e.g., 5-fold cross-validation, leave-one-out cross-validation) may be used during training, wherein the given dataset is actually repeatedly partitioned into different training and validation sets during training. The model (e.g., the best model, the model with the highest accuracy, etc.) is selected from vectors of models on automatically separated composite subsets. In some implementations, the dataset generator 172 may explicitly partition historical data (e.g., historical property data 144, historical target performance data 154, and corresponding historical heat transfer management data 164) into a training set (e.g., 60% of the historical data), a validation set (e.g., 20% of the historical data), and a test set (e.g., 20% of the historical data). In this implementation, some operations of the dataset generator 172 are described below regarding... Figure 2 and Figure 4A Detailed Description. In some embodiments, the prediction system 110 (e.g., via prediction component 114) generates multiple sets of features (e.g., training features). In some instances, a first set of features corresponds to a first set of types of property data 142 and target performance data 152 (e.g., a first type of operation associated with a first set of sensors, a first combination of values, a first pattern in values), the property data and target performance data corresponding to each of the datasets (e.g., training set, validation set, and test set), and a second set of features corresponds to a second set of types of property data 142 and target performance data 152 (e.g., a second type of operation associated with a second set of sensors different from the first set of sensors, a second combination of values ​​different from the first combination, a second pattern different from the first pattern), the property data and target performance data corresponding to each of the datasets.

[0048] Server machine 180 includes a training engine 182, a validation engine 184, a selection engine 185, and / or a testing engine 186. In some embodiments, engines (e.g., training engine 182, validation engine 184, selection engine 185, and testing engine 186) refer to hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing device, etc.), software (such as instructions that run on a processing device, general-purpose computer system, or dedicated machine), firmware, microcode, or a combination thereof. Training engine 182 is capable of training machine learning model 190 using one or more sets of features associated with a training set from dataset generator 172. In some embodiments, training engine 182 generates multiple trained machine learning models 190, each training machine learning model 190 corresponding to a different set of parameters of the training set (e.g., property data 142 and target performance data 152) and corresponding responses (e.g., heat transfer management data 162). In some embodiments, multiple models are trained on the same parameters with different objectives for the purpose of modeling multiple effects. In some instances, a first trained machine learning model is trained using property data 142 and target performance data 152 of all substrate support systems (e.g., substrate support systems 1-5), a second trained machine learning model is trained using a first subset of property data 142 and target performance data 152 (e.g., substrate support systems 1, 2, and 4), and a third trained machine learning model is trained using a second subset of property data 142 and target performance data 152 that partially overlaps with the first subset of features (e.g., substrate support systems 1, 3, 4, and 5).

[0049] The validation engine 184 is capable of validating the trained machine learning model 190 using a set of corresponding features from the validation set generated by the dataset generator 172. For example, a first trained machine learning model 190 trained using a first set of features from the training set is validated using a first set of features from the validation set. The validation engine 184 determines the accuracy of each trained machine learning model 190 based on the set of corresponding features from the validation set. The validation engine 184 evaluates and labels (e.g., discards) trained machine learning models 190 with accuracy that does not meet a threshold accuracy. In some embodiments, the selection engine 185 is capable of selecting one or more trained machine learning models 190 with accuracy that meets the threshold accuracy. In some embodiments, the selection engine 185 is capable of selecting the trained machine learning model 190 with the highest accuracy among the trained machine learning models 190.

[0050] The testing engine 186 can test the trained machine learning model 190 using a corresponding set of features from the test set generated by the dataset generator 172. For example, a first trained machine learning model 190 trained using a first set of features from the training set is tested using a first set of features from the test set. The testing engine 186 determines the trained machine learning model 190 with the highest accuracy among all trained machine learning models based on the test set.

[0051] In some implementations, machine learning model 190 (e.g., for classification) refers to model artifacts built by training engine 182 using a training set that includes data inputs and corresponding target outputs (e.g., correctly classifying the conditions or ordinal levels of the corresponding training inputs). Patterns mapping data inputs to target outputs (correct classifications or levels) can be discovered in the dataset and provided to machine learning model 190 with images capturing such patterns. In some implementations, machine learning model 190 uses one or more of the following: Gaussian Process Regression (GPR), Gaussian Process Classification (GPC), Bayesian Neural Network, Neural Network Gaussian Process, Deep Belief Network, Gaussian Mixture Model, or other probabilistic learning methods. Non-probabilistic methods may also be used, including one or more of the following: Support Vector Machine (SVM), Radial Basis Function (RBF), clustering, Nearest Neighbor algorithm (k-NN), linear regression, random forest, neural network (e.g., artificial neural network), etc. In some implementations, the machine learning model 190 is a multivariate analysis (MVA) regression model.

[0052] Prediction component 114 provides current property data 146 and current target performance data 156 (e.g., as input) to a trained machine learning model 190 and runs the trained machine learning model 190 (e.g., runs on the input to obtain one or more outputs). Prediction component 114 is capable of determining (e.g., extracting) prediction data 168 from the trained machine learning model 190 and determining (e.g., extracting) uncertainty data indicating the confidence level of prediction data 168 corresponding to current heat transfer management data 166. In some embodiments, prediction component 114 or heat transfer management component 122 uses uncertainty data (e.g., an uncertainty function or an acquisition function derived from an uncertainty function) to determine whether to use prediction data 168 to perform a correction action or whether to further train model 190.

[0053] For illustrative and not limiting purposes, aspects of this disclosure describe using historical data (i.e., previous data, historical property data 144, historical target performance data 154, and historical heat transfer management data 164) to train one or more machine learning models 190 and feeding current property data 146 and current target performance data 156 into one or more trained probabilistic machine learning models 190 to determine prediction data 168. In other embodiments, heuristic or rule-based models are used to determine prediction data 168 (e.g., without using trained machine learning models). In other embodiments, non-probabilistic machine learning models may be used. Prediction component 114 monitors historical property data 144, historical target performance data 154, and historical heat transfer management data 164. In some embodiments, regarding Figure 2 Any information described in data input 210 is monitored or used in a heuristic or rule-based model.

[0054] In some implementations, the functionality of client device 120, prediction server 112, server machine 170, and server machine 180 is provided by a smaller number of machines. For example, in some implementations, server machines 170 and 180 are integrated into a single machine, while in other implementations, server machine 170, server machine 180, and prediction server 112 are integrated into a single machine. In some implementations, client device 120 and prediction server 112 are integrated into a single machine.

[0055] Generally, 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, if appropriate. Furthermore, functionality attributable to specific components may be performed by different or multiple components operating together. For example, in some embodiments, prediction server 112 determines correction actions based on prediction data 168. In another instance, client device 120 determines prediction data 168 based on data received from a trained machine learning model.

[0056] Furthermore, the function of a particular component can be performed by different or multiple components operating together. In some implementations, one or more of the prediction server 112, server machine 170, or server machine 180 are accessed as a service, which is provided to other systems or devices through an appropriate application programming interface (API).

[0057] In some implementations, "user" refers to a single individual. However, other implementations of this disclosure cover "user" as an entity controlled by multiple users and / or automated sources. In some instances, a collection of independent users as a group of administrators is considered a "user".

[0058] Although embodiments of this disclosure are described based on predicted data 168 for determining heat transfer management of a substrate support system in a manufacturing facility (e.g., a substrate processing facility), in some embodiments, this disclosure can also be broadly applied to heat transfer management. Embodiments can be broadly applied to determining heat transfer management based on different types of data.

[0059] Figure 2 The following are examples of methods for building machine learning models according to certain implementations (e.g., Figure 1 The dataset generator 272 (e.g., model 190) of the dataset. Figure 1 The dataset generator 272). In some implementations, the dataset generator 272 is Figure 1 Part of server machine 170. (Through) Figure 2 The datasets generated by dataset generator 272 can be used to train machine learning models (see, for example, see...). Figure 4E ) to provide a substrate support system (e.g., see Figure 4F Heat transfer management.

[0060] Dataset generator 272 (e.g., Figure 1 The dataset generator 172) builds models for machine learning (e.g., Figure 1 The dataset (model 190). The dataset generator 272 uses historical property data 244 and historical target performance data 254 (e.g., Figure 1 Historical property data 144 and historical target performance data 154) and historical heat transfer management data 264 (e.g., Figure 1 (164) Establish a dataset based on historical heat transfer management data. Figure 2 System 200 shows a dataset generator 272, a data input 210, and a target output 220 (e.g., target data).

[0061] In some implementations, dataset generator 272 generates a dataset (e.g., training set, validation set, test set) including one or more data inputs 210 (e.g., training input, validation input, test input) and one or more target outputs 220 corresponding to the data inputs 210. The dataset also includes mapped data that maps the data inputs 210 to the target outputs 220. Data inputs 210 are also referred to as “features,” “attributes,” or “information.” In some implementations, dataset generator 272 provides the dataset to training engine 182, validation engine 184, or testing engine 186, wherein the dataset is used to train, validate, or test machine learning model 190. Further details... Figure 4A This describes some implementation methods for generating training sets.

[0062] In some embodiments, the dataset generator 272 generates data input 210 and target output 220. In some embodiments, data input 210 includes one or more sets of historical property data 244 and historical target performance data 254. In some embodiments, historical property data 244 and historical target performance data 254 include one or more operations (e.g., associated with sensor data from one or more types of sensors, combinations of sensor data from one or more types of sensors, patterns of sensor data from one or more types of sensors, and / or the like).

[0063] In some implementations, dataset generator 272 generates a first set of data inputs corresponding to a first set of historical property data 244A and historical target performance data 254A to train, validate, or test a first machine learning model, and dataset generator 272 generates a second set of data inputs corresponding to a second set of historical property data 244B and historical target performance data 254B to train, validate, or test a second machine learning model.

[0064] In some implementations, the dataset generator 272 discretizes (e.g., segments) one or more of the data input 210 or the target output 220 (e.g., in a classification algorithm for a regression problem). Discretization of the data input 210 or the target output 220 (e.g., by sliding window segmentation) transforms continuous values ​​of variables into discrete values. In some implementations, the discrete values ​​of the data input 210 indicate discrete historical property data 244 and historical target performance data 254 to obtain the target output 220 (e.g., discrete historical heat transfer management data 264).

[0065] The data input 210 and target output 220 used for training, validating, or testing the machine learning model include information about a specific facility (e.g., a specific substrate manufacturing facility). In some instances, historical property data 244, historical target performance data 254, and historical heat transfer management data 264 are used for the same manufacturing facility.

[0066] In some embodiments, the information used to train the machine learning model comes from a specific type of manufacturing equipment 124 of a manufacturing facility with specific characteristics, and allows the trained machine learning model to determine the outcome of the manufacturing equipment 124 of the specific group based on inputs of current parameters (e.g., current property data 146 and current target performance data 156) associated with one or more parts sharing the characteristics of the specific group. In some embodiments, the information used to train the machine learning model is for parts from two or more manufacturing facilities, and allows the trained machine learning model to determine the outcome of the parts based on inputs from one manufacturing facility.

[0067] In some implementations, after generating a dataset and using the dataset to train, validate, or test the machine learning model 190, the machine learning model 190 is further trained, validated, or tested (e.g., Figure 1 (current heat transfer management data 166) or adjustment (e.g., adjusting weights associated with the input data of the machine learning model 190, such as connection weights in a neural network).

[0068] Figure 3 The illustration according to certain embodiments is used to generate prediction data 368 (e.g., Figure 1 A block diagram of system 300 (predicting data 168). System 300 is used to train a machine learning model (e.g., Figure 1 Model 190) determines the predicted data 368 for use in the thermal transfer management of the substrate support system.

[0069] In box 310, system 300 (e.g., Figure 1 The prediction system 110) executes historical data (e.g., used for...) Figure 1 The data partitioning of model 190 (historical property data 344, historical target performance data 354, and historical heat transfer management data 364) (e.g., through...) Figure 1The server machine 170 uses a dataset generator 172 to generate a training set 302, a validation set 304, and a test set 306. In some instances, the training set is 60% of the historical data, the validation set is 20% of the historical data, and the test set is 20% of the historical data. The system 300 generates multiple sets of features for each of the training set, validation set, and test set. In some instances, if the historical data includes features derived from 20 substrate processing systems and 100 heatmaps (e.g., heatmaps of 20 substrate processing systems), the first set of features is substrate processing systems 1-10, the second set of features is substrate processing systems 11-20, and the training set is heatmaps... Figure 1-6 0, the validation set is heatmap 61-80, and the test set is heatmap 81-100. In this example, the first set of features for the training set will be from the heatmap. Figure 1-6 Parameters of substrate processing system 1-10.

[0070] In box 312, system 300 uses training set 302 to perform model training (e.g., via...). Figure 1 The training engine 182). In some embodiments, the system 300 uses multiple sets of features from the training set 302 (e.g., a first set of features from the training set 302, a second set of features from the training set 302, etc.) to train multiple models. For example, the system 300 uses a first set of features from the training set (e.g., the substrate processing systems 1-10 and the thermal...) Figure 1-6 0) to train a machine learning model to generate a first trained machine learning model, and using a second set of features from the training set (e.g., substrate processing systems 11-20 and thermal...) Figure 1-6 0) to generate a second trained machine learning model. In some embodiments, the first trained machine learning model and the second trained machine learning model are combined to generate a third trained machine learning model (e.g., in some embodiments, this may be a predictor that is better than the first or second trained machine learning model itself). In some embodiments, the sets of features used in the comparison models overlap (e.g., the first set of features is substrate processing systems 1-15 and the second set of features is substrate processing systems 5-20). In some embodiments, hundreds of models are generated, including models with various permutations of features and combinations of models.

[0071] In box 314, system 300 performs model validation using validation set 304 (e.g., via...). Figure 1The system 300 uses a corresponding set of features from the validation set 304 to validate each of the trained models. For example, the system 300 uses a first set of features from the validation set (e.g., substrate processing systems 1-10 and heatmaps 61-80) to validate a first trained machine learning model, and uses a second set of features from the validation set (e.g., substrate processing systems 11-20 and heatmaps 61-80) to validate a second trained machine learning model. In some implementations, the system 300 validates hundreds of models generated at block 312 (e.g., models with various permutations of features, combinations of models, etc.). At block 314, the system 300 determines the accuracy (e.g., through model validation) of each of the one or more trained models and determines 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, the flow returns to block 312, where the system 300 performs model training using different sets of features from the training set. In response to determining that one or more of the trained models have an accuracy that meets a threshold, the process continues to box 316. System 300 discards trained machine learning models with an accuracy below the threshold (e.g., based on a validation set).

[0072] In box 316, system 300 performs model selection (e.g., via...). Figure 1 The selection engine 185 determines which of one or more trained models that meets a threshold accuracy has the highest accuracy (e.g., selected model 308, based on validation in box 314). In response to the determination that two or more trained models meeting the threshold accuracy have the same accuracy, the process returns to box 312, where system 300 performs model training using a further refined training set corresponding to a further refined feature set to determine the trained model with the highest accuracy.

[0073] In box 318, system 300 performs model tests using test set 306 (e.g., via...). Figure 1The system 300 uses a test engine 186 to test the selected model 308. The system 300 uses a first set of features from the test set (e.g., operations 1-10 for products 81-100) to test the first trained machine learning model to determine if the first trained machine learning model meets a threshold accuracy (e.g., based on the first set of features from the test set 306). In response to the selected model 308's accuracy not meeting the threshold accuracy (e.g., the selected model 308 is overfitted to the training set 302 and / or validation set 304 and is not applicable to other datasets, such as the test set 306), the process continues to box 312, where the system 300 performs model training (e.g., retraining) using different training sets corresponding to different sets of features (e.g., operations). In response to determining that the selected model 308 has an accuracy that meets the threshold accuracy based on the test set 306, the process continues to box 320. In at least box 312, the model learns patterns in historical data for prediction, and in box 318, the system 300 applies the model to the remaining data (e.g., the test set 306) to test the predictions.

[0074] At frame 320, system 300 uses a trained model (e.g., selected model 308) to receive current property data 346 and current target performance data 356 (e.g., Figure 1 The current property data 146 and the current target performance data 156) and the prediction data 368 (e.g., extracted) for heat transfer management determined by the trained model (e.g., extracted) Figure 1 The predicted data 168 is used to perform correction actions. In some embodiments, the current property data 346 and the current target performance data 356 correspond to features of the same type in the historical property data 344 and the historical target performance data 354. In some embodiments, the current property data 346 and the current target performance data 356 correspond to features of the same type as a subset of the feature types in the historical property data 344 and the historical target performance data 354 used to train the selected model 308.

[0075] In some implementations, current data is received. In some implementations, the current data includes current heat transfer management data 366 (e.g., Figure 1 The current heat transfer management data 166) and / or current property data 346 and current target performance data 356. In some embodiments, at least a portion of the current data is obtained from a metering device (e.g., Figure 1 The metering device 128) or the data is received via user input. In some embodiments, model 308 is retrained based on current data. In some embodiments, the new model is trained based on current heat transfer management data 366, current property data 346, and current target performance data 356.

[0076] In some implementations, one or more of blocks 310-320 occur in various orders and / or together with other operations not presented or described herein. In some implementations, one or more of blocks 310-320 will not be performed. For example, in some implementations, one or more of the following will not be performed: data partitioning in block 310, model validation in block 314, model selection in block 316, and / or model testing in block 318.

[0077] Figures 4A to 4F This is a flowchart of methods 400A-F associated with heat transfer management according to certain embodiments. In some embodiments, methods 400A-F are executed by processing logic, which includes hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing device, etc.), software (such as instructions that run on a processing device, general-purpose computer system, or dedicated machine), firmware, microcode, or a combination thereof. In some embodiments, methods 400A-F are executed at least in part by prediction system 110 and / or client device 120. In some embodiments, methods 400A are executed at least in part by prediction system 110 (e.g., ...). Figure 1 Server machine 170 and dataset generator 172 Figure 2 The dataset generator 272) executes the method. In some embodiments, the prediction system 110 uses method 400A to generate a dataset for at least one of training, validating, or testing a machine learning model. In some embodiments, one or more of methods 400B-D are executed via client device 120 (e.g., heat transfer management component 122). In some embodiments, method 400E is executed via server machine 180 (e.g., training engine 182, etc.). In some embodiments, method 400F is executed via prediction server 112 (e.g., prediction component 114). In some embodiments, non-transitory storage media 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 execute one or more of methods 400A-F.

[0078] For simplicity, methods 400A-F are depicted and described as a series of operations. However, the operations according to this disclosure may occur in various orders and / or simultaneously, and may occur together with other operations not presented and described herein. Furthermore, in some embodiments, not all of the shown operations are performed to implement method 400A-F according to the disclosed objectives. Moreover, those skilled in the art will understand and recognize that method 400A-F may alternatively be represented by a state diagram or events as a series of interrelated states.

[0079] Figure 4APredictive data is generated based on datasets used to generate machine learning models in certain implementations (e.g., Figure 1 The flowchart of the method 400A for predicting data 168) is shown.

[0080] See Figure 4A In some implementations, at block 402, the processing logic of method 400A initializes the training set T to an empty set.

[0081] In box 404, the processing logic generates a first data input (e.g., a first training input, a first validation input), which includes historical property data and historical target performance data.

[0082] At block 406, the processing logic produces a first target output from one or more data inputs (e.g., a first data input). In some implementations, the first target output is historical heat transfer management data.

[0083] At box 408, the processing logic generates image data that indicates the input / output image, as appropriate. The input / output image (or image data) refers to the data input (e.g., one or more data inputs as described herein), the target output of the data input (e.g., where the target output identifies historical heat transfer management data 164), and the association between the data input and the target output.

[0084] In box 410, the processing logic adds the image data generated in box 408 to the dataset T.

[0085] At box 412, the processing logic branches based on whether the dataset T is sufficient for at least one of training, validating, and / or testing the machine learning model 190 (e.g., the uncertainty of the trained machine learning model meets a threshold uncertainty). If yes, execution proceeds to box 414; otherwise, execution continues back to box 404. It should be noted that in some implementations, the sufficiency of the dataset T is determined simply based on the number of input / output mappings in the dataset, while in some other implementations, in addition to or instead of the number of input / output mappings, the sufficiency of the dataset T is determined based on one or more other criteria (e.g., a measure of the diversity of data instances, accuracy, etc.).

[0086] At box 414, the processing logic provides a dataset T (e.g., to server machine 180) to train, validate, and / or test the machine learning model 190. In some implementations, the dataset T is a training set and is provided to a training engine 182 on server machine 180 to perform training. In some implementations, the dataset T is a validation set and is provided to a validation engine 184 on server machine 180 to perform validation. In some implementations, the dataset T is a test set and is provided to a test engine 186 on server machine 180 to perform testing. In the case of a neural network, for example, input values ​​(e.g., numerical values ​​associated with data input 210) of a given input / output mapping are input into the neural network, and output values ​​(e.g., numerical values ​​associated with target output 220) of the input / output mapping are stored in the output nodes of the neural network. The connection weights in the neural network are then adjusted according to a learning algorithm (e.g., backpropagation, etc.), and the procedure is repeated for other input / output mappings in the dataset T.

[0087] Following box 414, the machine learning model (e.g., machine learning model 190) can be trained using training engine 182 of server machine 180, validated using validation engine 184 of server machine 180, or tested using test engine 186 of server machine 180. The trained machine learning model is implemented via prediction component 114 (of prediction server 112) to generate predictive data (e.g., predictive data 168) for heat transfer management.

[0088] Figure 4B This is a method 400B associated with heat transfer management according to certain implementations.

[0089] At block 420 of method 400B, processing logic identifies property data associated with the substrate support system. The substrate support system may include components including one or more ceramic disks (e.g., electrostatic chucks including clamping electrodes and heaters, ceramic disks forming gas channels), cooling plates (e.g., forming cooling channels), and adhesive materials for bonding one or more ceramic disks and cooling plates (e.g., bonding a first ceramic disk to a second ceramic disk and a second ceramic disk to a cooling plate). The property data may include measurement data of one or more components of the substrate support system (e.g., height, width, thickness, channel height, channel width, corrugated protrusion height, corrugated protrusion width, corrugated protrusion period, etc.).

[0090] In some implementations, the property data includes sensor data received from one or more sensors associated with the substrate support system and / or simulation data associated with the substrate support system.

[0091] In some embodiments, the property data includes measurement data of one or more components of the substrate support system and / or processing chamber data of the processing chamber (e.g., the substrate support system disposed in the processing chamber). In some embodiments, the processing chamber data includes one or more of the following: flow rate data associated with the processing gas flowing into the processing chamber, exhaust port location data associated with the exhaust port of the processing chamber, and / or pressure data associated with the pressure of the processing chamber.

[0092] At block 422, processing logic identifies target performance data associated with the substrate support system. In some embodiments, the target performance data includes one or more of the following: a target thermal map, a target etched map, or a target deposition map associated with the upper surface of the substrate support system. In some embodiments, the target performance data is associated with etch rate, deposition rate, uniformity (e.g., etch and / or deposition uniformity), and / or the like. In some embodiments, the target thermal map is a substantially uniform thermal map. In some embodiments, the target thermal map is a substantially uniform thermal map. In some embodiments, the target thermal map includes a first portion at a first temperature and a second portion at a second temperature (e.g., an intentionally asymmetrical thermal map).

[0093] At block 424, the processing logic causes thermal transfer management of the substrate support system based on property data and target performance data. In some embodiments, the processing logic determines thermal transfer management data based on the property data and target performance data and causes thermal transfer management based on the thermal transfer management data. In some embodiments, thermal transfer management includes the execution of one or more material operations (e.g., see...). Figure 4C (See box 434). In some embodiments, heat transfer management includes the configuration of the substrate support system (e.g., see box 434). Figure 4D (444)

[0094] Figure 4C It is a method 400C associated with heat transfer management according to certain implementations.

[0095] In box 430, the processing logic identifies the property data associated with the substrate support system. This can be used with... Figure 4B The same as or similar to box 420.

[0096] In box 432, the processing logic identifies target performance data associated with the substrate support system. This can be used with... Figure 4B The same as or similar to box 422.

[0097] At block 434, the processing logic causes one or more material operations to be performed on the substrate support system based on property data and target performance data. The one or more material operations may include one or more of the following: removing material from the substrate support system; adding material to the substrate support system; or surface treatment of the substrate support system. In some embodiments, the material operations are performed on the upper surface of the cooling plate of the substrate support system. In some embodiments, the material operations are performed on the inner surface of the cooling channels forming the cooling plate of the substrate support system.

[0098] In some implementations, one or more material operations include causing the gas passage to have a variable gas passage width, a variable gas passage height, or to be configured to flow one or more of different gas composition types (e.g., helium (He), nitrogen (N2), argon, etc.).

[0099] Figure 4D It is a method 400D associated with heat transfer management according to certain implementations.

[0100] In box 440, the processing logic identifies the property data associated with the substrate support system. This can be used with... Figure 4B The same as or similar to box 420.

[0101] In box 442, the processing logic identifies target performance data associated with the substrate support system. This can be used with... Figure 4B The same as or similar to box 422.

[0102] At block 444, the processing logic causes the substrate support system to be configured based on property data and target performance data. In some embodiments, block 444 includes determining region configuration data associated with the substrate support system based on the property data and target performance data and causing the substrate support system to be configured based on the region configuration data. In some embodiments, causing the substrate support system to be configured based on the region configuration data includes causing the substrate support system to be manufactured based on the region configuration data (e.g., forming areas of channels, forming continuous heaters and pixelated heaters, etc.). In some embodiments, causing the substrate support system to be configured based on the region configuration data includes causing the substrate support system to be controlled based on the region configuration data (e.g., controlling one or more operating cycles, voltages, temperatures, etc. of heaters such as continuous heaters and / or pixelated heaters).

[0103] Figure 4E It is used to train predictive data to determine heat transfer management (e.g., Figure 1 Machine learning models (e.g., prediction data 168) Figure 1 The method of model 190) 400E.

[0104] See Figure 4EIn box 450 of method 400E, the logic identifies historical data. This can be combined with... Figure 4B The same as or similar to box 420.

[0105] In box 452, process the logic to identify historical target performance data. This can be combined with... Figure 4B The same as or similar to box 422.

[0106] In box 454, the processing logic identifies historical heat transfer management data. Historical heat transfer management data may indicate one or more of the following: historical material handling, historical area configuration, historical area manufacturing, historical area control, and / or similar.

[0107] In box 456, the processing logic uses data inputs including historical property data and historical target performance data, and a target output including historical heat transfer management data, to train a machine learning model to produce a trained machine learning model. The heat transfer management data time is determined (e.g., Figure 4B Box 424) can be used Figure 4C A trained machine learning model. In some implementations, the trained machine learning model is a neural network.

[0108] Figure 4F It is used to use trained machine learning models (e.g., Figure 1 Model 190) is used for heat transfer management method 400E.

[0109] See Figure 4F In box 460 of method 400F, the processing logic identifies the current property data. This can be combined with... Figure 4B The same as or similar to box 420.

[0110] In box 462, the processing logic identifies the current target performance data. This can be compared with... Figure 4B The same as or similar to box 422.

[0111] In box 462, the processing logic provides the current property data and the current target performance data as input to the trained machine learning model (e.g., through...). Figure 4E (Block 456 training).

[0112] In box 464, the processing logic receives the output associated with the predicted data from the trained machine learning model.

[0113] In box 466, the processing logic determines the current heat transfer management data based on the predicted data. The current heat transfer management data may indicate one or more of the following: current material handling, current area configuration, current area manufacturing, current area control, and / or similar.

[0114] Figures 5A to 5UA substrate support system 500 according to certain embodiments is shown (e.g., Figure 1 The substrate support system 125 is part of it.

[0115] This disclosure can be associated with heat transfer management in substrate support systems (e.g., wafer support systems, electrostatic chucks (ESCs), ESC stacks). This can be accomplished by spatially adjusting heat transfer across the substrate support system without using complex and expensive multi-region pixelated heaters. The substrate support system may include layers of uniform thickness (e.g., ceramic, adhesive material, base plate). The substrate support system may form channels of uniform thickness (e.g., gas channels and cooling channels). This disclosure can improve substrate thermal uniformity by spatially adjusting the heat transfer coefficient across the substrate support system by one or more of the following: introducing corrugations in the channels; using hybrid heaters; and / or incorporating layers of non-uniform thickness with reduced symmetry (e.g., asymmetric layers).

[0116] Conventional systems suffer from insufficient substrate thermal uniformity. This can be attributed to insufficient heat transfer coefficient from the coolant to the cooling base in the smooth channel and / or inadequate radial heaters to compensate for azimuth non-uniformity. This disclosure mitigates the drawbacks of conventional systems by spatially adjusting the heat transfer coefficient across the substrate support system to improve wafer uniformity. This can be achieved through one or more of the following: heat transfer management through channel engineering modifications (e.g., Figure 5G to Figure 5N ); by incorporating intentional asymmetric features to modulate heat transfer (e.g., Figures 5A to 5F , Figures 50 to 5P ); and / or through heat transfer regulation via a hybrid heater (e.g., Figures 5S to 5U ).

[0117] The substrate support system 500 may include a ceramic disk 530 with uniform thickness (e.g., and accommodating clamping electrodes 532, heaters 534, and / or gas channels 552), a uniform adhesive material 550 for securing the ceramic disk 530 to the cooling plate 540, a gas channel 552 with uniform thickness, and / or a cooling plate 540 with uniform thickness forming an embedded cooling channel 520.

[0118] Figures 5A to 5C A view of the substrate support system 500A before material handling is shown according to certain embodiments. Figure 5A A top view of a component (e.g., a cooling plate) of the substrate support system 500A is shown. Figure 5B A cross-sectional side view of a component (e.g., a cooling plate) of the substrate support system 500A is shown. Figure 5CA thermal map 510A is shown associated with the substrate support system 500A (e.g., the cooling plate, the upper surface of the substrate support system, the substrate disposed on the substrate support system, etc.). Key 512 shows the temperature range of thermal map 510 from high (e.g., the top of key 512) to low (e.g., the bottom of key 512). In some embodiments, target performance data for the substrate support system 500 includes thermal maps associated with the substrate support system 500 within a threshold temperature range (e.g., substantially the same temperature). In some embodiments, target performance data for the substrate support system 500 includes thermal maps associated with the substrate support system 500 that substantially meet a desired thermal map (e.g., substantially the same temperature, specific temperatures of different portions of the substrate support system and / or the substrate, etc.). The substrate support system 500A may have high thermal non-uniformity prior to material handling.

[0119] Figures 5D to 5F The diagram shows a view associated with the substrate support system 500B after material handling, according to certain embodiments. In some embodiments, the material handling includes material addition 502 and / or material removal 504.

[0120] Symmetry can be disrupted by one or more of the following: removing material from stacked individual layers (e.g., ceramic disks, cooling plates); adding material (e.g., soldering aluminum or copper to a machined ceramic disk or bonding ceramic frit to a ceramic disk or combination thereof); and / or using dissimilar bonding materials (e.g., using at least two bonding materials with different thermal conductivity). In some embodiments, material can be added and / or removed based on statistical thermal unit properties. In some instances, removing a high thermal conductivity metal from a cooling plate (e.g., a corresponding increase in the bonding material) can result in lower effective thermal conductivity (e.g., an increase in substrate temperature). In some instances, adding material with higher thermal conductivity can increase cooling (e.g., a decrease in fluid temperature). Low thermal non-uniformity can be achieved by intentionally adding and / or removing material to alter the heat transfer coefficient.

[0121] In some implementations, symmetry can be disrupted by removing material from the substrate facing the ceramic top surface based on static properties (e.g., by laser or mechanical means). In some implementations, the removed material may include isolated features (e.g., pits) or a set of features.

[0122] Figure 5D A top view of a component (e.g., a cooling plate) of the substrate support system 500B is shown. Figure 5E A cross-sectional side view of a component (e.g., a cooling plate) of the substrate support system 500B is shown. Figure 5FA thermal map 510B is shown that is associated with a substrate support system 500B (e.g., a cooling plate, the upper surface of the substrate support system, the substrate disposed on the substrate support system, etc.). Key 512 shows the temperature range of the thermal map from high to low. In some embodiments, the thermal map 510B substantially meets the target performance data of the substrate support system (e.g., within a threshold temperature range, substantially the same temperature, substantially meeting the desired thermal map, specific temperatures of different portions of the substrate support system and / or the substrate, etc.). In some embodiments, the substrate support system 500A (e.g., before material handling) does not meet the target performance data, while the substrate support system 500B (e.g., after material handling) meets the target performance data.

[0123] Figure 5G to Figure 5K A cross-sectional side view of a substrate support system 500 (e.g., a cooling plate) according to certain embodiments is shown.

[0124] Some conventional fluid channels are smooth. Fluid channels can be used for fluid flow (e.g., for conveying liquids or gases). In some instances, liquid coolant flows in fluid channels within a cooling plate (e.g., a cooling block for metal machining). Gas flow can occur through gas channels machined within a cooling plate (e.g., a metal base plate) and / or a ceramic disc (e.g., a ceramic workpiece) bonded to the cooling plate.

[0125] It is known that there may be insufficient heat transfer from the coolant to the cooling plate and / or insufficient heat transfer from the substrate to the cooling plate (e.g., cooling base).

[0126] In some embodiments, this disclosure introduces corrugations in the channel (e.g., a cooling channel). The channel may be located within a cooling plate (e.g., a metal cooling plate), a ceramic disc (e.g., a ceramic workpiece), and / or an intermediate ceramic block bonded to the ceramic disc and the cooling plate (e.g., a metal base plate). In some embodiments, the channel cross-section is rectangular or of different shapes.

[0127] Ripples can disrupt fluid flow (e.g., causing turbulence), which leads to an increase in the area of ​​fluid mixing and recirculation. The area near the ripples may become less irritated. Disturbances can alter the Nusselt number and friction factor, which in turn modulate heat transfer efficiency.

[0128] Ripples (e.g., protrusions, depressions) can be defined by height (H), width (w), and number of ripples (e.g., i = 0 to N). Multiple ripples can be separated by a period (P). In some embodiments, ripples include one or more of protrusions, depressions, surface textures, surface roughness, etc.

[0129] Corrugations can exist on one or more surfaces forming the channels. The number of corrugations on each surface can be different. Corrugations on opposite surfaces may be misaligned. The height (CH) and width (CW) of individual channels can vary across the substrate support system. Corrugations may not be limited to a rectangular cross-section.

[0130] In some implementations, corrugations alter the thermohydrodynamic characteristics of the fluid flow. The channel (e.g., a gas channel, a cooling channel) can be a single or multiple regions with non-uniform heights. The channel (e.g., a gas channel, a cooling channel) can have at least one corrugation at any stage of the channel (e.g., an inflow page, an outflow page). The corrugations can be periodic or aperiodic in height, width, or both. The channel can transport gas or liquid. The channel can be part of a cooling plate (e.g., a metal cooling base) and / or part of a ceramic disk (e.g., a ceramic plate bonded to a wafer support substrate).

[0131] In some embodiments, the substrate support system 500 (e.g., a cooling plate) forms cooling channels 520. Fluid can flow in one direction through the first cooling channel 520 and in the opposite direction in the second cooling channel 520. In some embodiments, the cooling channels 520 are corrugated (e.g., see...). Figures 5G to 5H In some embodiments, the cooling channel 520 has a threshold surface roughness.

[0132] See Figure 5G The cooling channel 520 may be corrugated and may have a channel width 522 and a channel height 524. An "X" indicates fluid inflow to the page, and a circle with a dot in the middle indicates fluid outflow from the page.

[0133] See Figure 5H The cooling channel 520 may have different dimensions including one or more of the following: channel width 522 (e.g., CWi), channel height 524 (CHi), protrusion width 526 (e.g., top protrusion width (wti), bottom protrusion width (wbi), left protrusion width (wli), right protrusion width (wri)), protrusion cycle 528 (e.g., top protrusion cycle (Pti), bottom protrusion cycle (Pbi), left protrusion cycle (Pli), right protrusion cycle (Pri)), and / or protrusion height 529 (e.g., top protrusion height (Hti), bottom protrusion height (Hbi), left protrusion height (Hli), right protrusion height (Hri)).

[0134] See Figures 5I to 5K The substrate support system 500 may form one or more channels (e.g., cooling channel 520, gas channel 552).

[0135] The channel can be an additive manufacturing channel formed through the following: an aluminum (Al) matrix with magnesium (Mg) or silicon (Si) dopants (e.g., similar to Al6061); a metal matrix including particles; aluminum silicon carbide (Al-SiC) (e.g., a matrix composite comprising an Al matrix with SiC particles); reinforced aluminum carbon (e.g., carbon nanotubes (CNTs), nanotubes, etc.); ceramic / functionally graded ceramics; and / or functionally graded metal-insulator matrices.

[0136] The cross-section of the channel (e.g., cooling channel 520, gas channel 552) may have at least one of the following: fins, a progressive approximation of a circular channel, a circular channel, or a computer-generated regenerative channel (e.g., a cooling fluid, such as compressed gas, expands in the channel to provide cooling). The cooling plate 540 may have both cooling channel 520 and gas channel 552.

[0137] See Figure 5I In some embodiments, the cross-section of the cooling channel 520 may be fin-shaped.

[0138] See Figure 5J In some embodiments, the cross-section of the cooling channel 520 may be approximated by a stepped shape representing a circular channel.

[0139] See Figure 5K In some embodiments, the cross-section of the cooling channel 520 may be a circular channel.

[0140] Figures 5L to 5P A cross-sectional side view of a substrate support system 500 (e.g., an ESC stack) according to certain embodiments is shown.

[0141] As is known, ESC stacks may have uniform cooling channels (e.g., gas channels and cooling channels in the cooling plate, or gas channels, ceramic disks and cooling channels in the cooling plate).

[0142] The substrate support system 500 may include a ceramic disk 530, a cooling plate 540, and an adhesive material 550 for bonding the ceramic disk 530 and the cooling plate 540 together. The ceramic disk 530 may include clamping electrodes 532 and heaters 534 (e.g., resistance heaters, multi-zone heaters, four-zone heaters, etc.). The cooling plate 540 (e.g., a base plate) may form cooling channels 520 through which fluid flows (e.g., to remove heat from the cooling plate 540). The substrate support system 500 may form gas channels 552 through which gas flows over the cooling plate 540, the adhesive material, and the ceramic disk 530 (e.g., to the upper surface of the substrate support system 500).

[0143] See Figure 5LThe gas channel 552 may form branches 554 of varying sizes (e.g., variable gas channel width and / or height, with or without corrugations) flowing through the ceramic disk 530. In some embodiments, the branches 554 have substantially the same width and varying heights. In some embodiments, branches 554 closer to the periphery of the substrate support system 500 have lower heights, and branches further away from the periphery of the substrate support system 500 (e.g., closer to the center) have greater heights. For a substantially constant width of the branches 554, increasing the channel height of the branches 554 reduces heat transfer efficiency compared to edges that have higher heat transfer due to thinner channels.

[0144] See Figures 5M to 5N In some embodiments, the substrate support system 500 includes a ceramic disk 530A, a ceramic disk 530B, and a cooling plate 540. The ceramic disks 530A and 530B can be bonded together using an adhesive material 550A. The ceramic disk 530B can be bonded to the cooling plate 540 using an adhesive material 550B. The ceramic disk 530A may include a clamping electrode 532 and a heater 534. The cooling plate 540 may include a cooling channel 520. The ceramic disk 530B may include branches 554 of a gas channel 552 (e.g., a gas channel layer).

[0145] See Figure 5M The gas channel 552 can be guided through the cooling plate 540, through the adhesive material 550B, into the ceramic disk 530B, and then connect to branches 554 of different sizes formed by the ceramic disk 530B, through the adhesive material 550A, and then through the ceramic disk 530A to the upper surface of the substrate support system 500. This can be a single-area gas channel.

[0146] See Figure 5N Multiple gas channels 552 may have the following trajectory: passing through cooling plate 540 (e.g., between cooling channels 520), through adhesive material 550B, to branches 554 of different sizes formed by ceramic disk 530B, through adhesive material 550A, and subsequently through ceramic disk 530A to the upper surface of substrate support system 500. This may be a multi-zone gas channel (e.g., a multi-zone gas channel passing through cooling plate, adhesive material, and ceramic disk).

[0147] In some embodiments, ceramic blocks 550A-B are single ceramic blocks. In some embodiments, ceramic blocks 550A-B are more than two ceramic blocks joined together by an adhesive material.

[0148] In some embodiments, the cooling plate 540 has a substantially flat surface. See also Figures 50 to 5NIn some embodiments, protrusions (e.g., metal protrusions, material additive 502) are formed on the cooling plate 540 having protrusions. In some embodiments, the protrusions (e.g., metal protrusions) are formed by machining the cooling plate 540 (e.g., the protrusions may be machined into the body of the cooling plate 540). In some embodiments, the protrusions (e.g., metal protrusions) are formed by adding material to the cooling plate 540 and / or removing material from the cooling plate 540. In some embodiments, the protrusions are made of the same or different material as the cooling plate 540. Having protrusions can increase effective adhesive thermal conductivity to provide faster cooling.

[0149] See Figure 5O The upper surface of a component (e.g., cooling plate 540) of the substrate support system 500 may include a material additive 502 (e.g., a protrusion). The protrusion may be intentionally non-uniform to achieve desired thermal performance on the upper surface of the substrate support system 500.

[0150] See Figure 5P The substrate support system 500 may include a cooling plate 540, which includes a material additive 502 (e.g., a protrusion) that interfaces with the adhesive material 550.

[0151] Figures 5Q to 5R A cross-sectional side view of a portion (e.g., a cooling plate 540) of a substrate support system 500 according to certain embodiments is shown.

[0152] In some embodiments, the cooling plate may have a single channel. In some embodiments, the cooling plate 540 may have multiple channels (e.g., multiple rows of cooling channels 520, at least two layers of cooling channels 520). For multiple channels, the coolant flow may be in the same direction or in different directions. The channel shape may be different (e.g., the channel may not have a rectangular cross-section). The channels may have a common inlet 542 and a common outlet 544. The flow rates in the multiple channels may be different.

[0153] In some embodiments, the cooling plate 540 has multiple rows of cooling channels 520 (e.g., the first row of cooling channels 520 is above the second row of cooling channels 520). An "X" indicates fluid inflow into the page, and a circle with a dot in the center indicates fluid outflow from the page.

[0154] See Figure 5Q The cooling plate 540 may include cooling channels 520 above each other with fluid flowing into the page (e.g., see cooling channel 520 with an “X” indicator) and cooling channels 520 above each other with fluid flowing out of the page (e.g., see cooling channel 520 with a circle with a dot in the middle).

[0155] See Figure 5RThe cooling plate 540 may include cooling channels 520 that are above each other and have opposing flow (for example, see cooling channel 520 with an “X” indication above cooling channel 520 with a circle indicating the center dot, and vice versa).

[0156] Figures 5S to 5U A top view of a substrate support system 500 according to some embodiments is shown. In some embodiments, the region may be annular (e.g., a ring having an outer circumference and an inner circumference), disc-shaped (e.g., substantially circular), or segmented (e.g., formed by different segments, multiple segments forming a ring or disc, etc.).

[0157] Some electrostatic chucks may have micro-region pixelated heaters and multi-region main heaters in different planes. Pixelated heaters can be used to provide fine-tuning (e.g., not the main heating power). Some electrostatic chucks may only have four-region heaters.

[0158] Figures 5S to 5U The substrate support system 500 may be a hybrid heater including a continuous heater and a pixelated heater (e.g., an azimuth distribution of the heaters). The continuous heater and the pixelated heater may not be limited to the same plane. The continuous heater and the pixelated heater may overlap spatially. The pixelated heater may provide a main heater and tunability. In some embodiments, the pixelated heater provides tunability and primary heating, and the continuous heater provides secondary heating.

[0159] The substrate support system 500 may have different portions 560 (e.g., a portion 560 on the upper surface). For example, the substrate support system 500 may have an outer portion 560A, a middle outer portion 560B, a middle inner portion 560C, and an inner portion 560D. Each of the portions 560 may have one or more heating regions configured for temperature tuning. In some embodiments, one or more of the portions 560 may be further divided into additional heating regions (e.g., by pixelated heaters).

[0160] See Figure 5S The substrate support system 500 may include an outer portion 560A serving as a continuous heater, a middle outer portion 560B serving as a continuous heater, a middle inner portion 560C serving as a pixelated heater (e.g., four heaters), and an inner portion 560D serving as a pixelated heater (e.g., four heaters). Figure 5S As shown, the central pixelated heater region (e.g., X>1) can be surrounded by a Z(>1) group of annular heater regions.

[0161] See Figure 5TThe substrate support system 500 may include an outer portion 560A serving as a pixelated heater (e.g., four heaters), a middle outer portion 560B serving as a pixelated heater (e.g., four heaters), a middle inner portion 560C serving as a continuous heater, and an inner portion 560D serving as a continuous heater. Figure 5T As shown, the central circular heater can be surrounded by X (>1) annular heater regions, which are surrounded by Z (>1) pixelated heaters.

[0162] See Figure 5U The substrate support system 500 may include an outer portion 560A serving as a pixelated heater (e.g., four heaters), a middle outer portion 560B serving as a continuous heater and overlapping the middle portion of the outer portion 560A, a middle inner portion 560C serving as a pixelated heater (e.g., four heaters), and an inner portion 560D serving as a continuous heater. Figure 5T As shown, there may be alternating pixelation and continuous heater regions.

[0163] Figure 6 This is a block diagram illustrating a computer system 600 according to certain embodiments. In some embodiments, the computer system 600 is one or more of the following: client device 120, prediction system 110, server machine 170, server machine 180, or prediction server 112.

[0164] In some embodiments, computer system 600 is connected (e.g., via a network, such as a Local Area Network (LAN), intranet network, extranet network, or the Internet) to other computer systems. In some embodiments, computer system 600 operates in a client-server environment within the capacity of a server or client computer, or as a peer computer in a peer-to-peer or distributed network environment. In some embodiments, computer system 600 is provided via a personal computer (PC), tablet PC, set-top box (STB), personal digital assistant (PDA), cellular phone, network device, server, network router, switch or bridge, or any device capable of executing a set of instructions (continuously or otherwise) that specifies actions to be taken by that device. Furthermore, the term "computer" should include any set of computers that independently or jointly execute a set of instructions (or multiple sets of instructions) to perform any one or more methods described herein.

[0165] In another aspect, the computer system 600 includes a processing device 602 that communicates with each other via a bus 608, volatile memory 604 (e.g., random access memory (RAM)), non-volatile memory 606 (e.g., read-only memory (ROM) or electronically erasable programmable ROM (EEPROM)), and a data storage device 616.

[0166] In some embodiments, the processing device 602 is provided by one or more processors, such as general-purpose processors (e.g., complex instruction set computing (CISC) microprocessors, reduced instruction set computing (RISC) microprocessors, very long instruction word (VLIW) microprocessors, microprocessors implementing other types of instruction sets, or microprocessors implementing combinations of various types of instruction sets) or special-purpose processors (e.g., application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), digital signal processors (DSPs), or network processors).

[0167] In some embodiments, the computer system 600 further includes a network interface device 622 (e.g., coupled to a network 674). In some embodiments, the computer system 600 also includes a video display unit 610 (e.g., an LCD), an alphanumeric input device 612 (e.g., a keyboard), a cursor control device 614 (e.g., a mouse), and a signal generation device 620.

[0168] In some embodiments, the data storage device 616 includes a non-transitory computer-readable storage medium 624 on which instructions 626 encoding any one or more of the methods or functions described herein are stored, including Figure 1 The instruction encoding components (e.g., heat transfer management component 122, prediction component 114, etc.) are used to implement the method described herein.

[0169] In some embodiments, instruction 626 is also wholly or partially resident in volatile memory 604 and / or processing device 602 during its execution via computer system 600. Therefore, in some embodiments, volatile memory 604 and processing device 602 may also constitute machine-readable storage media.

[0170] Although computer-readable storage medium 624 is shown as a single medium in the illustrative example, the term "computer-readable storage medium" should 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" should also include any tangible medium capable of storing or encoding a set of instructions for execution by a computer, which causes the computer to perform any one or more of the methods described herein. The term "computer-readable storage medium" should include, but is not limited to, solid-state memory, optical media, and magnetic media.

[0171] In some embodiments, the methods, components, and features described herein are implemented through discrete hardware components or integrated into the functionality of other hardware components, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), or similar devices. In some embodiments, the methods, components, and features are implemented through functional circuitry within firmware modules or hardware devices. In some embodiments, the methods, components, and features are implemented as any combination of hardware devices and computer program components, or as a computer program.

[0172] Unless otherwise specifically stated, terms such as “identify,” “cause,” “add,” “remove,” “process,” “provide,” “obtain,” “determine,” “train,” “predict,” “receive,” “update,” or similar refer to actions and processes performed or implemented by a computer system that manipulate and transform data represented as physical (electronic) quantities within computer system caches and memory into other data similarly represented as physical quantities within computer system memory or caches or other such information storage, transmission, or display devices. In some embodiments, as used herein, the terms “first,” “second,” “third,” “fourth,” etc., are designations for distinguishing different elements and do not have ordinal meaning based on their numerical designation.

[0173] The examples described herein also relate to an apparatus for performing the methods described herein. In some embodiments, this apparatus is specifically configured to perform the methods described herein, or includes a general-purpose computer system selectively programmed by a computer program stored in a computer system. Such a computer program is stored in a computer-readable tangible storage medium.

[0174] The methods and illustrative examples described herein are not inherently related to any particular computer or other device. In some implementations, various general-purpose systems are used in accordance with the teachings described herein. In some implementations, more specialized devices are configured to perform each of the methods described herein and / or their individual functions, routines, subroutines, or operations. Examples of structures for various such systems are set forth in the foregoing description.

[0175] The above description is intended to be illustrative and not restrictive. Although this disclosure has been described with reference to specific illustrative examples and embodiments, it will be appreciated that this disclosure is not limited to the described examples and embodiments. The scope of this disclosure should be determined by referring to the appended claims together with the full scope of the equivalents given by the claims.

Claims

1. A method comprising: Identify property data associated with the substrate support system; Identify target performance data associated with the substrate support system; and The thermal transfer management of the substrate support system is based on the aforementioned property data and target performance data.

2. The method of claim 1, wherein causing the heat transfer management comprises: causing one or more material operations to be performed on the substrate support system.

3. The method of claim 1, wherein the property data comprises one or more of the following: Sensor data received from one or more sensors associated with the substrate support system; or Simulation data associated with the substrate support system.

4. The method of claim 1, wherein the property data comprises one or more of the following: measurement data of one or more components of the substrate support system; or processing chamber data of the processing chamber, wherein the substrate support system is disposed in the processing chamber, and the processing chamber data comprises one or more of the following: flow rate data associated with processing gas flowing into the processing chamber, exhaust port location data associated with exhaust port of the processing chamber, or pressure data associated with pressure of the processing chamber.

5. The method of claim 1, wherein the target performance data includes one or more of the following: a target thermal map, a target etched map, or a target deposition map associated with the upper surface of the substrate support system.

6. The method of claim 2, wherein the one or more material operations comprise one or more of the following: Remove material from the substrate support system; Add material to the substrate support system; or The surface treatment of the substrate support system.

7. The method of claim 2, further comprising: The property data and the target performance data are provided as input to the trained machine learning model; Obtain the output associated with the predicted data from the trained machine learning model; and Based on the predicted data, determine the one or more material operations to be performed on the substrate support system to meet the target performance data.

8. The method of claim 7, wherein the trained machine learning model is trained based on data input including historical property data and historical target performance data of the historical substrate support system, and target output including historical material operations on the historical substrate support system.

9. The method of claim 2, wherein the one or more materials are manipulated to form intentionally asymmetric features in the substrate support system.

10. The method of claim 2, wherein the substrate support system comprises: A ceramic disk that serves one or more of the following functions: accommodating a clamping electrode; accommodating a heater; or forming a gas passage. Cooling plate, the cooling plate forming a cooling channel and a gas channel; and An adhesive material is used to couple the ceramic disk to the cooling plate.

11. The method of claim 10, wherein the one or more material operations include forming protrusions on the cooling plate.

12. The method of claim 10, wherein the one or more material operations comprise corrugating the cooling channels to cause disturbances in the fluid flow to modulate heat transfer efficiency.

13. The method of claim 10, wherein the one or more material operations comprise causing the gas passage to have a variable gas passage width, a variable gas passage height, or to be configured to flow one or more of different gas composition types.

14. The method of claim 10, wherein the one or more material operations comprise causing the gas passage to be a multi-region gas passage through the cooling plate, the adhesive material, and the ceramic disk.

15. The method of claim 10, wherein the one or more material operations comprise causing the cooling channel to be one or more of the following: fin; The stepped approximation represents a circular channel; Circular passage; Stacking; or A computer-generated regeneration channel.

16. A non-transitory machine-readable storage medium for storing instructions, which, when executed, cause a processing device to perform an operation comprising: Identify property data associated with the substrate support system; Identify target performance data associated with the substrate support system; and Based on the property data and the target performance data, one or more material operations are performed on the substrate support system.

17. The non-transitory machine-readable storage medium of claim 16, wherein the one or more material operations comprise one or more of the following: Remove material from the substrate support system; Add material to the substrate support system; or The surface treatment of the substrate support system.

18. The non-transitory machine-readable storage medium of claim 16, wherein the operation further comprises: The property data and the target performance data are provided as input to the trained machine learning model; Obtain the output associated with the predicted data from the trained machine learning model; and Based on the predicted data, determine the one or more material operations to be performed on the substrate support system to meet the target performance data.

19. A system comprising: Memory; Processing device, coupled to the memory, the processing device being used for: Identify property data associated with the substrate support system; Identify target performance data associated with the substrate support system; as well as Based on the property data and the target performance data, one or more material operations are performed on the substrate support system.

20. The system of claim 19, wherein the one or more material operations comprise one or more of the following: Remove material from the substrate support system; Add material to the substrate support system; or The surface treatment of the substrate support system.

21. A method comprising: Identify property data associated with the substrate support system; Identify target performance data associated with the substrate support system; Based on the property data and the target performance data, determine the region configuration data associated with the substrate support system; and This results in the configuration of the substrate support system based on the regional configuration data.

22. The method of claim 21, wherein the property data comprises one or more of the following: Sensor data received from one or more sensors associated with the substrate support system; or Simulation data associated with the substrate support system.

23. The method of claim 21, wherein causing the substrate support system to be configured based on the region configuration data comprises: causing the substrate support system to be manufactured based on the region configuration data.

24. The method of claim 21, wherein the region configuration data is associated with a plurality of regions of the substrate support system, the plurality of regions including one or more annular regions, disc-shaped regions, or segmented regions.

25. The method of claim 21, wherein causing the substrate support system to be configured based on the region configuration data comprises: causing the substrate support system to be controlled based on the region configuration data.

26. The method of claim 21, wherein the property data comprises one or more of the following: Measurement data of one or more components of the substrate support system; or Processing chamber data, wherein the substrate support system is disposed in the processing chamber, the processing chamber data includes one or more of the following: flow rate data associated with the processing gas flowing into the processing chamber, exhaust port location data associated with the exhaust port of the processing chamber, or pressure data associated with the pressure of the processing chamber.

27. The method of claim 21, wherein the target performance data comprises one or more of the following: a target thermal map, a target etched map, or a target deposition map associated with the upper surface of the substrate support system.

28. The method of claim 21, further comprising: The property data and the target performance data are provided as input to the trained machine learning model; Obtain the output associated with the predicted data from the trained machine learning model; and Based on the predicted data, the regional configuration data is determined to cause the substrate support system to meet the target performance data.

29. The method of claim 28, wherein the trained machine learning model is trained based on data input including historical property data and historical target performance data of the historical substrate support system, and target output including historical area configuration data associated with the historical substrate support system.

30. The method of claim 21, wherein the substrate support system comprises: A ceramic disk that serves one or more of the following functions: accommodating a clamping electrode; accommodating a heater; or forming a gas passage. Cooling plate, the cooling plate forming a cooling channel and a gas channel; and An adhesive material is used to couple the ceramic disk to the cooling plate.

31. The method of claim 21, wherein the substrate support system comprises a hybrid heater system, the hybrid heater system comprising a plurality of heaters, a first subset of the plurality of heaters being continuous heaters and a second subset of the plurality of heaters being pixelated heaters.

32. The method of claim 31, wherein the plurality of heaters comprises a first heater in a first plane and a second heater in a second plane different from the first plane.

33. The method of claim 31, wherein at least a portion of the plurality of heaters overlap in space.

34. The method of claim 31, wherein the pixelated heater provides tunability and primary heating, and wherein the continuous heater provides secondary heating.

35. A non-transitory machine-readable storage medium for storing instructions, which, when executed, cause a processing device to perform an operation comprising: Identify property data associated with the substrate support system; Identify target performance data associated with the substrate support system; Based on the property data and the target performance data, determine the area configuration data associated with the substrate support system; as well as This results in the configuration of the substrate support system based on the regional configuration data.

36. The non-transitory machine-readable storage medium of claim 35, wherein causing the substrate support system to be configured based on the region configuration data comprises: causing the substrate support system to be manufactured based on the region configuration data.

37. The non-transitory machine-readable storage medium of claim 35, wherein causing the substrate support system to be configured based on the region configuration data comprises: causing the substrate support system to be controlled based on the region configuration data.

38. A system comprising: Memory; Processing device, coupled to the memory, the processing device being used for: Identify property data associated with the substrate support system; Identify target performance data associated with the substrate support system; Based on the property data and the target performance data, determine the area configuration data associated with the substrate support system; as well as This results in the configuration of the substrate support system based on the regional configuration data.

39. The system of claim 38, wherein, in order to cause the substrate support system to be configured based on the region configuration data, the processing apparatus is configured to cause the substrate support system to be manufactured based on the region configuration data.

40. The system of claim 38, wherein, in order to cause the substrate support system to be configured based on the region configuration data, the processing means is configured to cause the substrate support system to be controlled based on the region configuration data.

Images