Digital simulation for semiconductor manufacturing processes
Patent Information
- Application Number
- TW113111481
- Authority / Receiving Office
- TW · TW
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2023-03-28
- Filing Date
- 2024-03-27
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2044-03-26
Smart Images

Figure TWG2TB001920263_001 
Figure TWG2TB001920263_002 
Figure TWG2TB001920263_003
Abstract
Description
Technical Field
[0001] Embodiments of the present disclosure relate generally to manufacturing systems and, more particularly, to controlling semiconductor processing systems using model-based digital simulation. Prior Art
[0002] As equipment sizes shrink, the complexity of semiconductor wafer processing continues to increase. A typical process has multiple distinct steps, with some advanced processes, such as plasma etching, having twenty or more. Each step has multiple associated knobs to adjust for optimal performance. Therefore, theoretically, the scope for tuning and optimizing a specific process is enormous.
[0003] Process engineers use their experience and expertise to select a preliminary baseline process and fine-tune it based on a limited number of wafers (or portions of wafers, called coupons) dedicated to a design of experiments (DoE). The goal of DoE is to tailor the process to achieve the desired specifications on the wafers. However, dedicating an entire wafer or a portion of a wafer to DoE data collection consumes valuable resources. As a result, the process often adopted may be a viable process, but not necessarily the optimal solution.
[0004] Furthermore, over time, the quality and consistency of substrates processed at a manufacturing system may deviate from the target quality and consistency associated with a process recipe. However, it may be difficult for an operator of the manufacturing system to identify the deviation from the target quality and consistency, determine the defective component causing the deviation, and determine the appropriate action to address the deviation in quality and consistency. Summary of the Invention
[0005] The following is a simplified summary of the disclosure to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is not intended to identify key or critical elements of the disclosure, nor is it intended to delineate any scope of the specific embodiments of the disclosure or any scope of the claims. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that follows.
[0006] In one example, a method is provided that includes the steps of causing manufacturing equipment to release at least one gaseous substance at a specified flow rate through a plurality of regions of a pixelated showerhead for a manufacturing process; obtaining a gas distribution for each of the plurality of regions of the pixelated showerhead from a trained digital simulation associated with the manufacturing equipment; simulating a wafer characteristic based at least in part on the obtained gas distribution, the simulation predicting at least one characteristic of a processed wafer to be produced by the manufacturing process; and causing execution of one or more corrective actions associated with the manufacturing equipment based on the prediction.
[0007] In another example, a method is provided that includes the steps of causing a plurality of individually tunable heaters to heat a plurality of regions of an electrostatic chuck to a specified temperature for a manufacturing process, wherein at least one of the plurality of individually tunable heaters is associated with at least one region of the electrostatic chuck; obtaining a heat profile for each of the plurality of regions of the electrostatic chuck from a trained digital simulation associated with the electrostatic chuck; simulating a wafer characteristic based at least in part on the obtained heat profile, the simulation predicting at least one characteristic of a processed wafer to be produced by the manufacturing process; and causing performance of one or more corrective actions associated with the manufacturing process based on the prediction.
[0008] In yet another example, a method is provided that includes the steps of causing a plurality of individually tunable RF field generators to generate plasma across a plurality of regions in a processing chamber in which a fabrication process is performed, wherein at least one of the plurality of individually tunable RF field generators is associated with at least one region of the processing chamber; obtaining a plasma distribution for each of the plurality of regions of the processing chamber from a trained digital simulation associated with the processing chamber; simulating a wafer characteristic based at least in part on the obtained plasma distribution, the simulation predicting at least one characteristic of a processed wafer to be produced by the fabrication process; and causing performance of one or more corrective actions associated with the fabrication process based on the prediction.
[0009] In yet another example, a method is provided that includes the steps of causing a plurality of individually tunable heaters to heat a plurality of regions of an electrostatic chuck to a specified temperature for a manufacturing process, wherein at least one of the plurality of individually tunable heaters is associated with at least one region of the electrostatic chuck; causing a plurality of individually tunable RF field generators to generate plasma across a plurality of regions in a processing chamber in which the manufacturing process is performed, wherein at least one of the plurality of individually tunable RF field generators is associated with at least one region of the processing chamber; and A trained digital simulation associated with the electrostatic chuck is used to obtain heat distribution information for each of the plurality of regions of the electrostatic chuck; a trained digital simulation associated with the processing chamber is used to obtain plasma distribution information for each of the plurality of regions of the processing chamber; a wafer characteristic is simulated based in part on at least one of the heat distribution information or the plasma distribution information, the simulation predicting at least one characteristic of a processed wafer to be produced by the manufacturing process; and based on the prediction, one or more corrective actions associated with the manufacturing process are caused to be performed. Simple diagram description
[0010] In order to provide a detailed understanding of the aforementioned features of the present disclosure, a more detailed description of the present disclosure, briefly summarized above, may be obtained by reference to embodiments, some of which are illustrated in the accompanying drawings. It should be noted, however, that the drawings illustrate only exemplary embodiments and are therefore not to be considered limiting of the scope of the present disclosure, as the disclosure may admit to other equally effective embodiments.
[0011] FIG1 depicts an exemplary system architecture of the present disclosure.
[0012] FIG2 depicts a schematic top view of an example manufacturing system.
[0013] 3 depicts a schematic cross-sectional view of an exemplary processing chamber.
[0014] FIG4 depicts a schematic plan view of an exemplary panel for a showerhead.
[0015] 5 is a partial cross-sectional schematic side view detailing portions of a substrate support assembly having a spatially tunable heater.
[0016] 6A-6D are top views of a substrate support assembly illustrating example layouts for spatially tunable heaters.
[0017] FIG7 depicts a simplified timing diagram of a manufacturing process with parallel real-time simulation.
[0018] FIG8 depicts an exemplary environment for a manufacturing process with parallel real-time simulation.
[0019] FIG9 depicts another exemplary environment of a manufacturing process with parallel real-time simulation.
[0020] FIG10 depicts another exemplary environment for a manufacturing process with parallel real-time simulation.
[0021] FIG. 11 depicts an exemplary environment for real-time process control.
[0022] FIG12 depicts an exemplary method for controlling a manufacturing process in real time using autonomous parameter adjustments.
[0023] FIG13 depicts an example machine of a computer system.
[0024] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. It is contemplated that elements and features of one embodiment may be beneficially incorporated in other embodiments without further recitation. Implementation Method
[0025] This article describes techniques for automatically controlling a semiconductor manufacturing process in real time using a model-based digital simulation that operates in parallel with the physical manufacturing process. Manufacturing equipment can be used to produce products (e.g., electronic devices). For example, the manufacturing equipment can execute a recipe to produce an electronic device by supplying certain process gases from designated nozzles of a pixelated showerhead into a processing chamber at specified rates and times to generate a plasma for etching or deposition operations on a semiconductor wafer.
[0026] In some traditional systems, process engineers manually adjust process parameters if process output deviates from the desired process output. However, these manual adjustments vary depending on individual judgment. Furthermore, there is a time delay between noticing that a process drift is occurring, determining the possible cause of the drift, deciding on the process adjustments to make, and inputting the necessary adjustments into the manufacturing process controller.
[0027] The apparatus, systems, and methods disclosed herein enable real-time control of a manufacturing process using autonomous parameter adjustments while the manufacturing process is being executed. Real-time control of the process is achieved by running one or more computer simulations of the process from one or more trained machine learning models in parallel with the actual operation of the process. The one or more trained machine learning models can be trained for the process being executed by a specific manufacturing component. For example, the one or more trained machine learning models can be trained for: pixelated control of the temperature of a substrate support (e.g., an electrostatic chuck (ESC)); pixelated control of gas flow from a showerhead; pixelated control of local power used to generate a localized plasma (e.g., via the substrate support); and / or pixelated control of local power used to generate a localized plasma via an upper electrode (e.g., an electrode in a showerhead or a process chamber lid). Furthermore, a second set of machine learning models compares the simulation results with target results and determines one or more adjustments that can be made to the process while the process is being executed in parallel. The adjustments are then run through the computer simulation again to determine a predicted process outcome, which is then compared to the target outcome. If the adjustments are deemed to match the target results, the process controller is instructed to make these adjustments to the physical process in substantial real time.
[0028] Various aspects of the present disclosure relate to establishing a spatial model for predicting the performance of semiconductor processing equipment (also known as a "chamber") using spatial digital DoE. The spatial model is also capable of digitally generating a virtual recipe for any set of conditions within the process space. The term "virtual recipe" refers to a process recipe generated using a spatial digital model of a wafer (i.e., a virtual wafer) rather than a physical wafer or test piece. The term "virtual wafer" includes not only wafers, but also other objects, such as masks used to perform mask-based metrology. The spatial model uses machine learning to exploit the synergy between metrology data and various knobs. The term "knob" is used to refer to a variable that can be adjusted to control a process. Knobs are sometimes also called process control variables, or simply process variables.
[0029] Real-time monitoring data from an ongoing process is input into a digital simulation associated with the manufacturing process. In some embodiments, the digital simulation associated with the manufacturing process is a digital replica or digital "twin" of the manufacturing process, simulating the exact conditions of the manufacturing process. In other embodiments, the digital simulation may only simulate some conditions of the manufacturing process. An output representing predictive data can be obtained from the digital simulation. The output representing predictive data can be generated using a trained machine learning model, a heuristic model, or a rule-based model. Based on the predictive data, the processing device can execute one or more corrective actions associated with the manufacturing equipment.
[0030] In some embodiments, the flow rates of certain process gases in a pixelated showerhead of a processing chamber are adjustable. The predicted data may include one or more gas flow rate adjustments for certain nozzles or groups of nozzles in the pixelated showerhead. Corrective action may include adjusting the gas flow rate through one or more nozzles or groups of nozzles in the showerhead based on the predicted data.
[0031] In some embodiments, a pixelated electrostatic chuck of a processing chamber includes individually tunable heaters. The predicted data may include one or more of the following parameters: electrostatic chuck temperature, process time, gas type, power, RF bias, pressure, or temperature within the processing chamber. The corrective action may include adjusting the amount of power delivered to one or more individually tunable heaters or groups of heaters in the electrostatic chuck based on the predicted data.
[0032] In some embodiments, the amount or composition of the plasma generated in the processing chamber is adjustable. The predicted data may include information regarding process gases, temperatures, flow rates, gas ratios, pressures, power profile information, and other parameters used to generate the plasma in the processing volume. Corrective action may include adjusting the amount of power delivered to one or more electrodes or groups of electrodes to adjust the amount of RF bias generated in the processing volume.
[0033] Aspects of the present disclosure yield technical advantages that significantly reduce the time required to achieve optimal settings and reduce energy consumption. By using signal processing, digital simulation, and machine learning models to obtain predictive data and initiate corrective actions based on the predicted data, trial-and-error iterations can be reduced or eliminated entirely. The present disclosure can predict optimal parameter settings associated with etching or deposition processes, thereby avoiding inconsistent and abnormal products, unplanned user time, and damage to manufacturing equipment. While the exemplary embodiments discussed herein relate to showerhead manufacturing processes, the present disclosure is also applicable to other aspects of semiconductor manufacturing processes, including electrostatic chucks.
[0034] FIG1 is a block diagram illustrating an exemplary system architecture 100 according to certain embodiments. In some embodiments, computer system architecture 100 may be included as part of a manufacturing system for processing substrates (e.g., manufacturing system 200 of FIG2 ). Computer system architecture 100 includes simulation system 110, client device 120, manufacturing equipment 124, metrology equipment 128, and data storage 140. Simulation system 110 may include a digital representation server 190, a server machine 170, one or more machine learning models 180, and a prediction server 112. In various embodiments, computer system architecture 100 may include fewer or more components than depicted in the exemplary figures.
[0035] Manufacturing equipment 124 may include sensors 126 and controller 122, configured to capture data about substrates being processed at the manufacturing system. In some embodiments, manufacturing equipment 124 and sensors 126 may be part of a sensor system, including a sensor server (e.g., a field service server (FSS) at the manufacturing facility) and a sensor identification code reader (e.g., a front-opening pod (FOUP) radio frequency identification (RFID) reader of the sensor system). In some embodiments, metrology equipment 128 may be part of a metrology system, including a metrology server (e.g., a metrology database, a metrology folder, etc.) and a metrology identification code reader (e.g., a FOUP RFID reader of the metrology system).
[0036] Manufacturing equipment 124 can produce products according to recipes or execute tasks over a period of time via controller 122. While multiple controllers may exist for different parts of manufacturing equipment 124, for simplicity, a general controller 122 is depicted in the exemplary diagram. Manufacturing equipment 124 can include a substrate metrology subsystem comprising one or more sensors 126 configured to generate spectral data and / or positional data for a substrate embedded within the substrate metrology subsystem. Sensors 126 configured to generate spectral data (referred to herein as spectral sensing components) can include reflectance sensors, ellipsometry sensors, thermal spectral sensors, capacitive sensors, and the like. In some embodiments, the spectral sensing components can be included within the substrate metrology subsystem or another part of the manufacturing system. One or more sensors 126 (e.g., eddy current sensors) can also be configured to generate non-spectral data for the substrate. Additional details regarding manufacturing equipment 124 and the substrate metrology subsystem are provided with respect to Figures 2 and 3.
[0037] In some embodiments, sensors 126 may provide sensor data associated with manufacturing equipment 124 and provide input to the simulation system 110 and / or the controller of the manufacturing equipment 124. The sensor data may include values for one or more of the following: temperature (e.g., heater temperature, gas temperature, and / or ambient temperature), spacing (SP), pressure, high-frequency radio frequency (HFRF), electrostatic chuck (ESC) voltage, current, flow rate, power, voltage, etc. of one or more substances present in the manufacturing equipment. The sensor data may be associated with or indicate manufacturing parameters, such as hardware parameters, including settings or components of the manufacturing equipment 124 (e.g., size, type, etc.), or process parameters of the manufacturing equipment 124. The sensor data may be provided while the manufacturing equipment 124 is executing a manufacturing process (e.g., equipment readings while processing a product). The sensor data 142 may be different for each substrate.
[0038] Metrology equipment 128 can provide metrology data associated with substrates (e.g., wafers, etc.) processed by manufacturing equipment 124. The metrology data can include one or more of the following: thin film property data (e.g., wafer-to-wafer thin film properties), dimensions (e.g., thickness, height, etc.), dielectric constant, dopant concentration, density, defects, etc. In some embodiments, the metrology data can further include one or more surface profile property data values (e.g., etch rate, etch rate uniformity, critical dimensions of one or more features on the surface of the substrate, critical dimension uniformity across the entire substrate surface, edge placement error, etc.). The metrology data can be for finished or semi-finished products. The metrology data can be different for each substrate.
[0039] Metrology equipment 128 can include both on-tool and off-tool metrology. On-tool metrology can include measurements performed on the devices themselves within the die or on test structures with similar characteristics to these devices. Depending on the measurement technology used, test structures can include, but are not limited to, structures similar to logic or memory devices located on the wafer. In general, on-tool metrology performed on devices or test structures is referred to as "on-board metrology" (OBM).
[0040] OBM can be based on optical measurements (for example, in situ collection of optical emission spectra from a device or test structure, or macroscopic two-dimensional mapping using optical targets) or other types of measurements. These optical or other measurements can be performed inside the chamber (in situ), outside the chamber (ex situ), but still under vacuum, or at a factory interface on a process platform that may have multiple chambers (not necessarily under vacuum). One example is integrated metrology, such as optical critical dimension (OCD). The term "platform" broadly encompasses a system that includes multiple process tools and / or metrology tools, all of which may be identical or some tools may be different from others. For example, a platform may have multiple process chambers that perform the same process, or some chambers that perform one process (e.g., etch) and other chambers that perform a different process (e.g., deposition). In some other embodiments, a platform may include different types of process chambers (e.g., conductor or dielectric etch). A platform may also include metrology tools. Those skilled in the art will understand that the scope of this disclosure is not limited by the configuration of chambers and / or platforms.
[0041] Another type of on-tool metrology is called virtual metrology (VM), which includes data collected from various sensors inside or outside the chamber but on the same platform. VM can include time traces of various process variables such as pressure, temperature, RF power, current, voltage, flow control position, etc.
[0042] Off-tool metrology can include inline metrology (such as e-beam inspection and metrology). "Inline metrology" broadly encompasses measurements that can be performed outside the processing chamber without removing the wafer from the production line. An example of inline metrology is the scanning electron microscope (SEM), whose advanced versions offer high precision and a wide range of modalities. Advanced SEMs can include backscattered electron (BSE) sensors in addition to secondary emission detectors and are capable of measuring electron emission at various tilt angles and landing energies, ranging from a few hundred electron volts to tens of kiloelectron volts. SEMs have the ability to create extensive metrology data libraries in a non-destructive manner. Custom SEM-based inline metrology, including electron beam (e-beam) simulation, data collection, image characterization and feature extraction, and statistical analysis, can be referred to as "custom metrology." Advanced SEM tools can serve as the basis for high-precision, non-destructive three-dimensional feature profiling, which is the core of custom metrology.
[0043] Off-tool metrology can also include data from any additional non-inline or offline metrology (e.g., TEM) that may have been previously performed on a set of similar tools. The digital simulations disclosed herein can use machine learning algorithms to combine metrology data and extract meaningful relationships between the dimensional metrology of interest and the various knobs that control the process.
[0044] The client device 120 may include a computing device such as a personal computer (PC), a laptop, a mobile phone, a smartphone, a tablet, a pocket PC, an Internet-connected television ("Smart TV"), an Internet-connected media player (e.g., a Blu-ray player), a set-top box, an over-the-top (OTT) streaming device, an operator box, etc. In some embodiments, metrology data may be received from the client device 120. The client device 120 may display a graphical user interface (GUI), wherein the GUI enables a user to provide, as input, metrology measurements of substrates processed at the manufacturing system.
[0045] Data storage 140 can be a memory (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or another type of component or device capable of storing data. Data storage 140 can include multiple storage components (e.g., multiple drives or multiple databases) that can span multiple computing devices (e.g., multiple server computers). Data storage 140 can store sensor data 142, performance data 148, library data 154, simulation data 156, and prediction data 158. Sensor data 142 can include historical sensor data 144 and current sensor data 146.
[0046] Performance data 148 may include historical performance data 150 and current performance data 152. Historical sensor data 144 and historical performance data 150 may be historical data associated with previous manufacturing operations. Current sensor data 146 may be data associated with a currently ongoing manufacturing operation for which simulation data 156 and prediction data 158 are to be generated (e.g., to perform corrective actions). Current performance data 152 may also be used for the currently ongoing manufacturing operation and may be used to retrain a previously trained machine learning model.
[0047] The data storage 140 may store data associated with processing a substrate at the fabrication equipment 124. For example, the data storage 140 may store data collected by the sensor 126 at the fabrication equipment 124 before, during, or after a substrate process (referred to as process data).
[0048] Performance data 148 may include data associated with manufacturing equipment 124 and / or products produced by manufacturing equipment 124. In some embodiments, performance data 148 may include an indication of the lifespan (e.g., failure time) of components of manufacturing equipment 124, manufacturing parameters of manufacturing equipment 124, maintenance of manufacturing equipment 124, energy usage of components of manufacturing equipment 124, variations in components of manufacturing equipment 124 (e.g., components with the same part number), or the like.
[0049] Performance data 148 may include an indication of differences in components (e.g., of the same type or part number) within the manufacturing equipment. Performance data 148 may indicate whether these differences (e.g., jitter, slope, peak, etc.) result in product-to-product variations. Performance data 148 may also indicate whether these differences provide improved wafers. Performance data 148 may be associated with the quality of products produced by the manufacturing equipment 124. Metrology equipment 128 may provide performance data 148 (e.g., wafer property data, yield, metrology data) associated with products (e.g., processed wafers) produced by the manufacturing equipment 124. Performance data 148 may include values for one or more of the following: film property data (e.g., wafer-to-wafer film properties), dimensions (e.g., thickness, height, etc.), dielectric constant, dopant concentration, density, defects, etc. Performance data 148 may be data for finished or semi-finished products. Performance data 148 may be different for each product (e.g., each processed wafer). Performance data 148 may indicate whether a product meets a threshold quality (e.g., defective, non-defective, etc.). Performance data 148 may indicate reasons why a threshold quality is not met. In some embodiments, performance data 148 includes historical performance data 150 corresponding to historical property data of the product. Sensor data 142, performance data 148, and library data 154 may be used for supervised and / or unsupervised machine learning.
[0050] Process data may refer to historical process data (e.g., process data generated for a previous substrate processed at the manufacturing system) and / or current process data (e.g., process data generated for a current substrate processed at the manufacturing system). The data storage may also store spectral data or non-spectral data associated with a portion of a substrate processed at the manufacturing equipment 124. The spectral data may include historical spectral data and / or current spectral data. The current process data and / or current spectral data may be data for which predictive data is generated. In some embodiments, the data storage may store metrology data including historical metrology data (e.g., metrology measurements for a previous substrate processed at the manufacturing system).
[0051] The data storage 140 may also store contextual data associated with one or more substrates processed at the manufacturing system. The contextual data may include a recipe name, recipe step number, preventive maintenance indicator, operator, and the like. In some embodiments, the contextual data may also include an indication of differences between two or more process recipes or process steps. For example, a first process recipe may include an operation that sets the internal temperature of a process chamber to 100°C. A second process recipe may include a corresponding operation that sets the internal temperature of the process chamber to 110°C. The contextual data may include an indication of a difference in the internal temperature of the process chamber between the first and second process recipes.
[0052] The simulation system 110 may include a digital representation server 190, a server machine 170, and a prediction server 112. The prediction server 112, the digital representation server 190, and the server machine 170 may each include one or more computing devices, such as a rack server, a router computer, a server computer, a personal computer, a mainframe computer, a laptop computer, a tablet computer, a desktop computer, a graphics processing unit (GPU), an accelerator application-specific integrated circuit (ASIC) (e.g., a tensor processing unit (TPU)), etc.
[0053] Digital representation server 190 may be an algorithmic model that simulates various components of manufacturing equipment 124. For example, digital representation server 190 may be a digital replica (e.g., a digital twin) of a pixelated showerhead portion of manufacturing equipment 124. Digital representation server 190 may use supervised machine learning, semi-supervised learning, unsupervised machine learning, or any combination thereof to generate virtual representations of physical components and the dynamics of how manufacturing equipment 124 operates. Digital representation server 190 may be regularly updated via reinforcement learning using data from sensors 126, sensor data 142, performance data 148, library data 154, and / or any other data associated with generating and maintaining the digital model of manufacturing equipment 124.
[0054] The digital representation server 190 may include a matching network model 192 and a process chamber model 194. The matching network model 192 may be associated with the physical components and dynamics of the matching network and may include information such as a target reference profile and thresholds used to match a simulation to the target reference profile. The process chamber model 194 may be associated with the physical components and dynamics of the process chamber.
[0055] In some embodiments, the digital representation server 190 may generate simulation data 156. The simulation data 156 may include data used to determine how the manufacturing equipment 124 will perform based on current or simulated parameters. The simulation data 156 may further include predicted property data of a digital model of the manufacturing equipment 124 (e.g., a product to be produced or a product already produced). The simulation data 156 may further include predicted metrology data (e.g., virtual metrology data) of the product to be produced or a product already produced. The simulation data 156 may further include indications of anomalies (e.g., an abnormal product, an abnormal component, abnormal manufacturing equipment 124, abnormal energy usage, etc.) and one or more causes of the anomaly. The simulation data 156 may further include indications of the end of life of a component of the manufacturing equipment 124. The simulation data 156 may be comprehensive, encompassing any mechanical and electrical state of the manufacturing equipment.
[0056] The prediction server 112 may include a prediction component 114. In some embodiments, the prediction component 114 may receive simulation data 156 and current performance data 152 (e.g., process chamber flow rate, process chamber pressure, RF power, etc.) and generate outputs (e.g., prediction data 158) for performing corrective actions associated with the manufacturing equipment 124. In some embodiments, the prediction component 114 may use one or more trained machine learning models 180 to determine outputs for performing corrective actions based on one or more of the simulation data 156, the current performance data 152, and the current sensor data 146.
[0057] In some embodiments, prediction component 114 receives simulation data 156 and one or more of current performance data 152 and current sensor data 146, and provides some or all of these data as input to trained machine learning model 180. Prediction component 114 obtains output from trained machine learning model 180 that represents prediction data 158. Trained machine learning model 180 may include a single model or multiple models. In some embodiments, trained machine learning model 180 may use additional data from data store 140 (e.g., library data 154, performance data 148, sensor data 142, etc.), or other metrology data.
[0058] In some embodiments, simulation system 110 further includes a server machine 170. Server machine 170 can use a dataset generator to generate one or more datasets (e.g., a set of data inputs and a set of target outputs) for training, validating, and / or testing machine learning model 180. In particular, server machine 170 can include a training engine 172, a validation engine 174, a selection engine 176, and / or a testing engine 178. Engines (e.g., training engine 172, validation engine 174, selection engine 176, and testing engine 178) can refer to hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, a processing device, etc.), software (e.g., instructions running on a processing device, a general-purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof.
[0059] The training engine 172 can train one or more machine learning models 180 using one or more sets of features associated with a training set from a dataset generator. The training engine 172 can generate multiple trained machine learning models 180, where each trained machine learning model 180 corresponds to a different set of features in the training set (e.g., sensor data from a different set of sensors). For example, a first trained machine learning model can be trained using all features (e.g., X1-X5), a second trained machine learning model can be trained using a first subset of features (e.g., X1, X2, X4), and a third trained machine learning model can be trained using a second subset of features (e.g., X1, X3, X4, and X5), which can partially overlap with the first subset of features.
[0060] The validation engine 174 can validate the trained machine learning model 180 using a set of corresponding features from the validation set from the dataset generator. For example, a first trained machine learning model 180 trained using a first set of features from the training set can be validated using a first set of features from the validation set. The validation engine 174 can determine the accuracy of each of the trained machine learning models 180 based on multiple sets of corresponding features from the validation set. The validation engine 174 can discard trained machine learning models 180 whose accuracy does not meet a threshold accuracy.
[0061] In some embodiments, the selection engine 176 may be capable of selecting one or more trained machine learning models 180 whose accuracy satisfies a threshold accuracy. In some embodiments, the selection engine 176 may be capable of selecting the trained machine learning model 180 with the highest accuracy among the trained machine learning models 180.
[0062] The testing engine 178 can test the trained machine learning model 180 using a corresponding set of features from the test set from the dataset generator. For example, the first trained machine learning model 180 trained using the first set of features from the training set can be tested using the first set of features from the training set. The testing engine 178 can determine the trained machine learning model 180 with the highest accuracy among all the trained machine learning models based on the test set.
[0063] Machine learning model 180 may refer to a model artifact created by training engine 172 using a training set that includes data inputs and corresponding target outputs (correct answers for the corresponding training inputs). Patterns in the data set that map the data inputs to the target outputs (correct answers) may be identified, and a mapping capturing these patterns may be provided to machine learning model 180. Machine learning model 180 may utilize one or more of support vector machines (SVMs), radial basis functions (RBFs), clustering, supervised machine learning, semi-supervised machine learning, unsupervised machine learning, k-nearest neighbor algorithms (k-NNs), linear regression, random forests, neural networks (e.g., artificial neural networks), and the like.
[0064] The prediction component 114 can provide the simulated data 156 and the current sensor data 146 to the trained machine learning model 180 and can run the trained machine learning model 180 on the input to obtain one or more outputs. The prediction component 114 can determine (e.g., extract) prediction data 158 based on the output of the trained machine learning model 180.
[0065] For purposes of illustration and not limitation, aspects of the present disclosure describe training one or more machine learning models 180 and inputting simulation data 156 and sensor data 142 into the one or more trained machine learning models 180 to determine prediction data 158. In other embodiments, a heuristic model or a rule-based model is used to determine prediction data 158 (e.g., without using a trained machine learning model). Prediction component 114 can monitor historical sensor data 144 and / or historical performance data 150. Any information described with respect to the data from data storage 140 can be monitored or otherwise used in a heuristic model or a rule-based model.
[0066] In some embodiments, the functionality of client device 120, prediction server 112, digital representation server 190, and server machine 170 may be provided by a smaller number of machines. For example, in some embodiments, digital representation server 190 and server machine 170 may be integrated into a single machine, while in other embodiments, digital representation server 190, server machine 170, and prediction server 112 may be integrated into a single machine. In some embodiments, client device 120 and prediction server 112 may be integrated into a single machine.
[0067] Generally speaking, functions described in one embodiment as being performed by client device 120, prediction server 112, digital representation server 190, and server machine 170 may also be performed on prediction server 112 in other embodiments, as appropriate. Furthermore, functionality attributed to a particular component may be performed by a different component or components operating together. For example, in some embodiments, prediction server 112 may determine corrective actions based on prediction data 158. In another example, client device 120 may determine prediction data 158 based on output from a trained machine learning model.
[0068] Furthermore, the functionality of a particular component may be performed by a different component or multiple components operating together. One or more of the prediction server 112, the digital representation server 190, and the server machine 170 may be accessed as a service provided to other systems or devices via an appropriate application programming interface (API).
[0069] In some embodiments, a "user" may be represented as a single individual. However, other embodiments of the present disclosure include a "user" that is an entity controlled by multiple users and / or automated sources. For example, a group of individual users united as an administrator group may be considered a "user."
[0070] Embodiments of the present disclosure may be applied to data quality assessment, feature enhancement, model evaluation, virtual metrology (VM), predictive maintenance (PdM), extreme optimization, etc.
[0071] While embodiments of the present disclosure are discussed in terms of generating predictive data 158 to perform corrective actions in a manufacturing facility (e.g., a semiconductor manufacturing facility), embodiments are generally applicable to characterizing and monitoring components. Embodiments are generally applicable to characterizing and monitoring components based on different types of data.
[0072] Client device 120, manufacturing equipment 124, sensor 126, metrology equipment 128, prediction server 112, data storage 140, digital representation server 190, and server machine 170 may be coupled to each other via network 130 to generate prediction data 158 for performing corrective actions.
[0073] In some embodiments, network 130 is a public network that provides client devices 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 devices 120 with access to manufacturing equipment 124, metrology equipment 128, data storage 140, and other privately available computing devices. Network 130 may include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.
[0074] Client device 120 may include a corrective action component 160. Corrective action component 160 may receive user input regarding an indication associated with manufacturing equipment 124 (e.g., via a graphical user interface (GUI) displayed on client device 120). In some embodiments, corrective action component 160 transmits the indication to simulation system 110, receives output from simulation system 110 (e.g., prediction data 168), determines a corrective action based on the output, and causes the corrective action to be implemented. In some embodiments, corrective action component 160 obtains sensor data 142 associated with manufacturing equipment 124 (e.g., from data storage 140, etc.) and provides the sensor data 142 associated with manufacturing equipment 124 to simulation system 110. In some embodiments, corrective action component 160 stores sensor data 142 in data storage 140, and prediction server 112 retrieves sensor data 142 from data storage 140.
[0075] In some embodiments, prediction server 112 can store the output of trained machine learning model 180 (e.g., prediction data 168) in data storage 140, and client device 120 can retrieve the output from data storage 140. In some embodiments, corrective action component 160 receives instructions for corrective actions from simulation system 110 and causes the corrective actions to be implemented. Each client device 120 includes an operating system that allows a user to create, view, and / or edit data (e.g., instructions associated with manufacturing equipment 124, corrective actions associated with manufacturing equipment 124, etc.).
[0076] Performing a manufacturing process that results in a defective product can result in losses in terms of time, energy, product, parts, manufacturing equipment 124, and the costs of identifying the defect and discarding the defective product. By inputting sensor data 142 (e.g., manufacturing parameters being used or to be used to manufacture the product), receiving output from predictive data 168, and performing corrective actions based on predictive data 168, system architecture 100 can provide a technical advantage in avoiding the costs of producing, identifying, and discarding defective products.
[0077] Performing a manufacturing process that causes a component of manufacturing equipment 124 to fail may result in costs such as downtime, product damage, equipment damage, and expedited ordering of replacement parts. By inputting sensor data 142 (e.g., manufacturing parameters being used or to be used to manufacture a product), receiving output from predictive data 168, and performing corrective actions based on predictive data 168 (e.g., predicted operational maintenance, such as replacing, treating, or cleaning a component), system architecture 100 can provide a technical advantage in avoiding the costs of one or more of unexpected component failures, unplanned downtime, lost productivity, unexpected equipment failures, product scrap, and the like.
[0078] The corrective action may be associated with one or more of: computational process control (CPC), statistical process control (SPC) (e.g., performing SPC on an electronic component to determine a controlled process, performing SPC to predict the useful life of a component, performing SPC to compare to a 3-sigma chart, etc.), advanced process control (APC), model-based process control, preventative operational maintenance, design optimization, updating manufacturing parameters, feedback control, machine learning modification, etc. In some embodiments, as discussed in further detail herein, the corrective action may include adjusting one or more controllable variable tuning elements (e.g., knobs).
[0079] In some embodiments, the corrective action includes providing an alert (e.g., providing an alert to stop or not perform the manufacturing process if the predictive data 168 indicates that an anomaly is predicted, such as an anomaly of the product, component, or manufacturing equipment 124). In some embodiments, the corrective action includes providing feedback control (e.g., modifying a manufacturing parameter in response to the predictive data 168 indicating that an anomaly is predicted). In some embodiments, the corrective action includes providing machine learning (e.g., modifying one or more manufacturing parameters based on the predictive data 168). In some embodiments, the execution of the corrective action includes causing an update to one or more manufacturing parameters.
[0080] Manufacturing parameters may include hardware parameters (e.g., replacing components, using certain components, replacing process wafers, updating firmware, etc.) and / or process parameters (e.g., temperature, pressure, flow rate, current, voltage, gas flow, lift speed, etc.). In some embodiments, the corrective action includes initiating preventive maintenance (e.g., replacing, treating, or cleaning components of the manufacturing equipment 124). In some embodiments, the corrective action includes initiating design optimization (e.g., updating manufacturing parameters, manufacturing processes, or manufacturing equipment 124 for an optimized product). In some embodiments, the corrective action includes updating a recipe (e.g., placing the manufacturing equipment 124 in an idle mode, sleep mode, or warm-up mode).
[0081] FIG2 is a schematic top view of an example manufacturing system 200 according to aspects of the present disclosure. Manufacturing system 200 can perform one or more processes on a substrate 202. Substrate 202 can be any suitably rigid, fixed-size, planar article, such as a silicon-containing disk or wafer, a patterned wafer, a glass plate, or the like, suitable for fabricating electronic devices or circuit components thereon.
[0082] Manufacturing system 200 may include a process tool 204 and a factory interface 206 coupled to process tool 204. Process tool 204 may include a housing 208 containing a transfer chamber 210. Transfer chamber 210 may include one or more processing chambers (also referred to as process chambers) 214, 216, and 218 disposed around and coupled to the transfer chamber. Process chambers 214, 216, and 218 may be coupled to transfer chamber 210 via corresponding ports (e.g., slit valves or the like). Transfer chamber 210 may also include a transfer chamber robot 212 configured to transfer substrates 202 between process chambers 214, 216, and 218, a load lock 220, and the like. Transfer chamber robot 212 may include one or more arms, each of which includes one or more end effectors at the end of each arm. The end effectors may be configured to handle specific objects, such as wafers.
[0083] The processing chambers 214, 216, and 218 can be adapted to perform any number of processes on the substrate 202. The same or different substrate processes can be performed in each processing chamber 214, 216, and 218. The substrate processes can include atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, curing, pre-cleaning, metal or metal oxide removal, or the like. In some embodiments, the substrate processes can include a combination of two or more of the following: atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, curing, pre-cleaning, metal or metal oxide removal, or the like. Other processes can be performed therein. Each processing chamber 214, 216, and 218 can include one or more sensors configured to capture data from the substrate 202 before, after, or during a substrate process. For example, the one or more sensors can be configured to capture spectroscopic data and / or non-spectroscopic data from a portion of the substrate 202 during a substrate process. In other or similar embodiments, the one or more sensors may be configured to capture data associated with the environment within the processing chambers 214, 216, 218 before, after, or during substrate processing. For example, the one or more sensors may be configured to capture data associated with the temperature, pressure, gas concentration, etc. of the environment within the processing chambers 214, 216, 218 during substrate processing.
[0084] A load lock 220 may also be coupled to the housing 208 and the transfer chamber 210. The load lock 220 may be configured to interface and couple with the transfer chamber 210 on one side and interface and couple with the factory interface 206 on an opposite side. In some embodiments, the load lock 220 may have an environmentally controlled atmosphere that can change from a vacuum environment (in which substrates can be transferred to and from the transfer chamber 210) to an inert gas environment at or near atmospheric pressure (in which substrates can be transferred to and from the factory interface 206). The load lock 220 may include one or more sensors to capture data associated with the substrates 202 before, after, or during substrate processing in the processing chambers 214, 216, and 218. For example, the load lock 220 may include a vibration sensor (e.g., a piezoelectric sensor) configured to detect and monitor the amount of vibration occurring within the load lock 220 during substrate processing in the processing chambers 214, 216, and 218. In another example, the load lock 220 may include a temperature sensor (e.g., an infrared camera) to detect and monitor the temperature of the load lock 220 during substrate processing. It should be noted that while the embodiments of the present disclosure describe sensors such as vibration sensors or temperature sensors at the load lock 220 for monitoring the status of the load lock 220 during substrate processing, any type of sensor may be used at any station in the manufacturing system to monitor the status and / or health of the station during substrate processing.
[0085] The factory interface 206 can be any suitable housing, such as an equipment front-end module (EFEM). The factory interface 206 can be configured to receive substrates 202 from substrate carriers 222 (e.g., front-opening pods (FOUPs)) docked at various load ports 224 of the factory interface 206. A factory interface robot 226 (shown in phantom) can be configured to transfer substrates 202 between the carriers 222 (also known as containers) and the load lock 220.
[0086] The carrier 222 may be a substrate storage carrier or a replacement part storage carrier. According to previously described embodiments, the factory interface 206 and / or the carrier 222 may include one or more sensors (e.g., vibration sensors, temperature sensors, etc.) to capture data associated with the substrate 202 before, after, or during substrate processing in the processing chambers 214, 216, and 218.
[0087] The manufacturing system 200 may also be connected to a client device (e.g., client device 120 of FIG. 1 ) configured to provide information to a user (e.g., an operator) regarding the manufacturing system 200. In some embodiments, the client device may provide information to a user of the manufacturing system 200 via one or more graphical user interfaces (GUIs). For example, the client device may provide information regarding one or more modifications to be made to a process recipe for the substrate 202 via the GUI.
[0088] Manufacturing system 200 may also include a system controller 228. System controller 228 may be and / or include a computing device such as a personal computer, a server computer, a programmable logic controller (PLC), a microcontroller, or the like. System controller 228 may include one or more processing devices, which may be general-purpose processing devices such as a microprocessor, a central processing unit, or the like. More specifically, the processing device may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or a processor that implements another instruction set or a combination of instruction sets. The processing device may also be one or more special-purpose processing devices such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, or the like. System controller 228 may include data storage devices (e.g., one or more disk drives and / or solid-state drives), main memory, static memory, a network interface, and / or other components. System controller 228 may execute instructions to implement any one or more of the methodologies and / or embodiments described herein. In some embodiments, the system controller 228 may execute instructions to perform one or more operations at the manufacturing system 200 according to a process recipe. The instructions may be stored on a computer-readable storage medium, which may include main memory, static memory, secondary storage, and / or a processing device (during execution of the instructions).
[0089] The system controller 228 can receive data from sensors included on or within various parts of the fabrication system 200 (e.g., the processing chambers 214, 216, 218, the transfer chamber 210, the load lock 220, etc.). In some embodiments, the data received by the system controller 228 can include spectroscopic data and / or non-spectroscopic data of a portion of the substrate 202. In other or similar embodiments, the data received by the system controller 228 can include data associated with processing the substrate 202 in the processing chambers 214, 216, 218, as previously described. For the purposes of this description, the system controller 228 is described as receiving data from sensors included within the processing chambers 214, 216, 218. However, the system controller 228 can receive data from any part of the fabrication system 200 and can use the data received from that part in accordance with the embodiments described herein. In one illustrative example, the system controller 228 can receive data from one or more sensors in the processing chambers 214, 216, 218 before, after, or during substrate processing in the processing chambers 214, 216, 218. The data received from the sensors in various parts of the manufacturing system 200 can be stored in a data storage 250. The data storage 250 can be included as a component within the system controller 228, or can be a separate component from the system controller 228. In some embodiments, the data storage 250 can be the data storage 140 described with respect to FIG.
[0090] The manufacturing system 200 may further include a substrate measurement subsystem 240. The substrate measurement subsystem 240 may obtain spectral measurements of one or more portions of the substrate 202 using one or more metrology tools discussed herein before or after the substrate 202 is processed at the manufacturing system 200. In some embodiments, the substrate measurement subsystem 240 may obtain spectral measurements of the one or more portions of the substrate 202 in response to receiving a request for spectral measurements from the system controller 228. The substrate measurement subsystem 240 may be integrated within a portion of the manufacturing system 200. In some embodiments, the substrate measurement subsystem 240 may be integrated within the factory interface 206. In other or similar embodiments, the substrate measurement subsystem 240 may not be integrated within any portion of the manufacturing system 200 but may be a standalone component. In such embodiments, the substrate 202 measured at the substrate measurement subsystem 240 may be transferred to and from a portion of the manufacturing system 200 before or after the substrate 202 is processed at the manufacturing system 200.
[0091] The substrate metrology subsystem 240 can obtain spectral measurements of a portion of the substrate 202 by generating spectral data for the portion of the substrate 202. In some embodiments, the substrate metrology subsystem 240 is configured to generate spectral data, non-spectral data, positional data, and other substrate property data of the substrate 202 (e.g., thickness of the substrate 202, width of the substrate 202, etc.). After generating data for the substrate 202, the substrate metrology subsystem 240 can transmit the generated data to the system controller 228. In response to receiving the data from the substrate metrology subsystem 240, the system controller 228 can store the data in the data storage 250.
[0092] FIG3 depicts a schematic cross-sectional view of an exemplary processing chamber 300. As shown, the exemplary processing chamber 300 is configured as a plasma-enhanced chemical vapor deposition (PECVD) chamber. However, in some embodiments, the processing chamber 300 can also be configured to perform another plasma-enhanced processing operation (e.g., etching or physical vapor deposition) or a processing operation that does not involve plasma (e.g., chemical vapor deposition). The processing chamber 300 includes a chamber body 302, a substrate support 304 disposed within the chamber body 302, and a lid 306 coupled to the chamber body 302 and enclosing the substrate support 304 within a processing volume 320. The substrate support 304 is configured to support a substrate 354 thereon during processing. The substrate 354 is provided to the processing volume 320 via an opening 326.
[0093] As shown, showerhead 312 includes a plurality of individually tunable electrodes 358. Showerhead 312 is coupled to power source 328 via a matching circuit. In the exemplary embodiment, power source 328 is a radio frequency (RF) power source electrically coupled to electrode 308. Furthermore, power source 328 provides approximately 300 watts to approximately 3,000 watts at a frequency of approximately 50 kHz to approximately 13.6 MHz. In some embodiments, power source 328 can be pulsed during various operations. Although not explicitly depicted, in some embodiments, electrodes 358 can be coupled to one or more power sources other than power source 328. These one or more power sources can be RF power sources. For example, each electrode 358 can be coupled to a different power source to provide different power to each electrode 358. The electrodes 358 and corresponding power sources (e.g., power source 328) facilitate control of the plasma formed within processing volume 320.
[0094] The substrate support 304 comprises or is formed from one or more metal or ceramic materials. Exemplary metal or ceramic materials include one or more metals, metal oxides, metal nitrides, metal oxynitrides, or any combination thereof. For example, the substrate support 304 may comprise or be formed from aluminum, aluminum oxide, aluminum nitride, aluminum oxynitride, or any combination thereof.
[0095] As shown, one or more electrodes 322 are embedded within the substrate support 304, but may also be coupled to a surface of the substrate support 304. The electrodes 322 are coupled to a power source 336. In various embodiments, the power source 336 may be a DC power source, a pulsed DC power source, a radio frequency (RF) power source, a pulsed RF power source, or any combination thereof. The power source 336 is configured to drive the electrodes 322 with a drive signal to generate plasma within the processing volume 320. It is contemplated that the drive signal may be one of a DC signal and a varying voltage signal (e.g., an RF signal). Furthermore, the electrodes 322 may alternatively be coupled to a power source 328 instead of the power source 336, or the power source 336 may be omitted. In some embodiments, the power source 328 is configured to drive a showerhead electrode 358 with the drive signal to generate plasma within the processing volume 320. In one embodiment, the showerhead electrode 358 is part of an RF field generator.
[0096] Plasma is generated in the processing volume 320 via power supply 328 and power supply 336. A radio frequency field is generated by driving at least one of the showerhead electrode 358 and the electrode 322 with a drive signal to promote the formation of plasma within the processing volume 320. The presence of the plasma facilitates processing of the substrate 354, such as depositing a thin film onto the surface of the substrate 354 or etching material from the surface of the substrate 354. In some embodiments, some electrodes 322 and / or showerhead electrodes 358 can be individually controlled to adjust the formation of plasma within the processing volume 320. That is, some electrodes can be driven with greater power than other electrodes to adjust the amount or composition of plasma generated in localized sections of the processing volume 320. This individual control of electrodes or regions of electrodes can be used to adjust the uniformity or composition of the generated plasma throughout the processing volume 320.
[0097] The exhaust port 356 is coupled to a vacuum pump 357. The vacuum pump 357 removes excess process gases or byproducts from the processing volume 320 via the exhaust port 356 during and / or after processing.
[0098] The gas supply 311 includes one or more gas sources. The gas supply 311 is configured to deliver the one or more gases from the one or more gas sources to the processing volume 320 through the showerhead 312. Each of the one or more gas sources provides a process gas, such as silane, disilane, tetraethyl orthosilicate (TEOS), germane, metal halides (such as titanium tetrachloride, tantalum pentachloride, and tungsten hexafluoride), organometallics (such as tetrakis(dimethylamido) titanium and pentakis(dimethylamido) tantalum), ammonia, oxygen (O 2 ), hydrogen peroxide, hydrogen, diborane, chlorine (Cl 2 ), sulfur hexafluoride, hydrocarbons (typically C x H y ), and the like.
[0099] In some embodiments, the process gas may be ionized to form a plasma within the processing volume 320. For example, one or more of a carrier gas and an ionizable process gas may be provided to the processing volume 320 to process the substrate 354. When processing a 300 mm substrate, the process gas may be introduced into the processing chamber 300 at a flow rate of approximately 6,500 sccm to approximately 8,000 sccm, approximately 100 sccm to approximately 10,000 sccm, or approximately 100 sccm to approximately 1,000 sccm. Alternatively, other flow rates may be utilized. In some examples, a remote plasma source may be used to deliver the plasma to the processing chamber 300 and may be coupled to the gas supply 311.
[0100] Showerhead 312 has an opening 318 for receiving one or more process gases from gas supply 311 into processing volume 320. Process gases are supplied to processing chamber 300 via gas feed 314, and enter plenum 316 before flowing through opening 318. In some embodiments, different process gases that flow simultaneously during processing operations enter processing chamber 300 through separate gas feeds and separate plenums before entering processing chamber 320 through showerhead 312.
[0101] The flow of gas through the openings 318 of the showerhead 312 is regulated by one or more micro-electromechanical systems (MEMS devices) disposed in the showerhead 312. In some embodiments, the flow of gas through individual openings 318 and / or through a group of openings 318 (i.e., a zone) can be regulated by a single MEMS device. In some embodiments, the flow of gas through all openings 318 is regulated by a plurality of MEMS devices. In one example, each MEMS device regulates the flow of gas through one or more openings 318, such that the flow of gas through any opening 318 is regulated by a corresponding MEMS device.
[0102] The MEMS device regulates the gas flow rate by allowing a maximum gas flow rate through an individual opening 318 and / or through a group of openings 318. The MEMS device regulates the gas flow rate by preventing any gas from flowing through an individual opening 318 and / or through a group of openings 318. The MEMS device regulates the gas flow rate by controlling the gas flow rate through an individual opening 318 and / or through a group of openings 318 so that the gas flow rate is greater than zero and less than the maximum gas flow rate through the individual opening 318 and / or through the group of openings 318.
[0103] FIG4 is a schematic plan view of an exemplary faceplate 400 of a showerhead, such as showerhead 312 of process chamber 300. Regions 410 are defined on faceplate 400, where each region 410 includes one or more openings 412 through faceplate 400. In operation, the flow of process gas through the one or more openings 412 of a particular region 410 is controlled by one or more MEMS devices associated with that particular region 410. In operation, each region 410 has associated with it one or more devices (e.g., MEMS devices and / or heaters), which are controlled individually and / or in groups, as described herein. In one example, the operation of each MEMS device or heater can be controlled without changing the operating state of another MEMS device or heater associated with faceplate 400. In various embodiments, the number, size, and distribution of regions 410 can be configured based on the type of process to be performed in process chamber 300 and / or the specific process gases to be employed.
[0104] A controller (e.g., controller 122 of FIG. 1 ) can control the distribution of gas through one or more openings 412 or zones 410. The controller includes a central processing unit (CPU), a memory containing instructions, and support circuits for the CPU. The controller can be any form of computer processor used in industrial environments to control various chambers and equipment and / or subprocessors therein.
[0105] Memory, or non-transitory computer-readable media, is one or more readily available memories, such as random access memory (RAM), read-only memory (ROM), floppy disks, hard disks, flash drives, or any other form of local or remote digital storage. Support circuits are coupled to the CPU to support the CPU (processor). Support circuits include cache, power supplies, clock circuits, input / output circuitry, and subsystems. Operations and operating parameters are stored in the memory as software routines that are executed or called to transform the controller into a specific-purpose controller to control the operation of any portion of showerhead 312 or the entire showerhead 312, such as power and gas flow distribution. The controller is configured to perform any of the operations described herein. When executed, the instructions stored in the memory cause one or more of the operations described herein to be performed.
[0106] A controller can be designated to control the functionality of any portion of the processing chamber 300 or the entire processing chamber 300. As discussed herein, the controller can control the operation of the pixelated showerhead by controlling at least the power distributor and / or gas distributor of the pixelated showerhead. The power distributor controls the amount of power delivered to each power conductor of the pixelated showerhead by, for example, controlling the showerhead's electrical switches. The gas flow distributor controls the settings of the gas flow meter and the individual control of the gas valves coupled to the gas supplies of different gas species. The controller can control any one or more parameters controlled by the gas distributor.
[0107] In some embodiments, data from any of the sensors discussed herein and / or any sensors specifically associated with showerhead 312 can be used to provide feedback to the controller. For example, the data can include pressure data and / or temperature data.
[0108] The controller uses the information thus provided as input to process commands addressed to individual MEMS devices and / or groups of MEMS devices. In some such embodiments, these commands cause the individual MEMS devices and / or groups of MEMS devices to adjust the flow rate of a gas through the individual MEMS devices and / or groups of MEMS devices. In one example, such adjustment provides control over the flow distribution of one or more gases through showerhead 312. In some such embodiments, these commands cause the individual MEMS devices and / or groups of MEMS devices to adjust the temperature of a gas flowing through the individual MEMS devices and / or groups of MEMS devices. In one example, such adjustment provides control over the temperature distribution of one or more gases through showerhead 312.
[0109] The instructions in the controller's memory may include one or more machine learning models (e.g., model 180 of FIG. 1 ), which may be executed in addition to the operations described herein. As an example, the machine learning algorithm executed by the controller may optimize and change operating parameters based on one or more sensor measurements taken by the one or more sensors. The operating parameters may include process parameters such as pressure, temperature, gas flow rate, valve member position, and heater status.
[0110] As described herein, the one or more machine learning models can take into account variations in parameters such as gas flow rate, gas ratio, temperature, and pressure across the showerhead 312. In some embodiments, the one or more machine learning algorithms can measure any of these parameters to determine whether the appropriate amount of process gas at the appropriate pressure and temperature is being delivered to the appropriate area of the processing volume in accordance with the specified operation. In some embodiments, the one or more machine learning algorithms can cause the controller to initiate corrective actions to adjust the amount, pressure, or temperature of the process gas being delivered to the specified area of the processing volume.
[0111] The operational flexibility facilitated by the showerheads of the present disclosure is not available in conventional showerheads. Conventional showerheads are typically configured to achieve optimal distribution of a specific process gas by having a specific number of openings of a specific size arranged in a specific pattern. However, the number, size, and arrangement of openings that provide optimal distribution of one process gas does not necessarily provide optimal distribution of different process gases. In contrast, the showerheads of the present disclosure provide for the ability to tailor the relative flow rates of any process gas through an opening or group of openings to provide optimal distribution of any process gas within the processing chamber.
[0112] The operational flexibility facilitated by the showerhead of the present disclosure provides time and cost efficiencies compared to conventional operations. For example, some conventional processing sequences involve performing a first operation on a substrate in a first processing chamber using a first process gas, then transferring the substrate to a second processing chamber, and then performing a second operation on the substrate using a second, different process gas. In contrast, a processing chamber incorporating the showerhead of the present disclosure can be configured to perform a first operation using a first process gas and then a second operation using a second process gas. Consequently, the need for a second processing chamber is alleviated, as is the time required to transfer substrates between processing chambers.
[0113] Furthermore, the operational flexibility facilitated by the showerheads of the present disclosure provides inventory management efficiencies compared to conventional operations. Using the showerheads of the present disclosure can reduce or eliminate the need to stock different showerheads configured for use with different process gases for different process operations.
[0114] FIG5 is a schematic, partial cross-sectional side view detailing portions of a substrate support assembly 500 configured to provide azimuthal tuning of the temperature distribution across the substrate support assembly. The azimuthal tuning of the temperature distribution across the substrate support assembly 500 can be controlled by software routines running on the controller 122. The software routines can additionally or alternatively be stored and / or executed by a second controller (not shown) remote from the multi-chamber vacuum processing system, such as located in the processing chamber 300 or an advanced process controller (APC) 564.
[0115] The substrate support assembly 500 generally includes at least a substrate support 510. The substrate support 510 can be a vacuum chuck, an electrostatic chuck, a carrier, or other workpiece support surface. In one embodiment, the substrate support 510 is an electrostatic chuck, which will be described below as the electrostatic chuck 510. The substrate support assembly 500 can also include a cooling base 560. The cooling base 560 can alternatively be detachable from the substrate support assembly 500. The substrate support assembly 500 can be removably coupled to a support pedestal 505. The support pedestal 505 can include a pedestal base 544. The substrate support assembly 500 can be periodically removed from the support pedestal 505 to allow for refurbishment of one or more components of the substrate support assembly 500.
[0116] The electrostatic chuck 510 has a mounting surface 503 and a workpiece support surface 502 opposite the mounting surface 503, with the substrate 202 removably disposed on the workpiece support surface 502. The electrostatic chuck 510 generally includes a chucking electrode 507 embedded in a dielectric body 508. Although the chucking electrode 507 is shown near the mounting surface 503 of the electrostatic chuck 510, the chucking electrode 507 may also be embedded in other portions of the electrostatic chuck 510, such as just below the workpiece support surface 502. The chucking electrode 507 may be configured as a monopolar or bipolar electrode, or in other suitable arrangements. The chucking electrode 507 is coupled to a chucking power supply 506 via an RF filter 504, which provides RF power or DC power to electrostatically secure the substrate 202 to the workpiece support surface 502 of the electrostatic chuck 510. The RF filter 504 prevents the RF power used to form a plasma within a processing chamber (e.g., processing chamber 300 of FIG. 3 ) from damaging electrical equipment or creating electrical hazards outside the chamber.
[0117] The dielectric body 508 of the electrostatic chuck 510 can be fabricated from a ceramic material, such as AlN or Al2O3. Alternatively, the dielectric body 508 can be fabricated from a polymer, such as polyimide, polyetheretherketone, polyaryletherketone, and the like. A heater can be embedded within the dielectric body 508. The dielectric body 508 can include one or more primary resistive heaters 512 and / or a plurality of spatially tunable heaters 514. The primary resistive heaters 512 can be provided to raise the temperature of the substrate support assembly 500 to a temperature used for performing chamber processes (e.g., processing the substrate 202 and / or cleaning the interior of the processing chamber 300). The primary resistive heaters 512 can be configured to provide any one or more laterally separated heating zones, such as a plurality of concentric annular zones.
[0118] The spatially tunable heater 514 is complementary to the primary resistive heater 512 and is configured to adjust the local temperature of the electrostatic chuck 510 in a plurality of discrete locations (or zones) within any one or more of the plurality of laterally separated heaters defined by the primary resistive heater 512. For example, the spatially tunable heaters 514 can be arranged in a polar array, a Cartesian grid, a hexagonal grid, or other suitable matrix. Thus, the spatially tunable heaters 514 provide for local adjustment of the temperature distribution of a substrate 202 positioned on the substrate support assembly 500. Thus, the primary resistive heater 512 is used to maintain the temperature distribution of the workpiece support surface 502 on an overall macroscale, while the spatially tunable heaters 514 are used to adjust the temperature at discrete locations of the temperature distribution of the workpiece support surface 502 on a local microscale.
[0119] The main resistive heater 512 and the spatially tunable heater 514 are coupled to a heater power supply 522 via an RF filter 518. The heater power supply 522 can provide 900 watts or more of power to the heaters 512, 514. The controller 122 (which, in some embodiments, can be the same as the controller 122 of FIG. 1 ) can control the operation of the heater power supply 522, which is generally configured to provide power to the individual heaters 512, 514 to heat the substrate 202 to a predetermined temperature profile. In one embodiment, the main resistive heater 512 includes laterally separated heating zones, wherein the controller 122 enables one zone of the main resistive heater 512, or even a single spatially tunable heater 514, to be preferentially heated relative to adjacent heaters 512, 514. In some embodiments, each spatially tunable heater 514 can be independently controlled to provide a temperature that is different from the temperature of another spatially tunable heater 514. In some embodiments, a plurality (e.g., at least two or up to all) of the spatially tunable heaters 514 are independently powered simultaneously to provide a very stable temperature distribution without jumps, thereby forming a stable and easily controllable temperature distribution, which helps to enhance the uniformity and predictability of substrate processing results.
[0120] The electrostatic chuck 510 can include one or more temperature sensors 554. The temperature sensors 554 can measure the temperature at a plurality of discrete locations on the workpiece support surface 502. The temperature sensors 554 can provide temperature feedback information to the controller 122 to control the power applied by the heater power supply 522 to the primary resistive heater 512 and the spatially tunable heater 514. In addition, the feedback information can be used to control the operation of the cooling base 560.
[0121] The electrostatic chuck 510 can be disposed on a temperature-controlled cooling base 560. The temperature-controlled cooling base 560 is coupled to a thermally conductive fluid source 562. The thermally conductive fluid source 562 provides a thermally conductive fluid (e.g., a liquid, a gas, or a combination thereof) that circulates through one or more conduits 590 disposed in the cooling base 560. The controller 122 can control the flow of fluid through isolated adjacent conduits 590 to achieve localized control of heat transfer between different areas of the electrostatic chuck 510 and the cooling base 560, which helps control the lateral temperature distribution of the substrate 202.
[0122] The pedestal base 544 is disposed beneath the cooling base 560 and is configured to house a plurality of drive mechanisms configured to raise and lower a plurality of lift pins. Furthermore, the pedestal base 544 is configured to accommodate a plurality of fluid connections from the electrostatic chuck 510 and the cooling base 560. The pedestal base 544 is also configured to accommodate a plurality of electrical connections from the electrostatic chuck 510. A myriad of connections (e.g., fluid, electrical, data signals) can operate externally or internally to the substrate support assembly 500.
[0123] The system controller 122 can include a controller board 550, which can be disposed in the pedestal base 544. Alternatively, the controller board 550 can be disposed elsewhere within or outside the substrate support assembly 500. The controller board 550 can include a pulse width modulation (PWM) heater controller 516. The controller board 550 can also include an optical communication interface board 556. The controller board 550 can also optionally include a temperature sensor controller 552.
[0124] The temperature sensor controller 552 can be communicatively coupled to a temperature sensor 554 for measuring the temperature of the substrate 202. The temperature sensor controller 552 can convert the signal from the temperature sensor 554 into actual temperature readings at discrete locations around the workpiece support surface 502. The controller board 550 can also optionally include a metrology controller (not shown) to determine the thickness of the substrate 202 disposed on the workpiece support surface 502.
[0125] A PWM heater controller 516 can be connected to the heaters 512 and 514. The PWM heater controller 516 can modify the power supplied from the heater power supply 522 to the individual heaters 512 and 514. The PWM heater controller 516 can be programmed and calibrated by measuring the temperature at each spatially tunable heater 514. That is, each spatially tunable heater 514 has its own independent PWM control. The PWM heater controller 516 can control the temperature by adjusting the power parameters for the individual spatially tunable heaters 514. For example, providing 100% power to one of the spatially tunable heaters 514 can generate a heat output of approximately 5 degrees Celsius, increasing the temperature set by the primary heater for the surface location above the spatially tunable heater 514. Furthermore, providing 20% power to one or more spatially tunable heaters 514 can generate a heat output of approximately 1 degree Celsius, increasing the temperature set by the primary heater for the surface location above the spatially tunable heater 514. The power to several spatially tunable heaters 514 can be reduced to 20% and maintained at that level throughout the processing operation. Furthermore, the PWM heater controller 516 can operate one or more spatially tunable heaters 514 at a first power level (e.g., 80%) while simultaneously operating one or more individual spatially tunable heaters 514 at a second power level (e.g., approximately 20%). In one embodiment, the temperature can be adjusted by incrementally increasing the power to the spatially tunable heaters 514. For example, during the course of a process, a temperature increase can be achieved by increasing the power supplied to the spatially tunable heaters 514 by a percentage (e.g., a 9% increase). In another embodiment, the temperature can be adjusted by cycling the spatially tunable heaters 514 on and off while also cycling the other tunable heaters 514 on and off (either in phase or out of phase). In another embodiment, the temperature can be adjusted by a combination of adjusting the power levels as needed, cycling, and incrementally adjusting the power to the spatially tunable heaters 514 to maintain the desired temperature profile.
[0126] By varying the heat output of each individual spatially tuned heater 514 at once, a temperature map can be generated. This map can correlate the CD or substrate temperature distribution with the power profile of each spatially tunable heater 514. Thus, the spatially tunable heaters 514 can be used to generate a temperature distribution across the substrate based on a program that adjusts the power settings of the individual spatially tunable heaters 514. This logic can be placed directly in the PWM heater controller 516 or in an externally connected controller (e.g., controller 122). Thus, the PWM heater controller 516 is configured to independently control the output of one of the plurality of spatially tunable heaters 514 relative to another of the plurality of spatially tunable heaters 514 and the primary resistive heater 512.
[0127] The optical communication interface plate 556 can interface with the PWM heater controller 516. The optical communication interface plate 556 can also interface with other controllers, such as the temperature sensor controller 552. The optical communication interface plate 556 can have an optical connection to an optical converter 558. The optical interface plate 556 can send optical signals to the optical converter 558 to communicate with the controller 122. Thus, communication between the PWM heater controller 516 and the controller 122 can be performed using optical signals that do not contain harmful RF energy. For example, the controller 122 can send signals to control the heat output from one or more spatially tunable heaters 514 to adjust the temperature profile of the substrate 202 positioned on the electrostatic chuck 510. Signal commands from the controller 122 can enter the optical converter 558 for transmission to the optical communication interface plate 556, which then transmits the commands to the PWM heater controller 516 to control the heating profile of the substrate 202 during processing. One benefit of the optical communication interface plate 556 is that it prevents RF power from escaping the substrate support assembly 500 via the control circuitry.
[0128] The surface temperature of the substrate 202 in the processing chamber 300 may be affected by factors such as pumping process gases, the slit valve, the plasma, and other factors. The controller 122 may maintain a temperature profile of the substrate 202 to achieve high-quality processing results. The controller 122 may receive input from metrology equipment or other processing equipment to correct for deviations or unfavorable trends in the process results of the substrate 202. For example, one region of the substrate 202 may etch faster than another region of the substrate 202. The controller 122 may signal the spatially tunable heater 514 to adjust the surface temperature of the substrate 202 in regions with etch rate deviations. The controller 122 may activate the spatially tunable heater 514 in a pattern that resembles the shape of these regions. The spatially tunable heater 514 improves the temperature distribution on the surface of the substrate 202 produced by the primary resistive heater 512 by reducing the variation in the temperature distribution to approximately + / - 0.3 degrees Celsius. By using the spatially tunable heater 514, the temperature distribution can be made uniform or precisely varied in a predetermined manner across regions of the substrate 202 to achieve desired results. Using independently controllable spatially tunable heaters 514 to smooth or correct the temperature distribution generated by the main resistive heaters 512 enables local temperature uniformity across the substrate to be controlled to very tight tolerances, thereby enabling precise process and CD control when processing the substrate 202. Furthermore, the small size and high density of the spatially tunable heaters 514 relative to the main resistive heaters 512 enables temperature control at specific locations on the substrate support assembly without substantially affecting the temperature of adjacent areas, thereby compensating for hot and cold spots without introducing skew or other temperature asymmetries.
[0129] It is contemplated that in a given embodiment of a substrate support assembly 500 configured for use with a 300 mm substrate, there may be from about 200 to about 400, or even more, spatially tunable heaters 514. For substrates 450 mm or larger, it is contemplated that even more spatially tunable heaters 514 may be used. Exemplary distributions of spatially tunable heaters 514 are further described below with reference to FIG. 6A-6D .
[0130] Figures 6A-6D are top views of an electrostatic chuck 510, illustrating an example layout of cells 690 having spatially tunable heaters 514 therein. Figure 6A is a top view of the electrostatic chuck 510 of Figure 5, according to one embodiment. Figures 6B-6D are top views of the electrostatic chuck 510 of Figure 5, according to an alternative embodiment. Cells 690 may have material (or gaps) that act as thermal barriers 616 between adjacent cells 690. Thermal barriers 616 separate and reduce conduction between adjacent cells 690. Thus, by individually and independently controlling the power provided to each spatially tunable heater 514 in each cell 690, a method of localized temperature control is achieved that enables specific points on the substrate 202 to be heated or cooled, thereby achieving truly addressable lateral temperature profile tuning and control over the surface of the substrate 202.
[0131] Referring now to FIG. 6A , a thermal block 616 is positioned between each adjacent unit 690. Each unit 690 has at least one spatially tunable heater 514 associated therewith. The number of units 690 shown is for illustration only; any number of embodiments may have substantially more (or fewer) units 690. The number of spatially tunable heaters 514 may be at least an order of magnitude greater than the number of primary resistive heaters 512. The number of spatially tunable heaters 514 located throughout the substrate support assembly 500 may easily exceed several hundred.
[0132] Spatially tunable heaters 514 can be configured in a pattern 699 to efficiently generate heat distribution along the surface of electrostatic chuck 510. Pattern 699 can be symmetrical about midpoint 692 while providing gaps in and around holes 622 for lift pins or other mechanical, fluid, or electrical connections. Each spatially tunable heater 514 can be controlled by controller 122 via PWM controller 516 ( FIG. 5 ). PWM controller 516 can activate a single spatially tunable heater 514 defining a localized area 640, or multiple spatially tunable heaters 514 combined to define an inner wedge 662, a peripheral group 664, a pie-shaped area 630, or other desired geometric configurations (including non-contiguous configurations) to determine the desired temperature distribution. In this manner, the temperature can be precisely controlled at individual locations along the surface of electrostatic chuck 510, not limited to concentric rings or other center-to-edge configurations known in the art. While the illustrated pattern 699 includes discrete smaller cells, the pattern 699 may alternatively have larger and / or smaller cells, extend to the edge, or have other forms. In the embodiment depicted in FIG6A , the pattern 699 of the spatially tunable heater 514 is arranged in a polar array about a midpoint 692, which in some embodiments may coincide with the centerline of the electrostatic chuck 510.
[0133] FIG6B is a top view of the electrostatic chuck 510 depicted in FIG5 , according to another embodiment. The spatially tunable heaters 514 are arranged in a grid pattern, and thus the array defining the temperature control units 690 is also arranged in a grid pattern. While the grid pattern of the spatially tunable heaters 514 is shown as an X / Y (Cartesian) grid comprising columns and rows, the grid pattern of the spatially tunable heaters 514 may alternatively have some other uniformly packed form, such as a hexagonal, tightly packed pattern. It should be understood that, as discussed above, the spatially tunable heaters 514 can be activated in groups or individually.
[0134] FIG6C is a top view of the electrostatic chuck 510 depicted in FIG5 , according to another embodiment. FIG6C shows a plurality of spatially tunable heaters 514 arranged in a polar array. Optionally, one or more thermal blocks 616 may be positioned between the spatially tunable heaters 514 . The polar array pattern of the spatially tunable heaters 514 defines adjacent cells 690 , which are then also arranged in a polar array. Optionally, thermal blocks 616 may be used to isolate each cell 690 from adjacent cells 690 .
[0135] FIG6D is a top view of the electrostatic chuck 510 depicted in FIG5 , according to another embodiment. FIG6D shows a plurality of spatially tunable heaters 514 arranged in concentric channels. The concentric channel pattern of spatially tunable heaters 514 may optionally be separated by thermal blocks 616. It is contemplated that the spatially tunable heaters 514 and units 690 may be arranged in other orientations.
[0136] The number and density of spatially tunable heaters 514 facilitates controlling temperature uniformity across the substrate to within very tight tolerances, enabling precise process and CD control when processing substrates 202. Furthermore, individually controlling one spatially tunable heater 514 relative to another spatially tunable heater 514 enables temperature control at discrete locations along the surface of the substrate support assembly 500 without substantially affecting the temperature of adjacent or neighboring areas of the surface. This feature allows compensation for localized hot and cold spots without introducing skew or other temperature asymmetries. The spatially tunable heaters 514 can have individual temperature ranges between approximately 0.0 degrees Celsius and approximately 10.0 degrees Celsius, and can control temperature increases in increments of approximately 0.1 degrees Celsius. In one embodiment, the plurality of spatially tunable heaters 514 in the substrate support assembly 500, in combination with the primary resistive heater 512, have demonstrated the ability to control the temperature uniformity of substrates 202 processed thereon to less than approximately ±0.3 degrees Celsius. Thus, the spatially tunable heater 514 allows for both lateral and azimuthal tuning of the lateral temperature profile of a substrate 202 being processed on the substrate support assembly 500 .
[0137] The heaters 514, 512 may be controlled by software routines. The software may incorporate current substrate CD measurements and process results, as well as process recipes and other parameters, to tune the temperature profile of the substrate 202 undergoing processing in the processing chamber 300.
[0138] Figure 7 depicts a simplified timing diagram 700 of a physical manufacturing process with concurrent real-time simulation. In timing diagram 700, a machine controller 705 inputs process parameters into a process chamber 710 to implement a specific manufacturing process. As discussed herein, the manufacturing process can be one that utilizes a pixelated showerhead, an electrostatic chuck, or one that generates a pixelated plasma. The machine controller 705 also simultaneously sends the process parameters to a digital twin 715 of the process chamber. As discussed herein, sensor data is also transmitted from the process chamber 710 to the digital twin 715.
[0139] The process chamber then executes the manufacturing process at block 720. Simultaneously, the digital twin model 715 performs a real-time simulation 725 of the executing process based at least in part on some or all sensor inputs received from the process chamber, a model of the process chamber, one or more trained machine learning models, and metrology data. Finally, the process chamber generates wafer features at block 730. Simultaneously, the digital twin model generates a simulation of the produced wafer features at block 735. The simulation of the produced wafer features should almost exactly match the actual produced wafer features. In this way, the digital replica of the process chamber is able to generate a computer simulation of the manufacturing process while the process is in operation.
[0140] FIG8 depicts an exemplary environment 800 within which embodiments of the present disclosure may be implemented to implement process operations concurrently with simulation. Environment 800 depicts a physical process flow 860 occurring concurrently with a digital twin flow 865. While these specific process operations are shown for illustrative purposes, other operations may also be present in other embodiments.
[0141] During the physical process flow, a machine controller 805 controls at least one aspect of at least one pixelated showerhead (PSH) region. As discussed herein, a PSH region can be any one or more nozzles located on a showerhead faceplate, sometimes also referred to herein as a "zone." In the exemplary FIG8 , there can be any N PSH regions (e.g., 810A, 810B, through 810N). The physical process flow 860 concludes with the production of the actual wafer 855.
[0142] Digital twin process 865 can operate in real time, concurrently with physical process flow 860. In digital twin process 865, virtual sensor 815 is used to generate a simulation of each pixelated showerhead (PSH regions 820A, 820B, through 820N). Furthermore, a gas profile is generated for each of these PSH regions, such as profiles 825A, 825B, through 825N. A gas profile represents the distribution of one or more process gases present at a specific region of the processing chamber. In other embodiments, a plasma profile can be generated that represents the amount of plasma present at a specific region of the processing chamber. Furthermore, other types of profiles can be generated for each region at this step. The individual regional profiles are combined into an aggregate profile 830 representing the presence of the process gas across the entire substrate.
[0143] The summary profile 830 is combined with simulation data about the plasma 840, the electrostatic chuck 835, and the chamber 845 itself. These combined inputs are used to generate a simulated processed wafer 850, which is a computer simulation of an actual processed wafer 855 produced by the physical process. In various embodiments, the simulated processed wafer 850 is generated substantially simultaneously with the actual processed wafer 855.
[0144] FIG9 depicts an exemplary environment 900 within which embodiments of the present disclosure may be practiced to implement process operations concurrently with simulation. Environment 900 depicts a physical process flow 960 occurring concurrently with a digital twin flow 965. While these specific process operations are shown for illustrative purposes, other operations may also be present in other embodiments.
[0145] During the physical process flow, a machine controller 905 controls at least one aspect of at least one electrostatic chuck (ESC) zone. As discussed herein, an ESC zone can be any one or more heaters located on an ESC, sometimes referred to herein as a "zone." In the exemplary embodiment of FIG. 9 , there can be any N ESC zones (e.g., 910A, 910B, through 910N). The physical process flow 960 concludes with the production of the actual wafer 955.
[0146] Digital twin process 965 can be run in real time, concurrently with physical process flow 960. In digital twin process 965, virtual sensors 915 are used to generate simulations for each ESC region (ESC regions 920A, 920B, through 920N). Furthermore, temperature profiles are generated for each of these ESC regions, such as profiles 925A, 925B, through 925N. These temperature profiles represent the temperature distribution at one or more locations within a specific region of the processing chamber. In other embodiments, a plasma profile representing the amount of plasma present at a specific region of the processing chamber can be generated. Furthermore, other types of profiles can be generated for each region at this step. The individual regional profiles are combined into an aggregated profile 930 representing the temperature across the entire substrate.
[0147] The summary profile 930 is combined with simulation data about the plasma 940, the pixelated showerhead 935, and the chamber 945 itself. These combined inputs are used to generate a simulated processed wafer 950, which is a computer simulation of an actual processed wafer 955 produced by the physical process. In various embodiments, the simulated processed wafer 950 is generated substantially simultaneously with the actual processed wafer 955.
[0148] FIG10 illustrates an exemplary environment 1000 within which embodiments of the present disclosure may be implemented to implement process operations concurrently with simulation. Environment 1000 depicts a physical process flow 1060 occurring concurrently with a digital twin flow 1065. While these specific process operations are shown for illustrative purposes, other operations may also be present in other embodiments.
[0149] During the physical process flow, the machine controller 1005 controls at least one aspect of at least one plasma zone. As discussed herein, plasma generation in the process chamber can be driven by one or more individually controlled electrodes (e.g., electrode 358 in FIG. 3 ). The one or more electrodes can comprise a group of electrodes controlled together, sometimes referred to herein as an electrode "zone." In the exemplary FIG. 10 , there can be any N plasma zones (e.g., 1010A, 1010B, through 1010N). The physical process flow 1060 concludes with the production of the actual wafer 1055.
[0150] The digital twin process 1065 can operate in real time, concurrently with the physical process flow 1060. In the digital twin process 1065, the virtual sensor 1015 is used to generate a simulation of each plasma region (plasma regions 1020A, 1020B, through 1020N). Furthermore, a distribution for each of these plasma regions is generated, such as distributions 1025A, 1025B, through 1025N. In one embodiment, the distribution is the composition distribution of the plasma within the region. In another embodiment, the distribution is the density distribution of the plasma within the region. In further embodiments, the distribution can be the distribution of any parameter associated with the plasma. The distribution represents the condition of the parameter within a specific region of the processing chamber. The individual regional distributions are combined into an aggregated distribution 1030 representing the plasma conditions across the entire substrate.
[0151] The summary distribution 1030 is combined with simulation data about the PSH 1040, the ESC 1035, and the chamber 1045 itself. These combined inputs are used to generate a simulated processed wafer 1050, which is a computer simulation of an actual processed wafer 1055 produced by the physical process. In various embodiments, the simulated processed wafer 1050 is generated substantially simultaneously with the actual processed wafer 1055.
[0152] FIG11 illustrates an exemplary environment in which embodiments of the present disclosure may be practiced to implement real-time process control. Environment 1100 depicts gas flow simulation 1150 and showerhead controller operations 1155. While these specific process operations are shown for illustrative purposes, other operations may also be present in other embodiments.
[0153] In environment 1100, machine controller 1105 controls at least one aspect of at least one pixelated showerhead region. As discussed herein, a PSH region can be any one or more nozzles located on a showerhead panel. In the exemplary embodiment of FIG11 , there can be N PSH regions. Machine controller 1105 can be a controller specific to a showerhead or a global controller for manufacturing equipment.
[0154] The real-time sensor input for each PSH zone is input into the gas flow simulation 1150. That is, the PSH zone A sensor input 1110A, the PSH zone B sensor input 1110B, and the PSH zone N sensor input 1110N are all input into the gas flow simulation 1150. As discussed herein, any number of PSH zones can be defined on the showerhead panel, typically between 1 and 20 zones.
[0155] Gas flow simulation 1150 continues using virtual sensor input 1115 generated from the real-time sensor input of each PSH region. The output of the virtual sensor 1115 is used to generate a gas distribution or plasma distribution for each PSH region, such as PSH region distribution 1120A, PSH region distribution 1120B, and PSH region distribution 1120N.
[0156] The individual PSH area distributions are combined into a summary distribution 1125 for the entire wafer in a showerhead control process 1155. The summary distribution 1125 is then used to simulate specific wafer characteristics and generate simulated wafer characteristics 1130.
[0157] The automatic parameter adjustment module 1135 (which may be part of the matching network model 192 of FIG. 1 ) compares any differences between the summary distribution 1125 and the target reference distribution 1140, as well as any differences between the simulated wafer characteristics 1130 and the reference characteristics 1145. If there are differences in the simulated wafer characteristics and / or distribution compared to the reference wafer characteristics and / or distribution, the automatic parameter adjustment module 1135 may determine adjustments that may be made to one or more nozzles of the showerhead during the current physical process operation to correct the differences.
[0158] Before implementing process adjustments, they can be sent to the gas flow simulation 1150 via path 1160 to recalculate the simulated distributions for each PSH zone. This generates a new summary distribution and a new simulated wafer signature. The automatic parameter adjustment module 1135 compares the new summary distribution and simulated wafer signature with the reference distribution 1140 and reference wafer signature 1145. If the adjustment improves the performance of the physical process by at least a certain threshold amount, the parameter adjustment is sent to the machine controller 1105 to effectively adjust the parallel physical process operations. If the parameter adjustment does not improve the performance of the physical process by at least a certain threshold amount, the automatic parameter adjustment module 1135 can determine a different signature adjustment to make and send this further adjustment again to the gas flow simulation for each showerhead zone via path 1160.
[0159] In various embodiments, the automatic parameter adjustment module 1135 continues the feedback loop in an iterative manner until a satisfactory result is achieved, i.e., the simulated summary distribution and the simulated wafer characteristics match the corresponding reference values within a predetermined threshold. Although FIG11 is depicted from the perspective of PSH operation, a similar process can also be used to individually control regions of the electrostatic chuck and individually control regions of the generated plasma.
[0160] FIG12 illustrates an exemplary method 1200 for controlling a manufacturing process in real time using autonomous parameter adjustment. At block 1205, a controller initiates physical operations of a manufacturing process in a processing chamber according to a recipe. As discussed herein, the manufacturing process can be a process associated with a pixelated showerhead, an electrostatic chuck, or any other process associated with generating a plasma in a processing chamber. Although not depicted in the diagram, the controller can also transmit process information or recipe information to a digital simulation (which can be a digital twin model) operating in parallel with the physical process. The digital simulation can run on the controller's own processor or on a separate processor.
[0161] In one embodiment, at block 1205, a controller causes manufacturing equipment to release at least one gaseous substance at a specified flow rate through a plurality of zones of a pixelated showerhead for a manufacturing process. In another embodiment, at block 1205, a controller causes a plurality of individually tunable heaters to heat a plurality of zones of an electrostatic chuck to a specified temperature for the manufacturing process. Each of the plurality of individually tunable heaters can be associated with at least one zone of the electrostatic chuck.
[0162] In another embodiment, a controller causes a plurality of individually tunable RF field generators to generate plasma across a plurality of regions in a processing chamber where a manufacturing process is performed. Each of the individually tunable RF field generators can be associated with at least one region of the processing chamber. In a further embodiment, the controller causes manufacturing equipment to perform two or more of the above-described embodiments in combination.
[0163] At block 1210, a controller receives a simulated profile from a digital model operating in parallel with the physical manufacturing process. The digital model receives metrology data and / or sensor data from the processing chamber. Combined with a previously generated model of the chamber and the process parameters for the currently operating process, the digital model simulates what the processed semiconductor wafer will look like if the current trajectory continues. Alternatively, the model may predict what specific features of the semiconductor wafer will look like at the end of the process.
[0164] In one embodiment, the controller receives, at block 1210, a gas profile for each of the plurality of regions of the pixelated showerhead from a trained digital simulation associated with the fabrication equipment. In another embodiment, the controller receives, at block 1210, a heat profile for each of the plurality of regions of the electrostatic chuck from a trained digital simulation associated with the electrostatic chuck. In another embodiment, the controller receives, at block 1210, a plasma profile for each of the plurality of regions of the processing chamber from a trained digital simulation associated with the processing chamber. In a further embodiment, the controller receives two or more of a gas profile, a heat profile, or a plasma profile from the trained digital simulation.
[0165] At block 1215, the controller compares the simulated distribution generated by the digital model with a target reference distribution previously stored in memory (e.g., data storage 140 in FIG. 1 ). The target reference distribution can be the target distribution that the manufacturing process is attempting to achieve. At block 1220, the controller determines whether the simulated predicted distribution matches the target reference distribution within a predetermined tolerance level. If so, the controller maintains the current process parameters at block 1225 and continues the manufacturing process. Upon receiving a new simulated distribution from the digital twin model, the controller returns to block 1210. That is, the digital model continues to iterate while the physical process is occurring.
[0166] If the controller determines at block 1220 that the simulated predicted distribution does not match the target reference distribution within a predetermined tolerance level, the parameter adjustment module recommends one or more corrective actions (e.g., recipe adjustments) using one or more previously trained machine learning models at block 1230. The recommended recipe adjustments are predictions of recipe changes that can be implemented in the process chamber to attempt to bring the predicted simulated distribution more into line with the target reference distribution before wafer processing is terminated.
[0167] In one embodiment, the one or more corrective actions are changing the flow rate of at least one gaseous substance through at least one of the plurality of regions of the pixelated showerhead. In another embodiment, the one or more corrective actions are changing the heating rate applied by at least one individually tunable heater to at least one region of the ESC. In another embodiment, the one or more corrective actions are changing the heating rate applied by at least one individually tunable RF field generator. In a further embodiment, the one or more corrective actions are a combination of two or more of these embodiments.
[0168] The suggested recipe adjustments are re-entered into the digital model to generate a new simulated profile for the adjusted recipe, which is received by the controller at block 1235. At block 1240, the simulated profile of the adjusted recipe is again compared to the target reference profile, and at block 1245, the controller determines whether the simulated profile of the adjusted recipe is within a tolerance level of the target reference profile. If the simulated profile of the adjusted recipe differs from the target reference profile within a predetermined tolerance level, then at block 1250, the controller implements the adjusted recipe by adjusting one or more parameters of the process chamber and continues wafer processing. The controller then returns to block 1210 to continue wafer processing and receive the updated simulated predicted profile.
[0169] If the controller determines at block 1245 that the simulated predicted distribution of the adjusted recipe is not within the tolerance level of the target reference distribution, method 1200 returns to block 1230, and the parameter adjustment module recommends further recipe adjustments. In this manner, autonomous parameter adjustment enables real-time control of the manufacturing process. Furthermore, any potential process drift or deviation can be detected and compensated in substantial real time, thereby reducing the potential for defective product production. Furthermore, autonomous parameter adjustment enables the manufacturing process to operate and correct itself without human intervention.
[0170] FIG13 illustrates an example machine of a computer system 1300 within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein may be executed. In alternative embodiments, the machine may be connected (e.g., networked) to other machines in a LAN, an intranet, an extranet, and / or the Internet. The machine may operate as a server or a client machine in a client-server network environment, as a peer machine in a peer-to-peer (or distributed) network environment, or as a server or client machine in a cloud computing infrastructure or environment.
[0171] The machine may be a personal computer (PC), tablet PC, set-top box (STB), web appliance, server, network router, switch or bridge, or any machine capable of executing a set of instructions (sequentially or otherwise) that specify actions to be taken by the machine. Further, while a single machine is shown, the term "machine" should be construed to include any collection of machines that individually or collectively execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein.
[0172] The example computer system 1300 includes a processing device 1302, a main memory 1304 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous dynamic random access memory (SDRAM), etc.), a static memory 1306 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage device 1316, which communicate with each other via a bus 1308.
[0173] Processing device 1302 represents one or more general-purpose processing devices, such as a microprocessor, a central processing unit, or the like. More specifically, the processing device may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or a processor implementing another instruction set or a combination of instruction sets. Processing device 1302 may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, or the like. Processing device 1302 is configured to execute instructions to perform the operations and steps discussed herein.
[0174] The computer system 1300 may further include a network interface device 1322 for communicating via the network 1318. The computer system 1300 may also include a video display device 1310 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 1312 (e.g., a keyboard), a cursor control device 1314 (e.g., a mouse or a touchpad), a signal generating device 1320 (e.g., a speaker), a graphics processing unit (not shown), a video processing unit (not shown), and an audio processing unit (not shown).
[0175] The data storage device 1316 may include a machine-readable storage medium 1324 (also referred to as a computer-readable medium) having stored thereon one or more sets of instructions 1326 or software embodying any one or more of the methodologies or functions described herein. The instructions 1326 may also reside, completely or at least partially, within the main memory 1304 and / or within the processing device 1302 during execution of the instructions by the computer system 1300, with the main memory 1304 and the processing device 1302 also constituting machine-readable storage media.
[0176] In one embodiment, instructions 1326 include instructions for implementing functionality corresponding to the digital twin model. Although machine-readable storage medium 1324 is illustrated as a single medium in the example embodiment, the term "machine-readable storage medium" should be considered to include a single medium or multiple media (e.g., a centralized or distributed database and / or associated caches and servers) storing the one or more sets of instructions. The term "machine-readable storage medium" should also be considered to include any medium capable of storing or encoding a set of instructions for execution by a machine and causing the machine to perform any one or more methodologies of the present disclosure. Thus, the term "machine-readable storage medium" should be considered to include (but not be limited to) solid-state memory, optical media, and magnetic media.
[0177] Some portions of the foregoing detailed description have been presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of operations leading to a desired result. These operations are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, combined, compared, and otherwise manipulated. It proves convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.
[0178] It should be remembered, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless otherwise specifically indicated from the above discussion, it should be understood that throughout this description, discussions utilizing terms such as "identify" or "determine" or "execute" or "performing" or "collect" or "create" or "send" or similar terms refer to the actions and processes of a computer system or similar electronic computing device that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system's registers or memories or other such information storage devices.
[0179] The present disclosure also relates to an apparatus for performing the operations described herein. This apparatus may be specially constructed for its intended use, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer. This computer program may be stored on a computer-readable storage medium, such as, but not limited to, any type of disk (including floppy disks, optical disks, CD-ROMs, and magneto-optical disks), read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic or optical cards, or any other type of medium suitable for storing electronic instructions, each of which is coupled to a computer system bus.
[0180] The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may be used with programs according to the teachings herein, or it may prove convenient to construct a more specialized apparatus to perform the methods. The structures of various such systems will be as described in the description. Furthermore, this disclosure is not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the disclosure described herein.
[0181] The present disclosure may be provided as a computer program product or software, which may include a machine-readable medium having stored thereon instructions that can be used to program a computer system (or other electronic device) to perform processes in accordance with the present disclosure. A machine-readable medium includes any mechanism for storing information in a form readable by a machine (e.g., a computer). For example, a machine-readable (e.g., computer-readable) medium includes machine-readable (e.g., computer-readable) storage media such as read-only memory ("ROM"), random access memory ("RAM"), magnetic disk storage media, optical storage media, flash memory devices, and the like.
[0182] In the foregoing specification, embodiments of the present disclosure have been described with reference to specific example embodiments thereof. It will be apparent that various modifications may be made thereto without departing from the broader spirit and scope of embodiments of the present disclosure as set forth in the following claims. The specification and drawings are, therefore, to be regarded in an illustrative rather than a restrictive sense.
[0183] While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the present disclosure may be devised without departing from the basic scope thereof. It is contemplated that elements and features of any disclosed embodiment may be beneficially incorporated into one or more other embodiments. While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the present disclosure may be devised without departing from the basic scope thereof, and the scope thereof will be determined by the claims that follow.
[0184] 100: System Architecture 110:Simulation system 112: Prediction Server 114: Prediction Component 120: Client device 122: Controller 124: Manufacturing Equipment 126:Sensor 128:Measuring Equipment 130: Internet 140:Data Storage 142:Sensor data 144: Historical sensor data 146: Current sensor data 148: Performance Data 150: Historical performance data 152: Current performance data 154: Library Information 156:Simulation data 158: Forecast Data 160: Correction action parts 170: Server Machine 172: Training Engine 174: Verification Engine 176:Select Engine 178:Test Engine 180: Trained Machine Learning Model 190:Digital Display Server 192: Matching Network Model 194:Processing chamber model 200: Manufacturing System 202:Substrate 204:Process Tools 206:Factory Interface 208: Shell 210: Chamber 212: Chamber Robot 214: Processing chamber 216: Processing chamber 218: Processing chamber 220: Loading gate 222: substrate carrier 224: Loading port 226:Factory Interface Robot 228: System Controller 240: Substrate measurement subsystem 250:Data storage 300: Processing chamber 302: Chamber body 304: Substrate support 306: lid 311: Gas supply source 312:Showerhead 314: Gas feeder 316: Air supply chamber 318: Opening 320: Processing volume 322: Electrode 326: Opening 328: Power Supply 336: Power Supply 354:Substrate 356: Exhaust port 357: Vacuum Pump 358: Individually tunable electrodes 400: Panel 410: District 412: Open 500: substrate support assembly 502: Workpiece support surface 503: Mounting surface 504:RF filter 505: Support base 506: Fasten the power supply 507: Clamping electrode 508: Dielectric body 510: substrate support 512: Main resistance heater 514: Spatially tunable heater 516: Heater controller 518:RF filter 522: Heater power supply 544: base 550:Controller board 552: Temperature sensor controller 554: Temperature sensor 556: Optical communication interface board 558:Optical converter 560: Cooling base 562: Fluid Source 564: Advanced Process Controller (APC) 590: Catheter 600:unit 616: Thermal Blockage 622: Hole 630: Pie-shaped area 640: Local area 662: Internal wedge 664: Surrounding groups 690: Unit 692: Midpoint 699: Pattern 700: Timing diagram 705: Machine Controller 710: Process chamber 715: Digital Twin Model 720: Block 725: Real-time simulation 730: Block 735: Block 800: Environment 805: Machine Controller 815: Virtual Sensor 830: Summary Distribution 835:Electrostatic chuck 840: Plasma 845: Chamber 850:Simulated processed wafer 855: Actual processed wafer 860: Physical process flow 865: Digital Twin Process 900: Environment 905: Machine Controller 915: Virtual Sensor 930: Summary Distribution 935: Pixelated Showerhead 940: Plasma 945: Chamber 950:Simulated processed wafer 955: Actual processed wafer 960: Physical process flow 965: Digital Twin Process 1000: Environment 1005: Machine Controller 1015: Virtual Sensor 1030: Summary Distribution 1035:ESC 1040:PSH 1045: Chamber 1050:Simulated processed wafer 1055: Actual processed wafer 1060: Physical process flow 1065: Digital Twin Process 1100: Environment 1105: Machine Controller 1115: Virtual sensor input 1125: Summary Distribution 1130:Simulated wafer characteristics 1135: Automatic parameter adjustment module 1140: Target reference distribution 1145: Reference Feature 1150:Gas flow simulation 1155:Showerhead controller operation 1160: Path 1200: Method 1205: Block 1210: Block 1215: Block 1220: Block 1225: Block 1230: Block 1235: Block 1240: Block 1245: Block 1250: Block 1300: Computer Systems 1302: Processing Equipment 1304: Main memory 1306: Static Memory 1308: Bus 1310: Video display device 1312: Alphanumeric input device 1314: Cursor control device 1316: Data storage device 1318: Internet 1320:Signal generating equipment 1322: Network Interface Device 1324: Machine-readable storage media 1326: Instructions 1010A: Plasma region 1010B: Plasma region 1010N: Plasma area 1020A: Plasma area 1020B: Plasma area 1020N: Plasma area 1025A: Distribution 1025B: Distribution 1025N: Distribution 1110A: Sensor input 1110B: Sensor input 1110N:Sensor input 1120A:PSH regional distribution 1120B:PSH regional distribution 1120N:PSH regional distribution 810A:PSH area 810B:PSH area 810N:PSH area 820A:PSH area 820B:PSH area 820N:PSH area 825A: Distribution 825B: Distribution 825N: Distribution 910A:ESC area 910B:ESC area 910N:ESC area 920A:ESC area 920B:ESC area 920N:ESC area 925A: Distribution 925B: Distribution 925N: Distribution
[0185] Domestic storage information (please note the order of storage institution, date, and number) none
[0186] Overseas deposit information (please note the order of deposit country, institution, date, and number) none
Claims
1. A method for controlling a semiconductor manufacturing process, the method comprising the steps of: processing a wafer in a processing chamber while causing a manufacturing apparatus to release at least one gaseous substance at a specified flow rate through a plurality of regions of a pixelated showerhead for a manufacturing process; obtaining a gas distribution for each of the plurality of regions of the pixelated showerhead from a trained digital simulation associated with the manufacturing apparatus, wherein the trained digital simulation is performed simultaneously with the manufacturing process; simulating a wafer feature based at least in part on the obtained gas distribution, the simulation predicting at least one feature of a processed wafer to be produced by the manufacturing process; and, during the manufacturing process and before the completion of processing of the wafer, inducing the execution of one or more corrective actions associated with the manufacturing apparatus based on the prediction.
2. The method of claim 1 further includes the step of comparing the simulated wafer feature with a target reference wafer feature.
3. The method of claim 1 further includes the step of: determining, before inducing the execution of the one or more corrective actions, whether the simulated wafer feature deviates from a target reference wafer feature by at least a predetermined threshold amount.
4. The method of claim 1, wherein the one or more corrective actions associated with the manufacturing equipment include a change in the flow rate of the at least one gaseous substance through at least one of the plurality of regions of the pixelated shower head.
5. The method of claim 1 further includes the following steps: receiving measurement information from at least one in-situ measurement tool, an off-site measurement tool, or an airborne measurement tool before obtaining the gas distribution information from the digital simulation.
6. The method of claim 1, further comprising the step of: prior to inducing the execution of the one or more corrective actions, obtaining an updated gas distribution from each of the plurality of regions of the digitally simulated pixelated showerhead based on a simulation of the one or more corrective actions.
7. A method for controlling a semiconductor manufacturing process, the method comprising the steps of: simultaneously processing a wafer in a processing chamber, causing a plurality of individually tunable heaters to heat a plurality of regions of an electrostatic chuck to a specified temperature for a manufacturing process, wherein at least one of the plurality of individually tunable heaters is associated with at least one region of the electrostatic chuck; obtaining a heat distribution for each of the plurality of regions of the electrostatic chuck from a trained digital simulation associated with the electrostatic chuck, wherein the trained digital simulation is performed simultaneously with the manufacturing process; simulating a wafer feature based at least in part on the obtained heat distribution, the simulation predicting at least one feature of a processed wafer to be produced by the manufacturing process; and, during the manufacturing process and before the completion of processing of the wafer, inducing the execution of one or more corrective actions associated with the manufacturing process based on the prediction.
8. The method of claim 7 further includes the step of comparing the simulated wafer feature with a target reference wafer feature.
9. The method of claim 7 further includes the step of: determining, before inducing the execution of the one or more corrective actions, whether the simulated wafer feature deviates from a target reference wafer feature by at least a predetermined threshold amount.
10. The method of claim 7, wherein the one or more corrective actions associated with the manufacturing process include a change in a heating rate applied to the at least one region of the electrostatic chuck by at least one of the individually tunable heaters.
11. The method of claim 7 further includes the following steps: receiving measurement information from at least one in-situ measurement tool, an off-site measurement tool, or an airborne measurement tool before obtaining the heat distribution information from the digital simulation.
12. The method of claim 7 further comprises the step of: obtaining an updated heat distribution from each of the plurality of regions of the digital simulation of the electrostatic chuck before inducing the execution of the one or more corrective actions.
13. A method for controlling a semiconductor manufacturing process, the method comprising the steps of: while processing a wafer in a processing chamber, causing a plurality of individually tunable radio frequency (RF) field generators to generate a plasma across a plurality of regions in the processing chamber during the manufacturing process, wherein at least one of the plurality of individually tunable RF field generators is associated with at least one region of the processing chamber; obtaining a plasma distribution for each of the plurality of regions of the processing chamber from a trained digital simulation associated with the processing chamber, wherein the trained digital simulation is performed concurrently with the manufacturing process; simulating a wafer feature based at least in part on the obtained plasma distribution, the simulation predicting at least one feature of a processed wafer to be produced by the manufacturing process; and, during the manufacturing process and prior to completion of processing of the wafer, inducing the execution of one or more corrective actions associated with the manufacturing process based on the prediction.
14. The method of claim 13 further includes the step of comparing the simulated wafer feature with a target reference wafer feature.
15. The method of claim 13 further includes the step of: determining, before inducing the execution of the one or more corrective actions, whether the simulated wafer feature deviates from a target reference wafer feature by at least a predetermined threshold amount.
16. The method of claim 13, wherein the one or more corrective actions associated with the manufacturing process include adjusting the RF power applied by at least one of the individually tunable RF field generators to a corresponding region of the processing chamber.
17. The method of claim 13 further comprises the following steps: receiving measurement information from at least one in-situ measurement tool, an off-site measurement tool, or an airborne measurement tool before obtaining the plasma distribution information from the digital analog.
18. The method of claim 13 further comprises the step of: obtaining an updated plasma distribution from each of the plurality of regions of the digital simulation representing the processing chamber before inducing the execution of the one or more corrective actions.
19. A method for controlling a semiconductor manufacturing process, the method comprising the steps of: processing a wafer in a processing chamber while causing a plurality of individually tunable heaters to heat a plurality of regions of an electrostatic chuck to a specified temperature for a manufacturing process, wherein at least one of the plurality of individually tunable heaters is associated with at least one region of the electrostatic chuck; processing the wafer in the processing chamber while causing a plurality of individually tunable RF field generators to generate a plasma across the plurality of regions in the processing chamber during the manufacturing process, wherein at least one of the plurality of individually tunable RF field generators is associated with at least one region of the processing chamber; obtaining heat distribution information for each of the plurality of regions of the electrostatic chuck from a trained digital simulation associated with the electrostatic chuck, wherein the trained digital simulation is executed simultaneously with the manufacturing process; Plasma distribution information is obtained for each of the plurality of regions of the processing chamber from a trained digital simulation associated with the processing chamber, wherein the trained digital simulation is performed concurrently with the manufacturing process; a wafer feature is simulated in part based on at least one of the heat distribution information or the plasma distribution information, the simulation predicting at least one feature of a processed wafer to be produced by the manufacturing process; and during the manufacturing process and before the processing of the wafer is completed, one or more corrective actions associated with the manufacturing process are triggered based on the prediction.
20. The method of claim 19 further comprises the step of: prior to causing the execution of the one or more corrective actions, a simulation based on the one or more corrective actions is performed to obtain at least one of the updated thermal distribution information or at least one of the updated plasma distribution information from the digital simulation for each of the plurality of corresponding regions.
Citation Information
Patent Citations
Methods and apparatuses for etch profile optimization by reflectance spectra matching and surface kinetic model optimization
US20190049937A1
Substrate carrier with array of independently controllable heater elements
US20210335631A1
Information processing apparatus, information processing system, and part ordering method
US20220207492A1
