Smart seasoning for growth chambers

Abstract

A method for analyzing the completion of seasoning in a semiconductor processing chamber may involve training a model using seasoning cycle characteristic data obtained from an existing semiconductor processing chamber. The supervised learning process may label the characteristic data based on expert-determined identification of seasoning completion, and optionally, based on chamber open event information or preventive maintenance information. The trained model may be used to characterize another chamber during seasoning to determine whether seasoning is complete and / or when, for how long, or how many seasoning cycles may be performed before seasoning is complete.

Description

Technical Field

[0001] Cross - Reference to Related Applications

[0001] This application claims the benefit and priority of U.S. Non - Provisional Application No. 18 / 361,326, entitled "GROWTH CHAMBER SMART SEASONING", filed on July 28, 2023, the content of which is hereby incorporated by reference in its entirety for all purposes.

[0002]

[0002] This disclosure generally describes techniques for completing the seasoning of a semiconductor processing chamber after preventive maintenance (PM) in a shorter time compared to conventional techniques that use fixed times for seasoning. More specifically, this disclosure describes systems and methods for smart seasoning of a semiconductor processing chamber after PM by monitoring chamber characteristics during a seasoning cycle to determine when seasoning is complete, and thereby determining that the chamber may be qualified for wafer monitoring and manufacturing.

Background Art

[0003]

[0003] Epitaxial deposition can include any type of crystal growth or deposition that forms a new crystal layer in a known direction with respect to a seed layer. Many modern semiconductor devices utilize epitaxial growth on silicon substrate wafers. For example, by supplying a silicon - based precursor to a deposition chamber at a temperature in the range of 300°C to 1200°C, a progressive layer of a film that matches the crystal lattice of the underlying seed layer can be formed. Conventional epitaxial film deposition typically includes a baking step that can be performed at about 700°C or higher. Epitaxial growth is being extended to various device structures and can be used for selective deposition or growth on a particular surface.

Summary of the Invention

[0004]

[0004] The productivity of epitaxial film processes is becoming more critical and difficult to achieve as the temperature range decreases and newer structures are manufactured with advances in processing technology, making improvements in recovery time after preventive maintenance more valuable.

[0005]

[0005] In a first embodiment, a method for seasoning a semiconductor processing chamber is disclosed, such as after an open chamber event or a preventive maintenance event. For example, the method of this embodiment may include performing a seasoning cycle, monitoring the characteristics of the semiconductor processing chamber, and using a trained model to determine whether the seasoning of the semiconductor processing chamber is complete based on the chamber signal and historical data in a library. In some examples, the method of this embodiment includes using a semiconductor processing chamber to perform a seasoning cycle, or causing a semiconductor processing chamber to perform a seasoning cycle, determining the characteristics of the semiconductor processing chamber during the seasoning cycle, determining, based on the characteristics, whether the seasoning of the semiconductor processing chamber is not complete, causing the semiconductor processing chamber to perform an additional seasoning cycle, determining additional characteristics of the semiconductor processing chamber during the additional seasoning cycle, determining, based on the additional characteristics, whether the seasoning of the semiconductor processing chamber is complete, and generating an indicator to identify the completion of the seasoning of the semiconductor processing chamber. In some examples, a model, such as a trained machine learning model, may be used to evaluate the characteristics and determine whether the seasoning is complete. For example, the method of this embodiment may include, or may further include, providing characteristics to a trained machine learning model configured to receive characteristics of a semiconductor processing chamber as input and to provide an output regarding the completion of seasoning of the semiconductor processing chamber. The output may be received from the trained machine learning model and used to determine whether or not the seasoning of the semiconductor processing chamber is complete. The process of repeatedly running a seasoning cycle, measuring or monitoring the characteristics of the semiconductor processing chamber during the seasoning cycle, providing characteristics to the model, and determining whether or not seasoning is complete based on the model output may be repeated one or more times as needed until it is determined that seasoning is complete.

[0006]

[0006] In some non-limiting examples, the output from the model may indicate the completion of seasoning of the semiconductor processing chamber, the expected completion time of seasoning of the semiconductor processing chamber, the expected duration of seasoning of the semiconductor processing chamber, the expected number of additional seasoning cycles for completion of seasoning of the semiconductor processing chamber, or any combination thereof. In some non-limiting examples, the method of this embodiment may generate a separate indicator that identifies one or more of the expected completion time of seasoning of the semiconductor processing chamber, the expected duration of seasoning of the semiconductor processing chamber, the expected number of additional seasoning cycles for completion of seasoning of the semiconductor processing chamber, or any combination thereof. Such an indicator may be generated after it has been determined that seasoning of the semiconductor processing chamber is not complete and may be useful in conveying information about the seasoning process to the user.

[0007]

[0007] In some embodiments, the seasoning cycle may include controlling the semiconductor processing chamber to establish etching conditions within the semiconductor processing chamber, or controlling the semiconductor processing chamber to establish epitaxial growth conditions within the semiconductor processing chamber, or both, in either order. In any embodiment, the execution of the seasoning cycle may include controlling the semiconductor processing chamber to establish etching conditions within the semiconductor processing chamber, or controlling the semiconductor processing chamber to establish epitaxial growth conditions within the semiconductor processing chamber, or both. In some embodiments, the semiconductor processing chamber may be controlled for the seasoning cycle to establish both etching conditions and epitaxial growth conditions, in either order.

[0008]

[0008] In any embodiment, the characteristics of the semiconductor processing chamber may be monitored, measured, or otherwise determined during the execution of the seasoning cycle and / or used by a machine learning model in evaluating the completion of the seasoning process. In some non-limiting examples, the characteristics may include the temperature inside the semiconductor processing chamber during the seasoning cycle, the thickness of the epitaxial layer produced inside the semiconductor processing chamber, the power supplied to the heater associated with the semiconductor processing chamber, the power settings of the heater associated with the semiconductor processing chamber, the growth rate inside the semiconductor processing chamber during the seasoning cycle, the etching rate inside the semiconductor processing chamber during the seasoning cycle, the optical conditions inside the semiconductor processing chamber during the seasoning cycle, the pressure inside the semiconductor processing chamber during the seasoning cycle, the gas composition inside the semiconductor processing chamber during the seasoning cycle, the inflow or outflow rate to or from the semiconductor processing chamber during the seasoning cycle, changes in any of these, or any combination thereof, or optionally, one or more of the various physical or structural parameters associated with the semiconductor processing chamber (e.g., geometric details of the semiconductor processing chamber or any of its components or internal components), or design specifications of the semiconductor processing chamber or any of its components or internal components. In some non-limiting examples, the characteristics may include, correspond to, or be derived from sensor data from one or more optical sensors, heat sources, thermal sensors, piezo sensors, etc.Optionally, the characteristics may include, or correspond to, virtual sensor data derived from one or more design specifications of the semiconductor processing chamber, the temperature in the semiconductor processing chamber during the seasoning cycle, the thickness of the epitaxial layer produced in the semiconductor processing chamber, the power supplied to the heater associated with the semiconductor processing chamber, the power setting of the heater associated with the semiconductor processing chamber, the growth rate in the semiconductor processing chamber during the seasoning cycle, the etching rate in the semiconductor processing chamber during the seasoning cycle, the optical conditions in the semiconductor processing chamber during the seasoning cycle, the pressure in the semiconductor processing chamber during the seasoning cycle, the gas composition in the semiconductor processing chamber during the seasoning cycle, the inflow or outflow rate to or from the semiconductor processing chamber during the seasoning cycle, any of these, or any combination thereof, or optionally, various physical or structural parameters associated with the semiconductor processing chamber (e.g., geometric details of the semiconductor processing chamber or any of its components or internal components), the semiconductor processing chamber or any of its components or internal components.

[0009]

[0009] The methods of this embodiment may or may further include training a machine learning model. For example, a machine learning model may be trained before providing characteristics for producing an output. For example, training a machine learning model may include receiving training characteristics from multiple runs of seasoning cycles performed by one or more semiconductor processing chambers, such as one or more reference semiconductor processing chambers, and generating training data based on the training characteristics. In some embodiments, the methods of this embodiment include receiving inputs corresponding to labeling information for multiple runs of seasoning cycles. In embodiments, the training data is generated using labeling information that identifies the completion of seasoning in one or more semiconductor processing chambers. Exemplary labeling information may include or correspond to information indicating a seasoning cycle that has been determined to be completed (e.g., information that may be determined according to expert or user analysis). Exemplary labeling information may include, or correspond to, information about open chamber events or preventive maintenance events related to multiple seasoning cycles, such as information indicating the duration of chamber open periods, information indicating changes in components within the semiconductor processing chamber, and information indicating preventive maintenance procedures or protocols used (e.g., dry air purging, exposure to ambient air, exposure to water). Once training data is prepared, the method of this embodiment may further include running a supervised learning algorithm to train a machine learning model using the training data. Thus, the machine learning model may be specifically adapted to generate output information based on input characteristic information to provide a meaningful analysis of the completion of the seasoning process.

[0010]

[0010] In some embodiments, for example, when different processing kits and / or devices are used within the same semiconductor processing chamber, the training characteristics may be derived from or acquired using the same semiconductor processing chamber in which seasoning is being performed. For example, the training characteristics may correspond to characteristics obtained during a previous seasoning process of the semiconductor processing chamber.

[0011]

[0011] Optionally, the training of a machine learning model can be sequentially updated with each completed seasoning process. In some examples, the method of this embodiment may include performing one or more additional seasoning cycles using the semiconductor processing chamber, or causing the semiconductor processing chamber to perform one or more additional seasoning cycles, after it has been determined that the seasoning of the semiconductor processing chamber is complete, for example, to obtain additional characteristic information of the semiconductor processing chamber that has been seasoned, for use in confirming and / or updating the machine learning model. In such a situation, determining that the seasoning of the semiconductor processing chamber is complete may include comparing the characteristics of the semiconductor processing chamber with previously determined characteristics of the semiconductor processing chamber (such as the characteristics of the semiconductor processing chamber obtained when the semiconductor processing chamber was in a seasoned state).

[0012]

[0012] Optionally, the user can manually trigger to override the completion determined by the trained machine learning model and, after determining that the seasoning of the semiconductor processing chamber is complete, to use the semiconductor processing chamber to perform one or more additional seasoning cycles, or to cause the semiconductor processing chamber to perform one or more additional seasoning cycles. This may, in some cases, be useful for confirming the completion of seasoning before using the semiconductor processing chamber in a production environment where semiconductor devices are processed. In some cases, the method of this embodiment may include receiving an input indicating that an open chamber event or a preventive maintenance event is occurring, and in response to that, performing one or more additional seasoning cycles, or causing the semiconductor processing chamber to perform one or more additional seasoning cycles, and determining the characteristics of the semiconductor processing chamber during the one or more additional seasoning cycles. In this way, historical characteristic data may be acquired before the open chamber event or preventive maintenance event and used in determining whether seasoning is complete following the open chamber event or preventive maintenance event.

[0013]

[0013] In any way of this embodiment, the machine learning model may perform regression and / or classification analysis, for example, using various characteristics determined during one or more seasoning cycles. For example, regression analysis may analyze measured characteristic information or virtual sensor data to predict the behavior of seasoning. The prediction of such behavior may be indicated by training data and / or labeling information. Similarly, classification analysis may classify the seasoning process for a meaningful comparison with appropriate training data by analyzing characteristics determined during one or more seasoning cycles and incorporating, optionally, labeling information such as identifying information on open chamber events or preventive maintenance events. Whether using regression analysis or classification analysis, or both, or other analytical techniques, the machine learning model may be used to identify or map true positives in previous characteristic data related to known seasoning processes in the training dataset in order to establish high reliability in the machine learning model for predicting the completion of seasoning. In any way of this embodiment, the machine learning model may compare the characteristics of a seasoning cycle to the corresponding characteristics of the immediately preceding seasoning cycle in order to assess the completion of seasoning of a semiconductor processing chamber. Some embodiments include cases where a machine learning model determines the change in properties between the current cycle and the previous seasoning cycle, and evaluates the completion of seasoning the semiconductor processing chamber based on whether the change in properties falls below a threshold. In some cases, the threshold may be predetermined by a trained machine learning model. Optionally, the threshold may be adjusted by receiving inputs corresponding to a user-determined threshold or an override of the threshold.

[0014]

[0014] In another embodiment, the Specified Description provides a system, such as a semiconductor processing chamber and a system for operating the semiconductor processing chamber. For example, the system of this embodiment may be configured or adapted to season the semiconductor processing chamber and to automatically determine when the seasoning of the semiconductor processing chamber is complete, for example, based on sensor data or other characteristics measured or monitored during the seasoning process. In some embodiments, the system of this embodiment may be configured, adapted or programmed to perform any of the methods described herein. In some embodiments, the system of this embodiment may comprise one or more processors and one or more non-temporary computer-readable storage devices that, when executed by the one or more processors, contain instructions causing the one or more processors to perform an operation. In some embodiments, these operations may include causing a semiconductor processing chamber to perform a seasoning cycle; determining the characteristics of the semiconductor processing chamber during the seasoning cycle; providing the characteristics to a trained machine learning model configured to receive the characteristics of the semiconductor processing chamber as input and to provide an output regarding the completion of the seasoning of the semiconductor processing chamber; receiving an output from the trained machine learning model; determining, based on the output, that the seasoning of the semiconductor processing chamber is not complete; causing the semiconductor processing chamber to perform an additional seasoning cycle; continuing to determine the characteristics of the semiconductor processing chamber during the additional seasoning cycle; providing the characteristics of the semiconductor processing chamber during the additional seasoning cycle to the trained machine learning model; receiving an additional output from the trained machine learning model; determining, based on the additional output, that the seasoning of the semiconductor processing chamber is complete; and generating an indicator to identify the completion of the seasoning of the semiconductor processing chamber. In some embodiments, the system of this embodiment may further include one or more sensors that communicate data with one or more processors in order to determine the characteristics of the semiconductor processing chamber during the seasoning cycle. Non-exclusive examples of sensors include optical sensors, heat sources, thermal sensors, and piezoelectric sensors.It can be understood that a heat source can be useful as a sensor because its power consumption or power output can be measured by or during the operation of that heat source.

[0015]

[0015] In the systems and methods described herein, the characteristics may optionally include or correspond to time-series data acquired from one or more sensors. Optionally, the characteristics may include or correspond to a comparison using a first time-series of data acquired from one or more sensors during a first seasoning cycle and a second time-series of data acquired from one or more sensors during a second seasoning cycle immediately preceding the first seasoning cycle.

[0016]

[0016] In any embodiment, the semiconductor processing chamber may include one or more of a quartz dome, a flow module, or a susceptor. Optionally, the semiconductor processing chamber may include one or more sensors, a flow module, a heat source, a light source, a window, or any other suitable components. In some examples, the systems described herein may include, or further include, one or more sensors positioned to measure the properties of one or more surfaces of the quartz dome, the flow module, the susceptor, or the substrates within the semiconductor processing chamber.

[0017]

[0017] In other embodiments, computer program products are described herein. An exemplary computer program product may include, for example, a non-temporary computer-readable medium that, when executed by one or more processors, contains instructions causing one or more processors to perform an action. These actions may, but are not limited to, correspond to any of the methods described herein. For example, these operations may include causing a semiconductor processing chamber to perform a seasoning cycle, determining the characteristics of the semiconductor processing chamber during the seasoning cycle, providing the characteristics to a trained machine learning model configured to receive the characteristics of the semiconductor processing chamber as input and to provide an output regarding the completion of the seasoning of the semiconductor processing chamber, receiving an output from the trained machine learning model, determining, based on the output, that the seasoning of the semiconductor processing chamber is not complete, causing the semiconductor processing chamber to perform an additional seasoning cycle, continuing to determine the characteristics of the semiconductor processing chamber during the additional seasoning cycle, providing the characteristics of the semiconductor processing chamber during the additional seasoning cycle to the trained machine learning model, receiving an additional output from the trained machine learning model, determining, based on the additional output, that the seasoning of the semiconductor processing chamber is complete, and generating an indicator to identify the completion of the seasoning of the semiconductor processing chamber.

[0018]

[0018] As discussed elsewhere in this Specification, a trained machine learning model may be trained to model the seasoning of a semiconductor processing chamber following a chamber open event or a preventive maintenance event associated with the semiconductor processing chamber. Optionally, the method, system, or computer program product described herein may include, or be configured to perform, steps of obtaining user inputs corresponding to one or more of the following: the duration of the chamber open event or preventive maintenance event, changes in the components of the semiconductor processing chamber during the chamber open event or preventive maintenance event, environmental conditions during the chamber open event or preventive maintenance event, maintenance procedures or protocols applied to the semiconductor processing chamber during the chamber open event or preventive maintenance event, or a severity assessment of the chamber open event or preventive maintenance event. For example, a trained machine learning model may be further configured to use user inputs when generating an output regarding the completion of seasoning of the semiconductor processing chamber. Non-limiting examples include information about the chamber open period, information about changes in the components within the semiconductor processing chamber, and information about preventive maintenance procedures or protocols used (e.g., dry air purging, exposure to ambient air, exposure to water). This information can be usefully used in, or by, trained machine learning models to evaluate the progress of the seasoning process and provide insights into the rate at which seasoning is progressing. This is because the seasoning process may correspond to a physical process in which surface components are passivated (e.g., by the removal or desorption of moisture during periodic exposure to etching and growth conditions). In addition, various pieces of information regarding open chamber events or preventive maintenance events may indicate that a longer seasoning process is likely to occur. This is because a larger amount of water may be introduced into the semiconductor processing chamber during a chamber open event or scheduled maintenance event.

[0019]

[0019] Optionally, the output from the trained machine learning model may include, or correspond to, an indication of the completion or incompleteness of the seasoning of the semiconductor processing chamber and / or one or more variance values ​​for one or more characteristics or environment variables. One or more variance values ​​may be determined, for example, by comparing the characteristics of the semiconductor processing chamber during a first seasoning cycle with the characteristics of the semiconductor processing chamber during a second seasoning cycle immediately preceding the first seasoning cycle. In some examples, determining that the seasoning of the semiconductor processing chamber is complete based on the output from the trained machine learning model may include deriving a variance value between the output of the first seasoning cycle and the output of the second seasoning cycle immediately preceding the first seasoning cycle, and determining that this variance value is smaller than a target variance value. In some embodiments, an input corresponding to an acceptable range for the model may be received. This input indicates an acceptable deviation from the target variance value and may correspond to user overrides. This allows the user to control the number of seasoning cycles on the fly.

[0020]

[0020] Without intending to be bound by any particular theory, beliefs or understandings of the fundamental principles relating to the present invention and the processes or parameters described herein may be discussed herein. Regardless of the final accuracy of any mechanical description or hypothesis, it is recognized that the examples and embodiments described herein may still be operational and useful.

[0021]

[0021] A further understanding of the properties and advantages of various embodiments and examples can be achieved by referring to the remainder of this specification and the drawings. In the drawings, similar reference numerals are used throughout several drawings to point to similar components. In some cases, a sub-label is associated with the reference numeral to indicate one of several similar components. When a reference numeral is referred to without specifying an existing sub-label, it is intended to refer to all such several similar components.

Brief Description of the Drawings

[0022] [Figure 1]

[0022] Shows a schematic side cross-sectional view of a semiconductor processing chamber according to some embodiments. [Figure 2]

[0023] Shows a flowchart of a method for training a machine learning model to characterize the completion of seasoning of a semiconductor processing chamber according to some embodiments. [Figure 3]

[0024] Shows a block diagram of how to collect and process semiconductor processing chamber sensor characteristic data to generate virtual sensor data according to some embodiments. [Figure 4]

[0025] Shows a graph of time series data measured during a seasoning cycle by a semiconductor processing chamber according to some embodiments. [Figure 5]

[0026] Shows a bar graph displaying virtual sensor data used by a machine learning model during seasoning of a semiconductor processing chamber according to some embodiments. [Figure 6]

[0027] A - D show additional bar graphs displaying virtual sensor data used by a machine learning model during seasoning of a semiconductor processing chamber exposed to conditions of different opening events or preventive maintenance events according to some embodiments. [Figure 7]

[0028] Shows a flowchart of a method for training a model to predict the completion of semiconductor processing chamber seasoning according to some embodiments. [Figure 8]

[0029] Shows a flowchart of a method for seasoning a semiconductor processing chamber according to some embodiments. [Figure 9]

[0030] Shows a flowchart of using a trained machine learning model to evaluate the completion of seasoning of a semiconductor processing chamber according to some embodiments. [Figure 10]

[0031] This illustrates an exemplary computer system in which various embodiments can be implemented. [Modes for carrying out the invention]

[0023]

[0032] Thermal chemical vapor deposition (CVD) is one of the highest temperature processes in semiconductor manufacturing. During the deposition process, precursor gases may flow into the semiconductor processing chamber at temperatures ranging from approximately 300°C to approximately 1200°C during the deposition, etching, or chamber cleaning processes. The temperature setpoint may vary throughout the process according to the processing recipe, and the temperature within the semiconductor processing chamber may rise / fall between different setpoints based on the thermal properties of the chamber and as different gases flow into and out of the chamber. The rate and quality of epitaxial growth or etching may depend not only on the absolute temperature and gas mixture but also on the consistent behavior of the semiconductor processing chamber. In general, the epitaxial growth and / or etching performance within a semiconductor processing chamber may vary based on the chamber's processing history. To obtain best performance, the semiconductor processing chamber should exhibit very stable conditions between various epitaxial growth or etching cycles.

[0024]

[0033] At a more detailed level, epitaxial growth and etching behavior within a semiconductor processing chamber can be altered by the presence of water or impurities (e.g., halide salts) inside or on the surface of the chamber or its components. Semiconductor processing chambers may be opened periodically or undergo preventive maintenance within production or research facilities, which could introduce contaminants to the surface or cause other changes. For example, exposure to air could allow water to enter the semiconductor processing chamber, and changes to the processing equipment during preventive maintenance or troubleshooting could introduce water or other impurities, or alter the balance or state of passivated surfaces. Various open chamber events or preventive maintenance events could expose the semiconductor processing chamber to various environmental conditions (e.g., relative humidity) over varying durations, or alter various components, potentially destabilizing the conditions.

[0025]

[0034] To restore performance following an open chamber or preventive maintenance event, a semiconductor processing chamber may be subjected to a cycling process, generally referred to herein as “seasoning,” in which the temperature and gas mixture within the chamber are repeatedly cycled, such as between etching conditions and epitaxial growth conditions. Seasoning can be useful for removing moisture and / or impurities introduced into or absorbed inside or above the semiconductor processing chamber, establishing passivated surfaces of the semiconductor processing chamber or its components, or returning the semiconductor processing chamber to pre-open chamber event conditions or preventive maintenance event conditions. Since the number of seasoning cycles required to return to pre-open chamber event conditions or preventive maintenance event conditions may vary depending on the duration of the event, environmental conditions during the event, and changes in internal components during the event, a typical seasoning protocol performs a number of fixed seasoning cycles on the semiconductor processing chamber. This may be sufficient to establish stable conditions after any open chamber event or preventive maintenance event.

[0026]

[0035] However, in many cases, the numerous fixed seasoning cycles in conventional seasoning processes can be excessive for establishing stable conditions. This is because some semiconductor processing chambers can be seasoned using fewer seasoning cycles, even after several chamber opening events or preventive maintenance events. Since each seasoning cycle takes time, running excessive seasoning cycles on a semiconductor processing chamber can reduce available production time. This disclosure solves these and other problems by establishing a technique for monitoring the conditions inside a semiconductor processing chamber to determine when sufficient seasoning has been achieved and / or when the semiconductor processing chamber has returned to or established stable conditions suitable for production. Furthermore, embodiments described herein include using a trained machine learning model to evaluate the monitored conditions and automatically determine when seasoning is complete, or to provide the user with feedback on the expected duration or number of seasoning cycles required to return the semiconductor processing chamber to stable conditions.

[0027]

[0036] Figure 1 shows a schematic side cross-sectional view of an exemplary semiconductor processing chamber 100 according to several embodiments. The semiconductor processing chamber 100 may represent a deposition chamber. In some embodiments, the semiconductor processing chamber 100 may include or comprise an epitaxial deposition or growth chamber or an etching chamber. The semiconductor processing chamber 100 may be used to grow or process an epitaxial film on a substrate 102. The semiconductor processing chamber 100 may generate a precursor flow across the upper surface 150 of the substrate 102. Throughout this disclosure, an epitaxial deposition chamber may be used as a concrete example of a semiconductor chamber. However, the embodiments described herein may be equally applicable to any semiconductor processing chamber. Thus, any specific reference to an epitaxial deposition chamber may be considered more generally to any semiconductor processing chamber.

[0028]

[0037] The semiconductor processing chamber 100 may include an upper body 156, a lower body 148 positioned below the upper body 156, and a flow module 112 positioned between the upper body 156 and the lower body 148. The upper body 156, the flow module 112, and the lower body 148 may form the chamber body. Within the chamber body, a substrate support 106, an upper window 108, a lower window 110 (e.g., a lower dome), a plurality of upper heat sources 141, and / or a plurality of lower heat sources 143 may be arranged. As shown, a controller 120 is communicable with the semiconductor processing chamber 100 and may be used to control processes and steps, such as the steps of the method described herein. The controller 120 and the semiconductor processing chamber 100 may be part of a larger substrate processing system or tool platform.

[0029]

[0038] The semiconductor processing chamber 100 may also include a plurality of upper heat sources 141 for heating the upper portion of the semiconductor processing chamber 100 and a plurality of lower heat sources 143 for heating the lower portion of the semiconductor processing chamber 100. The plurality of upper heat sources 141 may be positioned between the upper window 108 and the lid 154. The plurality of upper heat sources 141 may form part of the upper heating module 155. The plurality of lower heat sources 143 may be positioned between the lower window 110 and the chamber floor 152. The plurality of lower heat sources 143 may form part of the lower heating module 145. In the embodiment shown in Figure 1, the heat sources 141, 143 may be heating lamps. Other heat sources such as resistance heaters, light-emitting diodes (LEDs) and / or lasers may also be used without limitation. The heat sources 141, 143 may be connected to a reflector 175. The reflector 175 may be configured to redirect the thermal energy from the heat sources 141, 143 toward the processing space 136.

[0030]

[0039] The upper window 108 can be an upper dome and is at least partially formed of an energy-permeable material such as quartz. Therefore, the upper window 108 may also be referred to as a dome or a quartz dome. The lower window 110 is a lower dome and is at least partially formed of an energy-permeable material such as quartz. The lower window 110 may also be simply referred to as a dome or a quartz dome. The upper window 108 may include a concave or flat first surface 111 (in the embodiment shown in Figure 1, the first surface 111 is concave). The upper window 108 includes a convex second surface 113. The second surface 113 may face the substrate support 106. The upper window 108 may include an inner section 122 and / or an outer section 124. The first surface 111 and the second surface 113 may form at least a portion of the inner section 122. The inner section 122 may be transparent and the outer section 124 may be opaque. The outer section 124 can be at least partially housed within one or more side walls of the semiconductor processing chamber 100 (such as within the flow module 112).

[0031]

[0040] The processing space 136 and the purge space 138 may be located between the upper window 108 and the lower window 110. The processing space 136 and the purge space 138 may be part of an internal space at least partially defined by the upper window 108, the lower window 110, and one or more liners 163. The upper window 108 may at least partially define the processing space 136.

[0032]

[0041] The internal space may include a substrate support 106 located inside it. The substrate support 106 may be positioned between an upper window 108 and a lower window 110. The substrate support 106 may include a support surface 123 that supports the substrate 102. The substrate support 106 may also be referred to as a susceptor, platen, or platform. The substrate support 106 may include an upper surface on which the substrate 102 is placed. The substrate support 106 may also be attached to a shaft 118. The shaft 118 may be connected to a motion assembly 121. The motion assembly 121 may include one or more actuators and / or adjusters that cause movement and / or adjustment of the shaft 118 and / or the substrate support 106 within the processing space 136.

[0033]

[0042] The substrate support 106 may include a lift pin hole 107 located within the substrate support 106. The lift pin hole 107 may be sized to accommodate a lift pin 132 for lifting the substrate 102 from the substrate support 106, either before or after the deposition process. The lift pin 132 may rest on a lift pin stop 134 when the substrate support 106 is lowered from the processing position to the transfer position. The lift pin stop 134 may be connected to a second shaft 104. The substrate (such as the substrate 102) may be transferred to and / or from the internal space of the semiconductor processing chamber 100 through a transfer door 137 (such as a slit valve). When the transfer door 137 is opened, a transfer device may extend through the transfer door 137 into the internal space (with the substrate supported thereon). This allows the lift pin 132 to lift the substrate from the transfer device and place the substrate on the substrate support 106 for processing. After processing, the lift pins 132 can lift the substrate from the substrate support 106 and land the substrate in the transfer device. The transfer device can then retract through the open transfer door 137 to remove the substrate from the semiconductor processing chamber 100.

[0034]

[0043] The flow module 112 may include a plurality of gas inlets 114, a plurality of purge gas inlets 164, and one or more gas outlets 116. The plurality of gas inlets 114 and the plurality of purge gas inlets 164 may be located on the opposite side of the flow module 112 from the one or more gas outlets 116. One or more flow guides 117 may be located below the plurality of gas inlets 114 and / or one or more gas outlets 116. One or more flow guides may include, for example, one or more preheating rings. One or more flow guides 117 may be located above the purge gas inlets 164. One or more liners 163 may be located on the inner surface of the flow module 112 to protect the flow module 112 from reactive gases used during the deposition process and / or cleaning process. Each of the one or more gas inlets 114 and the one or more purge gas inlets 164 may be positioned to allow gas to flow parallel to the upper surface 150 of the substrate 102 located in the processing space 136. One or more gas inlets 114 may be fluidically connected to one or more process gas sources 151 and one or more scrubbing gas sources 153. One or more purge gas inlets 164 may be fluidically connected to one or more purge gas sources 162 and / or one or more scrubbing gas sources 153. One or more gas outlets 116 may be fluidically connected to an exhaust pump 157. One or more process gases supplied using one or more process gas sources 151 may contain one or more reactive gases (silicon (Si), phosphorus (P), and / or germanium (Ge)-containing gases) and / or one or more carrier gases (such as one or more of nitrogen (N2) and / or hydrogen (H2)). One or more purge gases supplied using one or more purge gas sources 162 may contain one or more inert gases (such as one or more of argon (Ar), helium (He), hydrogen (H2), and / or nitrogen (N2)). One or more cleaning gases supplied using one or more cleaning gas sources 153 may contain one or more hydrogen (H)-containing gases and / or chlorine (Cl)-containing gases. In some embodiments, one or more processing gases may contain silicon phosphide (SiP) and / or phosphine (PH3), and one or more cleaning gases may contain hydrochloric acid (HCl).It should be noted that these exemplary gases are non-exclusive, and other processing gases, purging gases, or cleaning gases may also be used.

[0035]

[0044] One or more gas outlets 116 may be further connected to or include an exhaust system 178. The exhaust system 178 may fluidly connect one or more gas outlets 116 to an exhaust pump 157. The exhaust system 178 may assist in the controlled deposition of layers on the substrate 102. The exhaust system 178 may be located on the opposite side of the semiconductor processing chamber 100 from the flow module 112.

[0036]

[0045] The controller 120 may include a central processing unit (CPU), memory containing instructions, and support circuitry for the CPU. For example, the controller 120 may include one or more processors. One or more processors may be distributed among a local controller of the semiconductor processing chamber 100, a tool server on a tool or platform operating multiple semiconductor processing chambers of different types, and / or a cloud-based or facility-based server via a network or wireless connection. One or more non-temporary computer-readable media may store instructions that cause one or more processors of the controller 120 to perform the processes described herein. Collectively, the computer-readable media and one or more processors, together with other components, may constitute the controller 120, but these are not limited to a single computer system. Instead, the controller 120 may be distributed among multiple different computer systems in various locations. In some embodiments, the controller 120 may be communicably connected to a dedicated controller. The controller 120 functions as a central controller. Alternatively, any processes performed or described herein may be distributed between the controller 120 and other computing systems. For example, processing steps, model training, and / or model storage may be performed by the controller 120, by another computing system located outside the facility, by a cloud-based computing system, or by any combination of the above or other systems. Some of these processing steps may also be distributed to individual sensors or control circuits within the chamber.

[0037]

[0046] An example of a computer system that may be used to implement at least a part of the controller 120 is described in detail below in Figure 10. Processing parameters (process gas pressure, process gas flow rate, and / or rotational position of the process kit) and processes may be stored as software routines in a computer-readable medium. When these software routines are executed or activated, the controller 120 functions as an application-specific controller and controls the processes of the various chambers / modules described herein. The controller 120 is configured to perform any of the processes described herein. The various processes described herein may be performed automatically using the controller 120, or automatically or manually by a user performing specific processes.

[0038]

[0047] The controller 120 can control various items directly or via other computers and / or controllers. For example, commands executed by the controller 120 can cause the semiconductor processing chamber 100 to perform operations in a recipe, such as flowing gas, increasing temperature, or receiving substrates. The controller 120 may be configured to control the rotational position, heating, and gas flow through the semiconductor processing chamber 100 by providing outputs to control units for the heat sources 141, 143, gas flow, and motion assembly 121. The control units may include control units for the upper heat source 141, the lower heat source 143, the processing gas source 151, the purge gas source 162, the motion assembly 121, and the exhaust pump 157.

[0039]

[0048] The controller 120 may be configured to adjust the output to the control unit based on sensor readings, a system model, and stored readings and calculation results. The controller 120 may include built-in software and one or more correction algorithms for calibrating the measured values. The controller 120 may include one or more machine learning algorithms and / or artificial intelligence algorithms to analyze sensor readings to determine or estimate when or whether a seasoning process is complete, to estimate optimized parameters for deposition, etching, purging, and / or cleaning operations, or to evaluate performance or stability within the semiconductor processing chamber 100. One or more machine learning algorithms and / or artificial intelligence algorithms may use, for example, regression models (such as linear regression models) or clustering techniques. The algorithms may be unsupervised or supervised.

[0040]

[0049] The lid 154 may include a number of sensors located inside or on top of it for measuring characteristics within the semiconductor processing chamber 100, such as measuring the temperature inside the semiconductor processing chamber, monitoring optical conditions inside the semiconductor processing chamber 100, or monitoring etching or growth conditions, speed, or characteristics inside the semiconductor processing chamber. For example, a central temperature sensor 172 may be located on the lid 154 and configured to measure the temperature of the central portion of the semiconductor processing chamber 100 or its vicinity. More specifically, the central temperature sensor 172 may be configured to measure the temperature of the upper window 108 or the quartz dome. The central temperature sensor 172 may also be configured to measure the temperature of the upper surface 150 of the substrate 102 and / or substrate support 106. One or more temperature sensors 173 may also be distributed around the semiconductor processing chamber 100. These temperature sensors 173 may be configured to measure specific locations around the semiconductor processing chamber 100. For example, these temperature sensors 173 may be directed to the periphery of the upper window 108, to internal components such as one or more liners 163, or to the periphery of the substrate support 106 or susceptor. Although not explicitly shown in Figure 1, the semiconductor processing chamber 100 may include other sensors distributed throughout the semiconductor processing chamber 100. For example, a temperature sensor may be located in the lower heating module 145 and positioned to measure the temperature around the substrate support 106 and / or the center of the bottom of the substrate support 106. In another example, one or more optical sensors may be located in the lid 154 and positioned to measure the optical properties of the substrate support 106, the upper window 108, the lower window 110, etc. In some embodiments, one or more piezoelectric or piezoresistive sensors may be located within the semiconductor processing chamber 100 to characterize the mass change or rate of change during processing. These sensors may be located, for example, in or on the substrate support 106 or the shaft 118. In some examples, the upper heat source 141 and the lower heat source 143 may function or operate as sensors. This is because the power consumption by the upper heat source 141 and the lower heat source 143 may be useful characteristics that can provide insights into the stability within the semiconductor processing chamber 100.

[0041]

[0050] More generally, multiple sensors may be oriented toward individual components of the semiconductor processing chamber 100. For example, the central temperature sensor 172 may be assigned to measure the temperature, particularly at the center of the upper window 108 or the quartz dome. Temperature or optical sensors may be positioned in a specific orientation and assigned to measure the temperature or optical properties beneath the susceptor or substrate support 106, and / or the temperature or optical properties above the substrate or susceptor / substrate support 106. Temperature or optical sensors may be assigned to a liner within the semiconductor processing chamber 100 and oriented toward the liner. These temperature sensor assignments are provided as examples only and are not intended to be limiting. Any of the components in Figure 1 or described above may be assigned specific sensors configured to measure the properties of that component (e.g., temperature, optical properties, etc.). As will be discussed later, this makes it possible to use the individual sensor data to analyze properties or changes in properties related to a particular component, or to evaluate the stability within the semiconductor processing chamber 100 at a general or finer level. This allows the controller to identify specific components that may prevent the semiconductor processing chamber 100 from operating as desired.

[0042]

[0051] The temperature sensor can be implemented using any type of temperature sensor. For example, in some embodiments, a pyrometer may be used. In some embodiments, a thermocouple may be used. In some embodiments, a resistance temperature detector (RTD) may be used. In some embodiments, a semiconductor-based temperature sensor may be used. In some embodiments, a temperature scanning system may be used, which includes a single sensor (e.g., a pyrometer) and is configured to scan temperature measurements across the chamber along a predetermined path through an upper / lower window. More generally, the temperature sensor can be implemented by any device configured to provide an output that changes in a temperature-dependent manner within the semiconductor processing chamber 100.

[0043]

[0052] The sensor may be used to sample the internal conditions (e.g., temperature or optical properties) of the semiconductor processing chamber 100 during various processes performed by the semiconductor processing chamber 100. For example, the sensor may be used to acquire measurement data that can be used to train a model.

[0044]

[0053] Figure 2 shows a flowchart of Method 200 for training a machine learning model to characterize the seasoning of a semiconductor processing chamber or to evaluate the completion of the seasoning process, according to several embodiments. The Method may be performed by a controller including one of the controllers and / or computer systems described herein. The processing chamber may include an epitaxial deposition chamber, or epi chamber.

[0045]

[0054] Method 200 may include, in block 205, receiving training characteristics from multiple runs of multiple seasoning cycles performed by one or more semiconductor processing chambers. Exemplary training characteristics may include time-series data acquired from one or more sensors in the semiconductor processing chamber, or virtual sensor data derived from time-series data. In some embodiments, a model is trained for a certain type of semiconductor processing chamber (e.g., an epi chamber), and training characteristics may be received from multiple different semiconductor processing chambers of this type (e.g., epi chambers), or from one specific semiconductor processing chamber. Alternatively, some embodiments train a model for a specific semiconductor processing chamber, and training characteristic data may be received from multiple runs of a process on the same semiconductor processing chamber. Temperature data will be understood as an example of a characteristic that can be used to evaluate the stability within the semiconductor processing chamber during seasoning and to determine the completion of seasoning. Embodiments and examples described herein may use any other characteristics of the chamber instead of temperature. For example, in any embodiment, temperature measurements may be replaced by, or combined with, other parameters such as optical measurements, heater power, growth rate, etching rate, etc. Therefore, the temperature data is used only as an example and is not intended to be limiting. Any embodiment that specifically refers to temperature data may, alternatively, use any other semiconductor processing chamber characteristics, without limitation.

[0046]

[0055] In some cases, training characteristics may include processed characteristics corresponding to a virtual sensor. This may occur, for example, when combining or processing one or more sets of sensor data to generate a dataset that can function similarly to a single set of sensor data. In some cases, sensor data acquired during subsequent seasoning cycles on a single chamber may provide virtual sensor data that can establish changes in characteristics between subsequent seasoning cycles. For example, if time-series measurements of a temperature sensor are obtained for different seasoning cycles, an analysis can be performed on the time series to establish variance or other metrics of how much the time series of one seasoning cycle differs from that of the previous seasoning cycle. In some cases, as seasoning progresses toward completion, the temperature time series or other sensor data may change less between seasoning cycles, so determining the variation in time-series data may provide a useful virtual sensor signal that can be used to assess the progress of seasoning.

[0047]

[0056] Method 200 may also include, in block 210, receiving labeling information that identifies the completion of seasoning for one or more semiconductor processing chambers during multiple seasoning cycles. In some cases, the labeling information may be received as user input to establish the completion date when an expert or other user determines when seasoning is complete based on training characteristics. That is, the label may identify a specific seasoning cycle in which seasoning is completed, or the number of seasoning cycles after seasoning is completed. In some cases, the labeling information may identify a threshold for a virtual sensor measurement, such as variance. Seasoning is automatically considered complete when this threshold is not met, for example, when one or more virtual sensor measurements fall below the threshold consecutively. In some embodiments, additional labeling information may be provided in addition to indicators of a specific seasoning cycle in which seasoning is completed, or several seasoning cycles after seasoning is completed. For example, the labeling information may also include chamber open event data or preventive maintenance event data to establish information about the conditions the semiconductor processing chamber was subjected to before the seasoning process. For example, information regarding environmental conditions (e.g., humidity, temperature) or the duration of a chamber opening event may be used to label training characteristics. Similarly, information regarding specific preventative maintenance performed on the semiconductor processing chamber (e.g., indications of changes in a particular part or cleaning process) may be used to label training characteristics.

[0048]

[0057] Method 200 may also include, in block 215, generating training data from training properties labeled with seasoning completion information, chamber open event data, and / or preventive maintenance data, and in block 220, running a supervised learning algorithm to train a model using the training data. In some cases, the labeled information may be used by the model to construct a formula that characterizes how the training properties change as a function of the seasoning cycle. This allows for an evaluation of predictions about when seasoning will be complete, given a small number of seasoning cycles and optionally based on chamber open event data and / or preventive maintenance data.

[0049]

[0058] Figure 3 shows a block diagram illustrating how virtual sensor data 312, exemplified as variance data, may be generated from characteristics measured during seasoning, according to several embodiments. The semiconductor processing chamber 302 may include a plurality of temperature sensors 304, a plurality of optical sensors 306, and / or a plurality of piezo sensors 308, such as those illustrated or described above with reference to Figure 1. As the seasoning cycle is performed by the semiconductor processing chamber 302, the temperature sensors 304, optical sensors 306, and / or piezo sensors 308 may record sensor data measurements in real time throughout the process. The sensor sampling stage 314 may generate time-series data 310 as a series of consecutive measurements from the temperature sensors 304, optical sensors 306, and / or piezo sensors 308. The time-series data 310 is then processed by a conversion stage 316, in which the time-series data 310 may be converted into virtual sensor data 312.

[0050]

[0059] For example, a seasoning process performed by the semiconductor processing chamber 302 may include multiple seasoning cycles. Each cycle may be associated with a specific set of processing conditions, such as epitaxial growth conditions or etching conditions. In some cases, a single cycle may include two subcycles (e.g., a growth condition subcycle and an etching condition subcycle). During a single cycle or subcycle, the characteristics of the semiconductor processing chamber 302 may be monitored by a temperature sensor 304, an optical sensor 306, and / or a piezo sensor 308. For example, specific processing conditions may include heating the semiconductor processing chamber to a specific temperature setpoint, but the temperature may change over time, such as while the temperature is reaching equilibrium. In some cases, chemical reactions may occur during etching or epitaxial growth, which may affect the temperature or the rate at which the temperature reaches equilibrium. In some examples, generating virtual sensor data may allow for comparison of time-series data of one cycle or subcycle with time-series data of another cycle or subcycle (e.g., the previous cycle or subcycle) to derive variations (e.g., variance) in time-series data between cycles or subcycles.

[0051]

[0060] Figure 4 presents examples of time-series data of temperature measurements at two different locations within a semiconductor processing chamber during several seasoning cycles, according to several embodiments. Only data for the first and every 10 cycles are shown in Figure 4 up to the 50th seasoning cycle; this limited amount of data is shown solely for clarity without obscuring other details. In practice, time-series data of this nature can be obtained for each seasoning cycle.

[0052]

[0061] The time-series data 310, which Figure 4 provides as an example, can be combined in the conversion step 316 to generate virtual sensor data 312. As mentioned above, an example of virtual sensor data that may be useful for evaluating the progress of seasoning in a semiconductor processing chamber is a comparative evaluation of sensor data from a particular seasoning cycle with sensor data from the immediately preceding seasoning cycle. For example, the temperature time-series data of seasoning cycle 2 (not shown in Figure 4) can be directly compared with the temperature time-series data of seasoning cycle 1. In some examples, the difference or ratio of the temperature time-series data can be calculated, resulting in a time series of difference or ratio data that can be used as virtual sensor data 312. In some cases, the time series of difference or ratio data can be combined into a single value, such as the sum or mean of the difference or ratio data, which may provide a single value that characterizes the seasoning cycle and how the time-series data of a particular characteristic changes between seasoning cycles. In other examples, the variance (e.g., the mean of the squared deviations from the mean) can be calculated. In some examples, this single value can be used as virtual sensor data 312. In some cases, multiple different comparisons of time-series data from various sensors can be obtained in this way and further combined according to various weighting coefficients to generate virtual sensor data 312. It should be understood that these are merely some examples of how virtual sensor data 312 may be generated, and are not intended to be limiting.

[0053]

[0062] Figure 5 presents an exemplary bar graph 500 showing virtual sensor data information as a function of the seasoning cycle, provided here as the variance between consecutive cycles for the first 50 cycles. Figure 5 also shows a threshold 510, which may indicate a variance at which seasoning can be considered complete. Figure 5 also shows a label 512 indicating that seasoning is complete around seasoning cycle 23 or 24.

[0054]

[0063] When the virtual sensor data shown in Figure 5 is used as training data, the user may provide inputs corresponding to label 512 that establish a point in time, such as in seasoning cycle 23 or 24, when seasoning is considered complete. For example, the position of label 512 may be input by the user. Additionally or alternatively, the user may establish virtual sensor values ​​at which seasoning is determined to be complete, such as when one or more (e.g., two, three, four, five, six, etc.) virtual sensor values ​​fall below a threshold 510, and provide inputs corresponding to the threshold 510 as labeling information.

[0055]

[0064] Figures 6A, 6B, 6C, and 6D present additional exemplary bar graphs showing virtual sensor data information as a function of the seasoning cycle for various seasoning processes. Figures 6A, 6B, 6C, and 6D also present threshold 610 and labeling information 612, where additional labeling information 614 also includes information regarding chamber open events or preventive maintenance events.

[0056]

[0065] In some cases, sensor data or virtual sensor data similar to those shown in Figures 4, 5, or 6A-6D may be acquired during the seasoning process and displayed in real time on an interface (e.g., a display) during seasoning as the data is collected or processed, allowing the user to monitor the progress of the seasoning. Optionally, the interface may be used, for example, during the training process, to acquire labeling information as input, to enable indicator positioning, to receive selections from drop-down menus, or to receive text input.

[0057]

[0066] Returning to the training method in flowchart 200 of Figure 2, this method may also include generating training data from training characteristics labeled with seasoning completion information. Figure 7 shows flowchart 700 for training a model to predict seasoning completion, according to several embodiments and examples. The training characteristic data 705 may be labeled using corresponding label data 710 for each semiconductor processing chamber. For example, a semiconductor processing chamber may perform a seasoning process and collect characteristic data 705 using one or more sensors. The label data 710 may be determined, for example, via user input. Here, the label data 710 may correspond to or indicate the seasoning cycle number in which seasoning has been determined to be complete, according to, for example, a user or expert's decision. In some cases, the label data 710 indicating the seasoning cycle number in which seasoning has been determined to be complete may be determined automatically, for example, based on one or more characteristic data 705 meeting or reaching a threshold. In some embodiments, the number of seasoning cycles that must meet a threshold before it is determined that seasoning is automatically complete may be adjustable or provided as labeling information. In some embodiments, the threshold may be adjustable or provided as labeling information.

[0058]

[0067] In addition, the chamber opening or preventive maintenance event information 715 can be determined, for example, via user input or derived from one or more data storage systems, and optionally used as additional labeling information for generating training data or training a machine learning model. As described above, the chamber opening or preventive maintenance event information 715 may be useful for associating with several seasoning cycles until seasoning is complete, because the various physical processes associated with the chamber opening or preventive maintenance event may affect several cycles until seasoning is complete. Making this chamber opening or preventive maintenance event available as labeling information may be useful for improving model training.

[0059]

[0068] Characteristic data 705 labeled with either or both of the label data 710 and / or chamber open or preventive maintenance event information 715 can be considered a training set 720 for training the model. Although only a single training set 720 is shown in Figure 7, it should be understood that multiple training sets can be provided to the training algorithm for the model from multiple chambers or from multiple processes from the same chamber.

[0060]

[0069] In the training method of flowchart 200 in Figure 2, the method may also include running a supervised learning algorithm to train a model using training data. Figure 7 shows how training stage 725 may include running training algorithm 730 on model 735. Model 735 may include a machine learning model trained to model the seasoning of a semiconductor processing chamber (e.g., an epitaxial chamber), such as after a chamber open event or a preventive maintenance event. Model 735 may describe the stability of surfaces within the semiconductor processing chamber. Thus, training algorithm 730 may receive characteristic data from component-specific sensors throughout the semiconductor chamber. For example, if training algorithm 730 adjusts the weights or parameters of model 735, these weights or parameters may be trained so that model 735 approximates the passivation behavior of various components of the semiconductor processing chamber between epitaxial growth or etching conditions. Training stage 725 may also be trained as a classification model that uses characteristic data and chamber open or preventive maintenance event information to output classifications based on how well the input data matches the training data. For example, Model 735 can be implemented using a neural network or a convolutional learning network to classify inputs, but this model is not intended to be limited to neural networks. Any type of model can be used, including data-mode models, physical models, and hybrid models.

[0061]

[0070] After the training phase 725 is completed, the process may generate a trained model 740. This trained model 740 may be used to predict whether the seasoning of the semiconductor processing chamber is complete, or other information regarding the completion of seasoning immediately after each seasoning cycle or subcycle is completed. Furthermore, the trained model 740 may be continuously retrained and / or improved over time as new training data becomes available. For example, in some embodiments, characteristic data may be periodically collected from the existing semiconductor processing chamber during seasoning, and confirmation from the user regarding whether the cycling is complete for labeling purposes may be requested or obtained. This information can be combined to form new labeled training data, and the training phase 725 may be repeated continuously or as needed to improve the trained model 740.

[0062]

[0071] Figure 8 shows a flowchart of Method 800 for seasoning a semiconductor processing chamber in several embodiments. This method can be performed using the trained model 740 described above. For example, in many different scenarios, it may be useful to use the trained model 740 to characterize the seasoning of a semiconductor processing chamber and to predict whether the seasoning process is complete and / or when the seasoning process will be completed. For example, when a new semiconductor processing chamber is added to a manufacturing facility, there may be concerns about whether the new semiconductor processing chamber can be in a stable state for performing controllable processing (e.g., etching or epitaxial growth). The trained model 740 may be used to determine how long it may take for the new semiconductor processing chamber to be seasoned and ready for use in manufacturing. Similarly, when an existing semiconductor processing chamber is used or released, or when preventive maintenance is performed, the trained model 740 may be used to determine how long it may take for the semiconductor processing chamber to be seasoned and ready for use in production again. For example, during one or more seasoning cycles, characteristic data can be acquired and provided as input to the trained model 735. The output of the trained model 735 may then indicate whether seasoning is complete, the expected number of seasoning cycles until seasoning is complete, and / or the duration until seasoning is complete.

[0063]

[0072] The method may include performing a seasoning cycle or subcycle using a semiconductor processing chamber, or causing a semiconductor processing chamber to perform a seasoning cycle or subcycle (805). This process is, for example, a process in which the temperature, pressure, and gas component mixture inside the chamber fluctuate or are fluctuated during processing, for example, to establish etching conditions or epitaxial growth conditions. This process is the same process described above in Figures 3 to 5 and can be used to generate various characteristic data. For example, the process may include sequentially establishing etching and growth conditions inside the semiconductor processing chamber. This may passivate the surface inside the semiconductor processing chamber and optionally remove undesirable impurities from inside the semiconductor processing chamber. The seasoning cycle is performed without a semiconductor substrate (e.g., wafer) present inside the semiconductor processing chamber and can repeatedly grow and / or etch material on the surface of other components of the semiconductor processing chamber, building a passivation layer that can serve to provide stable conditions inside the semiconductor processing chamber during etching or growth conditions.

[0064]

[0073] The method may also include determining the characteristics of the semiconductor processing chamber during a seasoning cycle, such as temperature measurements, optical measurements, and piezoelectric measurements (810). As described above, temperature sensors, optical sensors, or piezoelectric sensors may be associated with individual components and / or locations within the semiconductor processing chamber. These sensors may record various physical parameters during a seasoning cycle or subcycle as a set of time-series data. The method may optionally include deriving virtual sensor data from the measured characteristics. As described above, virtual sensor data may be derived by comparing time-series data from a given seasoning cycle with time-series data from the immediately preceding seasoning cycle. This may provide a measure of the variability of various characteristics from seasoning cycle to seasoning cycle.

[0065]

[0074] The method may further include providing characteristic data to a trained model (815). As described above, the trained model may be configured to receive characteristic data as input and provide one or more outputs related to the completion of seasoning. The outputs may indicate whether the seasoning process is complete or provide information about the expected completion of the seasoning process. Figure 9 shows a flowchart 900 of using a trained model 904 to predict the completion of seasoning in a semiconductor processing chamber, according to several embodiments. As detailed above, multiple labeled training data may be provided from one or more reference semiconductor processing chambers 902 to train or retrain the trained model 904. When the semiconductor processing chamber 906 is exposed to a chamber open event or a preventive maintenance event, the process described above may be performed to determine when seasoning is complete, without requiring the semiconductor processing chamber 906 to perform seasoning and undergo excessive seasoning cycles beyond what is necessary to ensure proper seasoning. During various seasoning cycles, characteristic data 908 may be acquired and provided as input to the trained model 904.

[0066]

[0075] In some embodiments, the trained model 904 may also be provided with chamber open event or preventive maintenance event information 910 as a secondary input. This information is provided supplementally and may be used by the trained model 904 to more accurately determine the completion of seasoning or to estimate the time or number of seasoning cycles required until seasoning completion is achieved. For example, the open event or preventive maintenance event information 910 can improve the accuracy of the trained model 904 because this additional information can lead to a more complex understanding of the amount of seasoning cycles that may be required to perform appropriate passivation treatment on any surfaces that may have been introduced during the chamber open event or preventive maintenance event, and / or to remove impurities.

[0067]

[0076] Returning to Figure 8, method 800 may further include receiving output from a trained model that provides seasoning completion information generated based on the provided characteristics (820). Referring to Figure 9, the output 912 of the trained model 904 may include binary values ​​(e.g., 0 / 1 or true / false) and / or scalar values ​​representing whether the seasoning is complete, the difference level, the deviation from the threshold indicating completion, the number of additional seasoning cycles expected to be required until the seasoning is complete, the expected duration until the seasoning is complete, or the expected time for the seasoning to be complete.

[0068]

[0077] Method 800 may further include determining, based on the output, whether the seasoning of the semiconductor processing chamber is complete (825). In some examples, the output may directly indicate whether the seasoning is complete, in the form of a binary value (yes / no). In other examples, the output may indicate, in the form of an integer value, the number of additional seasoning cycles expected until the seasoning is complete. The integer value may be zero if the seasoning is complete, and 1 or 2 or more if the seasoning is not complete.

[0069]

[0078] Even if it is not determined that the seasoning is complete, method 800 may return to step 805, where the semiconductor processing chamber performs an additional seasoning cycle, then measures characteristics during the additional seasoning cycle (810), provides additional characteristics to the trained machine learning model (815), receives output from the trained machine learning model based on the additional characteristics (820), and then evaluates again whether the seasoning is complete (825). If it is determined that the seasoning is complete, method 800 may then proceed to generate an indicator that identifies the completion of the seasoning (830).

[0070]

[0079] Referring again to Figure 9, an indicator of seasoning completion or an indicator related to seasoning completion may optionally be generated at any point during the seasoning process. The indicator may provide notification of whether seasoning is complete, an estimated time or number of additional seasoning cycles until seasoning is complete, an estimated duration of the seasoning process, or a timer countdown, which may optionally be updated in real time as additional characteristic data is acquired and the seasoning model is evaluated. Optionally, the indicator may provide characteristic data, such as in the form of time-series data, and such information may be provided to experts or other users, for example, for troubleshooting or other informational purposes. In some embodiments, the indicator may provide virtual sensor data, such as variability data, to show the progress of seasoning in a format that is more easily understood by experts or other users.

[0071]

[0080] It should be understood that the specific steps shown in Figures 2 and 8 provide a particular method for training and using a model to predict the performance of semiconductor processing chamber seasoning, according to various embodiments. Steps in other sequences may also be performed according to alternative embodiments. For example, alternative embodiments may perform the steps described above in a different order. Furthermore, the individual steps shown in Figures 2 and 8 include multiple substeps, which may be performed in various sequences depending on the individual step. In addition, additional steps may be added or removed depending on the particular application. Many variations, modifications, and alternatives are also included in the scope of this disclosure.

[0072]

[0081] Each of the methods described herein may be implemented by a computer system. Each step of these methods may be performed automatically by the computer system and / or provided with user-involved inputs / outputs. For example, a user may provide inputs for each step in the method. Each of these inputs may be in response to a particular output requesting such input, where the output is generated by the computer system. Each input may be received in response to a corresponding requested output. Furthermore, inputs may be received from a user, received as a data stream from another computer system, retrieved from a memory location, retrieved over a network, requested from a web service, and so on. Similarly, outputs may be provided to a user, provided as a data stream to another computer system, stored in a memory location, sent over a network, provided to a web service, and so on. In short, each step of the methods described herein may be performed by a computer system and may involve any number of inputs, outputs, and / or requests to and from the computer system. These may or may not involve a user. The above steps, which do not involve a user, can be said to be performed automatically by a computer system without human intervention. Therefore, in light of this disclosure, it will be understood that each step of each method described herein may be modified to include input and output with a user, or may be performed automatically by a computer system without human intervention, with any decisions made by a processor. Furthermore, some embodiments of each of the methods described herein may be implemented as a set of instructions stored in a tangible, non-temporary storage medium to form a tangible software product.

[0073]

[0082] Figure 10 shows an exemplary computer system 1000 in which various embodiments can be implemented. System 1000 can be used to implement any of the computer systems described above. As shown in the figure, computer system 1000 includes a processing unit 1004 that communicates with several peripheral subsystems via a bus subsystem 1002. These peripheral subsystems may include a processing acceleration unit 1006, an I / O subsystem 1008, a storage subsystem 1018, and a communication subsystem 1024. The storage subsystem 1018 includes a tangible computer-readable storage medium 1022 and system memory 1010.

[0074]

[0083] The bus subsystem 1002 provides a mechanism for various components and subsystems of the computer system 1000 to communicate with each other as intended. Although the bus subsystem 1002 is schematically illustrated as a single bus, alternative embodiments of the bus subsystem may utilize multiple buses. The bus subsystem 1002 may be one of several types of bus structures, including a memory bus or memory controller, a peripheral bus, and a local bus using any of the various bus architectures. For example, such architectures may include the Industry Standard Architecture (ISA) bus, the Microchannel Architecture (MCA) bus, the Extended ISA (EISA) bus, the Video Electronics Standards Institute (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus, which may be implemented as a mezzanine bus manufactured to the IEEE P1386.1 standard.

[0075]

[0084] The processing unit 1004, which can be implemented as one or more integrated circuits (e.g., conventional microprocessors or microcontrollers), controls the operation of the computer system 1000. One or more processors may be included in the processing unit 1004. These processors may include single-core processors or multi-core processors. In some embodiments, the processing unit 1004 is implemented as one or more independent or sub-processing units 1032 and / or 1034, each processing unit may include a single or multi-core processor. In other embodiments, the processing unit 1004 may also be implemented as a quad-core processing unit formed by integrating two dual-core processors onto a single chip.

[0076]

[0085] In various embodiments, the processing unit 1004 can execute various programs in response to program code and can maintain multiple programs or processes running simultaneously. At a given time, some or all of the program code being executed may reside in one or more processors 1004 and / or in the memory subsystem 1018. Through suitable programming, one or more processors 1004 can provide the various functions described above. The computer system 1000 may additionally include a processing acceleration unit 1006. This processing acceleration unit 1006 may include a digital signal processor (DSP), a special-purpose processor, and the like.

[0077]

[0086] The I / O subsystem 1008 may include user interface input devices and user interface output devices. User interface input devices may include pointing devices such as keyboards, mice or trackballs, touchpads or touchscreens integrated into displays, scroll wheels, click wheels, dials, buttons, switches, keypads, voice input devices including voice command recognition systems, microphones, and other types of input devices. In addition, user interface input devices may include voice recognition sensing devices that allow the user to interact with the voice recognition system through voice commands. User interface input devices may also include, but are not limited to, three-dimensional (3D) mice, joysticks or pointing sticks, gamepads and graphic tablets, as well as AV equipment such as speakers, digital cameras, digital video cameras, portable media players, webcams, image scanners, fingerprint scanners, barcode readers, 3D scanners, 3D printers, laser rangefinders, and eye-tracking devices. Furthermore, user interface input devices may include medical imaging input devices such as computed tomography scanners, magnetic resonance imaging scanners, positional radiography scanners, and medical ultrasound scanners. User interface input devices may also include audio input devices, such as MIDI keyboards and digital musical instruments, augmented reality (AR) input / output devices, and / or virtual reality (VR) input / output devices.

[0078]

[0087] User interface output devices may include non-visual displays such as display subsystems, indicator lights, or audio output devices. Display subsystems may include flat-panel devices such as cathode ray tubes (CRTs), liquid crystal displays (LCDs), or plasma displays, projection devices, touchscreens, etc. Generally, the use of the term “output device” is intended to include all possible types of devices and mechanisms for outputting information from computer system 1000 to a user or another computer. For example, user interface output devices may include, but are not limited to, various display devices that visually convey text, graphics, and audio / video information (e.g., monitors, printers, speakers, headphones, car navigation systems, plotters, audio output devices, and modems).

[0079]

[0088] The computer system 1000 may include a storage subsystem 1018. This storage subsystem 1018 includes software elements that are currently located in the system memory 1010. The system memory 1010 may store program instructions that are loadable and executable on the processing unit 1004, as well as data generated during the execution of these programs.

[0080]

[0089] Depending on the configuration and type of the computer system 1000, the system memory 1010 may be volatile (e.g., random access memory (RAM)) and / or non-volatile (e.g., read-only memory (ROM) or flash memory). RAM typically contains data and / or program modules that are immediately accessible and / or currently being manipulated and executed by the processing unit 1004. In some embodiments, the system memory 1010 may include several different types of memory, such as static random access memory (SRAM) or dynamic random access memory (DRAM). In some embodiments, the basic input / output system (BIOS) (which includes basic routines that help transfer information between elements within the computer system 1000, such as during startup) may be stored, generally in ROM. As an example, but not an limitation, the system memory 1010 may also contain application programs 1012 (which may include client applications, web browsers, intermediate-tier applications, relational database management systems (RDBMS), etc.), program data 1014, and the operating system 1016. For example, Operating System 1016 may include various versions of Microsoft Windows®, Apple Macintosh®, and / or Linux operating systems, various commercially available UNIX® or UNIX-like operating systems (including, but not limited to, various GNU / Linux operating systems, Google Chrome® OS, etc.), and / or mobile operating systems such as iOS, Windows® Phone, Android® OS, BlackBerry® 10 OS, and Palm® OS.

[0081]

[0090] The storage subsystem 1018 may also provide a tangible, computer-readable storage medium for storing basic programming and data structures that provide the functionality of several embodiments. Software (programs, code modules, instructions) that, when executed by the processor, provide the functionality described above, may be stored in the storage subsystem 1018. The software modules or instructions may be executed by the processing unit 1004. The storage subsystem 1018 may also provide a repository for storing data used according to several embodiments.

[0082]

[0091] The memory subsystem 1018 may also include a computer-readable storage medium reader 1020, which may be further connected to the computer-readable storage medium 1022. Together with the system memory 1010, and optionally in combination with the system memory 1010, the computer-readable storage medium 1022 may comprehensively represent remote, local, fixed, and / or removable storage devices and storage media for storing, transmitting, and retrieving computer-readable information, both temporarily and / or more permanently.

[0083]

[0092] The computer-readable storage medium 1022, which includes code or a portion of code, may also include any suitable medium, including storage and communication media, such as volatile and non-volatile, removable and non-removable media, which are implemented in any method or technique for storing and / or transmitting information. This may include tangible computer-readable storage media (e.g., RAM, ROM, electronically erasable programmable ROM (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD), or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices), or other tangible computer-readable media. This may also include non-tangible computer-readable media, which may be any other medium that can be used to transmit data signals, data transmission signals, or desired information and is accessible to the computer system 1000.

[0084]

[0093] For example, computer-readable storage media 1022 may include hard disk drives that read from or write to non-removable non-volatile magnetic media, magnetic disk drives that read from or write to removable non-volatile magnetic disks, and optical disk drives that read from or write to removable non-volatile optical disks, such as CD-ROMs, DVDs, and Blu-ray® discs, or other optical media. Computer-readable storage media 1022 may also include, but are not limited to, Zip® drives, flash memory cards, Universal Serial Bus (USB) flash drives, Secure Digital (SD) cards, DVD discs, digital videotapes, and the like. The computer-readable storage medium 1022 may also include non-volatile memory-based SSDs such as flash memory-based solid drives (SSDs), enterprise flash drives, and solid ROMs, volatile memory-based SSDs such as solid RAM, dynamic RAM, static RAM, DRAM-based SSDs, and magnetoresistive RAM (MRAM) SSDs, and hybrid SSDs that use a combination of DRAM-based SSDs and flash memory-based SSDs. The disk drive and its associated computer-readable medium may provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for the computer system 1000.

[0085]

[0094] The communication subsystem 1024 provides interfaces to other computer systems and networks. The communication subsystem 1024 functions as an interface for transmitting data from computer system 1000 to other systems and receiving data from other systems. For example, the communication subsystem 1024 may enable computer system 1000 to connect to one or more devices via the Internet. In some embodiments, the communication subsystem 1024 may include radio frequency (RF) transceiver components for accessing wireless voice and / or data networks (using, for example, cellular technology, 3G, 4G, or EDGE (enhanced data rates for global evolution)), Global Positioning System (GPS) receiving components, and / or other components. In some embodiments, the communication subsystem 1024 may provide wired network connectivity (e.g., Ethernet) in addition to or instead of the wireless interface.

[0086]

[0095] In some embodiments, the communication subsystem 1024 may also receive input communications in the form of structured and / or unstructured data feeds 1026, event streams 1028, event updates 1030, etc., for one or more users who may use the computer system 1000.

[0087]

[0096] For example, the communication subsystem 1024 may be configured to receive data feeds 1026 in real time from users of social networks and / or other communication services, such as web feeds like RSS (Rich Site Summary) feeds and / or real-time updates from one or more third-party sources.

[0088]

[0097] Furthermore, the communication subsystem 1024 may also be configured to receive data in the form of a continuous data stream. This data may include an event stream 1028 of real-time events and / or event updates 1030, which may have no explicit end and be substantially continuous or infinite. Examples of applications that generate continuous data may include, for example, sensor data applications, financial tickers, network performance measurement tools (e.g., network monitoring and traffic management applications), clickstream analysis tools, and automotive traffic monitoring.

[0089]

[0098] The communication subsystem 1024 may also be configured to output structured and / or unstructured data feeds 1026, event streams 1028, event updates 1030, etc., to one or more databases that can communicate with one or more streaming data source computers coupled to the computer system 1000.

[0090]

[0099] Computer system 1000 can be one of a variety of types, including handheld portable devices (e.g., iPhone® mobile phones, iPad® computing tablets, PDAs), wearable devices, PCs, workstations, mainframes, kiosks, server racks, or any other data processing systems.

[0091]

[0100] Due to the ever-changing nature of computers and networks, the description of computer system 1000 shown in the figure is merely an example. Many other configurations are possible, having more or fewer components than the system shown in the figure. For example, customized hardware may be used, and / or certain elements may be implemented in hardware, firmware, software (including applets), or a combination thereof. Furthermore, connections to other computing devices, such as network input / output devices, may be employed. Based on the disclosures and teachings presented herein, other ways and / or methods for implementing various embodiments should be made apparent.

[0092]

[0101] As used herein, the terms “about,” “approximately,” or “substantially” should be interpreted as being within the range expected by those skilled in the art in light of this specification.

[0093]

[0102] In the above description, for the sake of clarity, numerous specific details have been included to provide a complete understanding of various embodiments. However, it will be apparent that some embodiments can be carried out without some of these specific details. In other examples, well-known structures and devices are shown in the form of block diagrams.

[0094]

[0103] The above description provides only illustrative embodiments and does not limit the scope, applicability, or configuration of this disclosure. Rather, the foregoing description of various embodiments will provide a feasible disclosure for implementing at least one embodiment. It should be understood that various modifications may be made to the function and arrangement of elements without departing from the spirit and scope of some embodiments, as described in the appended claims.

[0095]

[0104] In the preceding description, certain details are provided to facilitate a complete understanding of the embodiments. However, it will be understood that embodiments can be carried out even without these specific details. For example, circuits, systems, networks, processes, and other components may be shown as components in the form of block diagrams to avoid obscuring the embodiments with unnecessary details. In other examples, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary details to avoid obscuring the embodiments.

[0096]

[0105] Furthermore, note that individual embodiments have been described as processes, shown as flowcharts, flow diagrams, data flow diagrams, structural diagrams, or block diagrams. While flowcharts may describe steps as sequential processes, many of these steps can be executed in parallel or simultaneously. Moreover, the order of the steps may be rearranged. A process terminates when a step is completed, but it may have additional steps not shown in the diagram. A process can correspond to a method, function, procedure, subroutine, subprogram, etc. If a process corresponds to a function, its termination may correspond to the function's return to the calling function or main function.

[0097]

[0106] The term “computer-readable medium” includes, but is not limited to, portable or fixed-storage devices, optical storage devices, wireless channels, and various other media capable of storing, containing, or carrying one or more instructions and / or data. A code segment or machine-executable instruction may represent a procedure, function, subprogram, program, routine, subroutine, module, software package, class, or any combination of instructions, data structures, or program statements. A code segment may be connected to another code segment or hardware circuit by passing information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc., may be passed, transferred, or transmitted via any suitable means, including memory sharing, message passing, token passing, network transmission, etc.

[0098]

[0107] Furthermore, embodiments may be implemented by hardware, software, firmware, middleware, microcode, hardware description language, or any combination thereof. When implemented by software, firmware, middleware, or microcode, program code or code segments for performing the required tasks may be stored in a machine-readable medium. One or more processors may perform the required tasks.

[0099]

[0108] While the features are described in the above specification with reference to specific embodiments, it should be noted that not all embodiments are limited thereto. Various features and aspects of several embodiments may be used individually or collectively. Furthermore, embodiments may be used in any number of environments and applications other than those described herein without departing from the broader spirit and scope of this specification. Accordingly, this specification and the drawings should be considered illustrative, not restrictive.

[0100]

[0109] Furthermore, for explanatory purposes, the methods have been described in a specific order. It should be understood that in alternative embodiments, these methods may be executed in an order different from that described. Furthermore, the methods described above may be executed by hardware components or embodied by a sequence of machine-executable instructions. It should be understood that these instructions can be used to cause a machine (e.g., a general-purpose or special-purpose processor, or a logic circuit programmed with instructions) to execute the methods described above. These machine-executable instructions may be stored in one or more machine-readable media (e.g., a CD-ROM or other type of optical disk, a floppy diskette, ROM, RAM, EPROM, EEPROM, magnetic or optical card, flash memory, or other type of machine-readable media suitable for storing electronic instructions). Alternatively, the methods may be executed by a combination of hardware and software.

Claims

1. A method for seasoning a semiconductor processing chamber, Performing a seasoning cycle using a semiconductor processing chamber, The characteristics of the semiconductor processing chamber are determined during the seasoning cycle, The aforementioned characteristics are provided to a trained machine learning model configured to receive characteristics of a semiconductor processing chamber as input and to provide an output regarding the completion of seasoning of the semiconductor processing chamber. Receiving the output from the trained machine learning model generated based on the aforementioned characteristics, Based on the output, determining that the seasoning of the semiconductor processing chamber is not complete, wherein the output indicates the expected completion time of the seasoning of the semiconductor processing chamber, the expected duration of the seasoning of the semiconductor processing chamber, or the expected number of additional seasoning cycles required to complete the seasoning of the semiconductor processing chamber. Performing additional seasoning cycles using the aforementioned semiconductor processing chamber, Determining additional characteristics of the semiconductor processing chamber during the additional seasoning cycle, To provide the characteristics of the semiconductor processing chamber during the additional seasoning cycle to the trained machine learning model, Receiving additional output from the aforementioned trained machine learning model, Based on the aforementioned additional output, it is determined that the seasoning of the semiconductor processing chamber is complete. To generate an indicator that identifies the completion of seasoning of the semiconductor processing chamber. Methods that include...

2. The method according to claim 1, further comprising generating an indicator that identifies one or more of the following: an expected completion time for seasoning the semiconductor processing chamber, an expected duration for seasoning the semiconductor processing chamber, or an expected number of additional seasoning cycles for completing the seasoning of the semiconductor processing chamber.

3. Based on the output, it is determined that the seasoning of the semiconductor processing chamber is not yet complete. To generate an indicator that identifies the expected completion time of seasoning the semiconductor processing chamber, the expected duration of seasoning the semiconductor processing chamber, or the expected number of additional seasoning cycles required to complete the seasoning of the semiconductor processing chamber. The method according to claim 1, further comprising:

4. Performing a seasoning cycle is Controlling the semiconductor processing chamber to establish etching conditions within the semiconductor processing chamber, Controlling the semiconductor processing chamber to establish epitaxial growth conditions within the semiconductor processing chamber. The method according to claim 1, including the method described in claim 1.

5. The method according to claim 1, wherein the characteristics include one or more of the following: the temperature in the semiconductor processing chamber during the seasoning cycle, the thickness of the epitaxial layer produced in the semiconductor processing chamber, the power supplied to the heater associated with the semiconductor processing chamber, the power setting of the heater associated with the semiconductor processing chamber, the growth rate in the semiconductor processing chamber during the seasoning cycle, the etching rate in the semiconductor processing chamber during the seasoning cycle, the optical conditions in the semiconductor processing chamber during the seasoning cycle, the pressure in the semiconductor processing chamber during the seasoning cycle, the gas composition in the semiconductor processing chamber during the seasoning cycle, the amount of gas flowing into or out of the semiconductor processing chamber during the seasoning cycle, changes in any of these, physical or structural parameters associated with the semiconductor processing chamber, or any combination thereof.

6. The process further includes training the machine learning model before providing the aforementioned characteristics to the machine learning model, and training the machine learning model is Receiving training characteristics from multiple executions of a seasoning cycle performed by one or more semiconductor processing chambers, The process of generating training data based on the aforementioned training characteristics, wherein the training data is generated using labeling information that identifies the completion of seasoning of one or more semiconductor processing chambers. To train the machine learning model using the aforementioned training data, a supervised learning algorithm is executed. The method according to claim 1, including the method described in claim 1.

7. The one or more semiconductor processing chambers and the semiconductor processing chamber are the same chamber, The method according to claim 6, wherein determining that the seasoning of the semiconductor processing chamber is complete includes comparing the characteristics of the semiconductor processing chamber with predetermined characteristics of the semiconductor processing chamber.

8. The method according to claim 1, wherein the machine learning model compares the characteristics of a seasoning cycle with the corresponding characteristics of the immediately preceding seasoning cycle in order to evaluate the completion of seasoning of the semiconductor processing chamber.

9. It is a system, One or more processors, One or more non-temporary computer-readable storage devices containing instructions, wherein the instructions, when executed by the one or more processors, To execute a seasoning cycle in the semiconductor processing chamber, The characteristics of the semiconductor processing chamber are determined during the seasoning cycle, The aforementioned characteristics are provided to a trained machine learning model configured to receive characteristics of a semiconductor processing chamber as input and to provide an output regarding the completion of seasoning of the semiconductor processing chamber. Receiving the output from the aforementioned trained machine learning model, Based on the output, determining that the seasoning of the semiconductor processing chamber is not complete, wherein the output indicates the expected completion time of the seasoning of the semiconductor processing chamber, the expected duration of the seasoning of the semiconductor processing chamber, or the expected number of additional seasoning cycles required to complete the seasoning of the semiconductor processing chamber. Performing an additional seasoning cycle in the semiconductor processing chamber, The determination of the characteristics of the semiconductor processing chamber continues during the additional seasoning cycle, To provide the characteristics of the semiconductor processing chamber during the additional seasoning cycle to the trained machine learning model, Receiving additional output from the aforementioned trained machine learning model, Based on the aforementioned additional output, it is determined that the seasoning of the semiconductor processing chamber is complete. To generate an indicator that identifies the completion of seasoning of the semiconductor processing chamber. A non-temporary computer-readable storage device that causes one or more processors to perform a process including the above A system that includes these features.

10. The system according to claim 9, further comprising one or more sensors that communicate data with one or more processors to determine the characteristics of the semiconductor processing chamber during a seasoning cycle.

11. The aforementioned characteristics include time-series data acquired from one or more sensors, or The system according to claim 10, wherein the characteristic includes a comparison using a first time series of data acquired from one or more sensors during a first seasoning cycle and a second time series of data acquired from one or more sensors during a second seasoning cycle immediately preceding the first seasoning cycle.

12. The aforementioned seasoning cycle is Controlling the semiconductor processing chamber to establish etching conditions within the semiconductor processing chamber, Controlling the semiconductor processing chamber to establish epitaxial growth conditions within the semiconductor processing chamber. The system according to claim 9, including the system described in claim 9.

13. The system according to claim 9, wherein the semiconductor processing chamber includes one or more of a quartz dome, a flow module, or a susceptor.

14. The system according to claim 13, further comprising one or more sensors positioned to measure the properties of one or more surfaces of the quartz dome, the flow module, the susceptor, or the substrate in the semiconductor processing chamber.

15. A non-temporary computer-readable medium containing instructions, wherein when the instructions are executed by one or more processors, To execute a seasoning cycle in the semiconductor processing chamber, The characteristics of the semiconductor processing chamber are determined during the seasoning cycle, The aforementioned characteristics are provided to a trained machine learning model configured to receive characteristics of a semiconductor processing chamber as input and to provide an output regarding the completion of seasoning of the semiconductor processing chamber. Receiving the output from the aforementioned trained machine learning model, Based on the output, determining that the seasoning of the semiconductor processing chamber is not complete, wherein the output indicates the expected completion time of the seasoning of the semiconductor processing chamber, the expected duration of the seasoning of the semiconductor processing chamber, or the expected number of additional seasoning cycles required to complete the seasoning of the semiconductor processing chamber. Performing an additional seasoning cycle in the semiconductor processing chamber, The determination of the characteristics of the semiconductor processing chamber continues during the additional seasoning cycle, To provide the characteristics of the semiconductor processing chamber during the additional seasoning cycle to the trained machine learning model, Receiving additional output from the aforementioned trained machine learning model, Based on the aforementioned additional output, it is determined that the seasoning of the semiconductor processing chamber is complete. To generate an indicator that identifies the completion of seasoning of the semiconductor processing chamber. A non-temporary computer-readable medium that causes one or more processors to perform a process including the above.

16. The non-temporary computer-readable medium according to claim 15, wherein the trained machine learning model is trained to model the seasoning of the semiconductor processing chamber following a chamber open event or preventive maintenance event related to the semiconductor processing chamber.

17. The non-temporary computer-readable medium according to claim 16, further comprising obtaining user input corresponding to one or more of the following: the duration of the chamber opening event or the preventive maintenance event; changes in the components of the semiconductor processing chamber during the chamber opening event or the preventive maintenance event; environmental conditions during the chamber opening event or the preventive maintenance event; maintenance procedures or protocols applied to the semiconductor processing chamber during the chamber opening event or the preventive maintenance event; or a severity rating of the chamber opening event or the preventive maintenance event, wherein the trained machine learning model is further configured to use the user input when generating the output relating to the completion of seasoning of the semiconductor processing chamber.

18. A non-temporary computer-readable medium according to claim 15, wherein the output from the trained machine learning model includes an indication of whether the seasoning of the semiconductor processing chamber is complete or incomplete, and one or more variance values ​​determined by comparing the characteristics of the semiconductor processing chamber during a first seasoning cycle with the characteristics of the semiconductor processing chamber during a second seasoning cycle immediately preceding the first seasoning cycle.

19. Based on the aforementioned additional output, it is determined that the seasoning of the semiconductor processing chamber is complete. To derive the variance value between the aforementioned output and the additional output, The determination that the aforementioned variance is smaller than the target variance. A non-temporary computer-readable medium according to claim 15, including the following:

20. The non-temporary computer-readable medium according to claim 19, further comprising the step of receiving an input corresponding to an acceptable value for the trained machine learning model that indicates an acceptable deviation from the target variance value.

Images