Automated control of process chamber components

Abstract

Methods, systems, and media for deposition control in a processing chamber are provided herein. In some embodiments, a method includes (a) obtaining information of a current point in time indicating a state of one or more components of the processing chamber during performing a deposition process on one or more wafers. The method may include (b) determining whether to adjust one or more control components of the processing chamber by providing an input based on the obtained information to a trained machine learning model configured to determine that adjustment is to be output, wherein the adjustments to the one or more control components cause variations in the deposition process. The method may include (c) transmitting instructions to a controller of the processing chamber to drive the adjustments to the one or more control components to be implemented.

Description

Incorporated by Reference

[0001] The PCT application form is filed concurrently with this specification as a part of this application. Each application identified in the concurrently filed PCT application form to which this application claims the benefit or priority is incorporated herein by reference in its entirety and for all purposes. Background Art

[0002] A multi-station processing chamber may suffer from imbalance issues between the multiple stations of the multi-station processing chamber. For example, there may be differences in gas flow, transmission power, etc. for different stations. This imbalance may cause undesirable differences in substrates being manufactured in different stations. For example, there may be differences in deposition thickness, etch depth, etc. In addition, it may be an inefficient use of resources to operate each station in the same manner. Therefore, it may be desirable to be able to control components of different stations of a multi-station processing chamber individually.

[0003] The background description provided here is for the purpose of generally presenting the context of the present disclosure. The work of the presently designated inventors is neither explicitly nor implicitly admitted to be prior art against the present disclosure to the extent that it is described in this background section and to aspects of the specification that could not be determined as prior art at the time of filing the application. Summary of the invention

[0004] Provided herein are automated control systems, apparatus, and methods for multi-station process chamber components.

[0005] In some embodiments, a method includes: (a) obtaining information at a current point in time, the information indicating the state of one or more components of the process chamber during a deposition process performed on one or more wafers, wherein the deposition process includes multiple deposition cycles performed in the process chamber. The method includes (b) determining whether to adjust one or more control components of the process chamber by providing an input based on the information obtained to a trained machine learning model configured to determine the adjustment as an output, wherein the adjustment of the one or more control components causes a change in the deposition process. The method includes (c) in response to determining that the one or more control components are adjusted, transmitting instructions to a controller of the process chamber, thereby causing the adjustment of the one or more control components to be implemented. The method includes (d) repeating (a) to (c) until the deposition process is completed.

[0006] In some examples, the process chamber is a multi-station process chamber. In some examples, the adjustment of the one or more control components causes a change in the deposition process that occurs in a first station of the multi-station process chamber relative to a second station of the multi-station process chamber. In some examples, in (d), (a) to (c) are repeated until the deposition process has been completed in each station of the multi-station process chamber.

[0007] In some examples, the process chamber is a single-station process chamber.

[0008] In some examples, the machine learning model is configured to determine the adjustment based at least in part on a determination of a plurality of predicted wafer characteristics of the one or more wafers undergoing the deposition process at the current point in time. In some examples, the predicted characteristics of the one or more wafers include a deposition thickness of each of the one or more wafers.

[0009] In some examples, determining whether to make adjustments to the one or more control components of the process chamber includes comparing the predicted characteristic for a given wafer to a target characteristic for the given wafer. In some examples, the predicted characteristic includes a virtual metrology measurement.

[0010] In some examples, the one or more components of the process chamber include one or more valves associated with one or more manifolds of the process chamber, each manifold being configured to flow gas to a station of the process chamber. In some examples, the information indicative of the state includes a duration that each of the one or more valves has been open. In some examples, the input based on the obtained information provided to the trained machine learning model includes an amount of gas provided to a given station operably coupled to a manifold of the one or more manifolds based on the duration that the corresponding valve has been open. In some examples, the method also includes determining the amount of gas based on the duration that the corresponding valve has been open and a gas flow rate.

[0011] In some examples, the trained machine learning model is configured to take as input at least one of: chamber pressure information; gas pressure information; ampoule temperature; non-ampoule gas feed; or carrier gas flow rate.

[0012] In some examples, obtaining the information in (a) is performed at a sampling rate greater than about 100 Hz.

[0013] In some examples, the one or more control components include one or more diverter valves that divert gas flowing through a manifold from a station of the process chamber.

[0014] In some examples, the process chamber is a multi-station process chamber, and wherein the adjustment of the one or more control components includes reducing the flow rate of the gas to the first station relative to the flow rate of the gas to the second station.

[0015] In some examples, the processing chamber is a multi-station processing chamber including at least a first station and a second station, and wherein the adjustment of the one or more control components includes changing a radio frequency (RF) power used to generate a plasma associated with the deposition process for the first station relative to the second station.

[0016] In some examples, the method further comprises: (e) obtaining post-processing metrology data for at least one of the one or more wafers; and (f) updating the trained machine learning model using the post-processing metrology data. In some examples, the post-processing metrology data comprises at least one of: resistivity data; quality of thin films grown for the at least one wafer; Fourier transform infrared (FTIR) spectral peaks; thickness of films deposited on the at least one wafer; refractive index; stress at a local location on the at least one wafer; stress associated with wafer bow of the at least one wafer; particle information; or material dielectric constant information.

[0017] In some examples, the method further includes: (e) determining a degradation state of at least one of the one or more components of the process chamber based on the predicted characteristic.

[0018] According to some embodiments, a computer program product is provided, comprising a non-transitory computer-readable medium on which computer-executable instructions are provided. The instructions may include instructions for the following operations: (a) obtaining information at a current point in time, the information indicating the state of one or more components of the process chamber during a deposition process performed on one or more wafers, wherein the deposition process includes a plurality of deposition cycles performed in the process chamber; (b) determining whether to adjust one or more control components of the process chamber by providing an input based on the information obtained to a trained machine learning model configured to determine the adjustment as an output, wherein the adjustment of the one or more control components causes a change in the deposition process; (c) in response to determining that the one or more control components are adjusted, transmitting instructions to a controller of the process chamber, thereby causing the adjustment of the one or more control components to be implemented; and (d) repeating (a) to (c) until the deposition process is completed.

[0019] In some examples, the process chamber is a multi-station process chamber. In some examples, the adjustment of the one or more control components causes a change in the deposition process that occurs in a first station of the multi-station process chamber relative to a second station of the multi-station process chamber. In some examples, in (d), (a) to (c) are repeated until the deposition process has been completed in each station of the multi-station process chamber.

[0020] In some examples, the process chamber is a single-station process chamber.

[0021] In some examples, the machine learning model is configured to determine the adjustment based at least in part on a determination of a plurality of predicted wafer characteristics of the one or more wafers undergoing the deposition process at the current point in time. In some examples, the predicted characteristics of the one or more wafers include a deposition thickness of each of the one or more wafers.

[0022] In some examples, determining whether to make adjustments to the one or more control components of the process chamber includes comparing the predicted characteristic for a given wafer to a target characteristic for the given wafer. In some examples, the predicted characteristic includes a virtual metrology measurement.

[0023] In some examples, the one or more components of the process chamber include one or more valves associated with one or more manifolds of the process chamber, each manifold being configured to flow gas to a station of the process chamber. In some examples, the information indicative of the state includes a duration that each of the one or more valves has been open. In some examples, the input based on the obtained information provided to the trained machine learning model includes an amount of gas provided to a given station operably coupled to a manifold of the one or more manifolds based on the duration that the corresponding valve has been open. In some examples, the instructions also include instructions for determining the amount of gas based on the duration that the corresponding valve has been open and a gas flow rate.

[0024] In some examples, the trained machine learning model is configured to take as input at least one of: chamber pressure information; gas pressure information; ampoule temperature; non-ampoule gas feed; or carrier gas flow rate.

[0025] In some examples, obtaining the information in (a) is performed at a sampling rate greater than about 100 Hz.

[0026] In some examples, the one or more control components include one or more diverter valves that divert gas flowing through a manifold from a station of the process chamber.

[0027] In some examples, the process chamber is a multi-station process chamber, and wherein the adjustment of the one or more control components includes reducing the flow rate of the gas to the first station relative to the flow rate of the gas to the second station.

[0028] In some examples, the processing chamber is a multi-station processing chamber including at least a first station and a second station, and wherein the adjustment of the one or more control components includes changing a radio frequency (RF) power used to generate a plasma associated with the deposition process for the first station relative to the second station.

[0029] In some examples, the instructions further include instructions for: (e) obtaining post-processing metrology data for at least one of the one or more wafers; and (f) updating the trained machine learning model using the post-processing metrology data. In some examples, the post-processing metrology data includes at least one of: resistivity data; quality of thin films grown for the at least one wafer; Fourier transform infrared (FTIR) spectral peaks; thickness of films deposited on the at least one wafer; refractive index; stress at a local location on the at least one wafer; stress associated with wafer bow of the at least one wafer; particle information; or material dielectric constant information.

[0030] In some examples, the instructions further include instructions for: (e) determining a degradation state of at least one of the one or more components of the process chamber based on the predicted characteristic. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A substrate processing apparatus is shown for depositing or etching a film on or over a semiconductor substrate using any number of processes in accordance with some embodiments.

[0032] Figure 2 is a schematic diagram of an example multi-station processing chamber according to some embodiments.

[0033] Figure 3 is a schematic diagram illustrating various control components associated with an example processing chamber according to some embodiments.

[0034] Figure 4 is a block diagram of a system for automatically controlling process chamber components.

[0035] Figure 5 is a flow chart of an example process for automatically controlling components of a processing chamber.

[0036] Figure 6 An exemplary computer system is presented that can be used to implement certain embodiments described herein.

[0037] Figure 7 An exemplary timing diagram for controlling process chamber components according to some embodiments is shown.

[0038] Figure 8 is a schematic diagram of a system that may be used in conjunction with an exemplary precursor recovery technique, according to some embodiments. DETAILED DESCRIPTION

[0039] In the following description, several specific details are set forth to provide a thorough understanding of the presented embodiments. The disclosed embodiments may be implemented without some or all of these specific details. In other instances, well-known processing operations are not described in detail so as not to unnecessarily obscure the disclosed embodiments. Although the disclosed embodiments will be described in conjunction with specific embodiments, it should be understood that these specific embodiments are not intended to limit the disclosed embodiments.

[0040] In some embodiments, control components associated with the processing chamber can be actuated to control the manufacturing process. For example, in the case where the processing chamber is a multi-station processing chamber, control components associated with each station of the multi-station processing chamber can be actuated separately so that the manufacturing process can be individually controlled in each station. In some embodiments, the manufacturing process can be a deposition process, such as an atomic layer deposition (ALD) process, a chemical vapor deposition process (CVD), etc. In some embodiments, the manufacturing process can be an etching process. In some embodiments, the manufacturing process can be a passivation process, in which the surface composition of the substrate is changed (e.g., using oxidation) to, for example, protect the feature sidewalls of the substrate during a subsequent etching process. In some embodiments, the manufacturing process can be an inhibition process, during which the growth rate of different positions of the feature (e.g., the top of the feature or the bottom of the feature) during the deposition period is changed.

[0041] In some embodiments, the control components may be independently activated for each station to provide uniformity across different stations. For example, the independent activation of the control components may cause the deposition process in a particular station to be stopped or blocked, while the deposition process in other stations continues. For example, by preventing deposition in a station with a faster growth rate while allowing deposition to continue in other stations with a slower growth rate, a more uniform deposition thickness may be achieved across multiple substrates that are deposited in different stations. As another example, the independent activation of the control components may cause the etching process in a particular station to be stopped or blocked while allowing the etching process in other stations to continue. For example, by preventing the etching process in a station with a faster etching rate while allowing the etching process to continue in other stations with a slower etching rate, a more uniform etching depth may be achieved across multiple substrates that are etched in different stations. It should be understood that although some embodiments described herein are described for multi-station processing chambers, the techniques described herein may be applied to single-station modules. For example, in some implementations, the techniques described herein can be applied in a single-station module to divert specific gas species used in a specific step of a recipe while allowing additional pulses, times, etc. to be used with respect to other steps of the recipe, thereby allowing the manufacturing process to be controlled within the single station. In addition, it should be understood that these techniques can be similarly applied to plasma-based and non-plasma-based manufacturing processes. The techniques described herein can be advantageous from an environmental perspective, from a cost and / or resource perspective, etc. For example, the techniques described herein can be used to save (e.g., by diverting) various process gases, which can be beneficial from an environmental and / or cost perspective.

[0042] In some embodiments, the control component may include an independent gas flow valve that operably couples the station to a specific gas source associated with the multi-station processing chamber. For example, the gas flow valve may be associated with a specific manifold used to provide gas from the gas source to a specific station during the manufacturing process, so that when the gas flow valve is set to an "open" or "exit" position, the station receives gas through the manifold; and when the gas flow valve is set to a "closed" or "turn" position, the station does not receive gas through the manifold. In some embodiments, a first gas flow valve associated with a first station and a second gas flow valve associated with a second station can operably couple the first station and the second station to a common gas source via a common manifold. By setting the first gas flow valve to a "closed" or "turn" position and setting the second gas flow valve to an "open" or "exit" position, the first station can be prevented from receiving gas via the manifold, while the second station can receive gas via the manifold. Therefore, through independent control of the first gas flow valve and the second gas flow valve, the manufacturing process can be stopped at the first station and carried out at the second station. It should be noted that similar gas flow valve actuation may be performed for single station modules to control the flow of specific gas species during individual steps of a multi-step recipe, for example.

[0043] In some embodiments, the control component may include independent RF switches that operably couple the stations to an RF generator associated with the multi-station processing chamber. For example, a first RF switch associated with a first station operably couples the first station to the RF generator, while a second RF switch associated with a second station operably couples the second station to the second RF generator. By setting the first RF switch to a "disabled" state while setting the second RF switch to an "enabled" state, the first station may be prevented from receiving RF power from the RF generator, while the second station may receive RF power from the RF generator. Thus, through independent control of the first RF switch and the second RF switch, the manufacturing process may be stopped at the first station and carried out at the second station.

[0044] In some implementations, it can be determined whether to adjust a particular control component based on the output of a trained machine learning model. It should be understood that when used herein, a trained machine learning model can refer to any suitable model type or architecture based on model reasoning, and can include neural networks, Bayesian reasoning models, regression, etc. The machine learning model can be configured to take state information about one or more components of a manufacturing tool as input, and generate predicted wafer characteristics of each wafer manufactured in different stations and / or adjustments to be made to the control components to produce consistency between the wafers being manufactured as output. It should be noted that the machine learning model is run in situ so that the actuation of the control components to produce wafer consistency is performed during the manufacturing process. In some implementations, the state information about one or more components can include valve timing information associated with the valves of the manufacturing tool and / or information derived from the valve timing information. For example, the valves can be valves associated with one or more manifolds, and each manifold is configured to flow gas to the processing station. The machine learning model may take as input the valve opening duration and / or the amount of gas flowing to a particular station (the gas may be a precursor gas used in a deposition step, a gas used in an oxidation step, a gas used in a passivation step, a gas used in an inhibition step, and / or any combination thereof), wherein the amount of gas may be determined based on the valve opening duration and the gas flow rate. Since wafer characteristics (e.g., wafer thickness) may be affected by the amount of gas flowing to the station where the wafer is being processed, the machine learning model is able to predict the wafer thickness at the current time based on the valve opening duration and / or the amount of gas flowing to the station. The predicted wafer characteristics may be considered as "virtual metrology" measurements. Such virtual metrology measurements may be used in conjunction with other applications, for example, as a replacement for various physical metrology measurements, where the acquisition of these physical metrology measurements may require a lot of time and / or resources, may be destructive to the wafer, etc. In addition, it should be noted that while gate timing information is generally used as an example of component measurements that can be used by the machine learning model, other components whose measurements may be used as input to the machine learning model may include RF-related components (e.g., RF switches, RF generators, etc.), pressure information, temperature information, flow rates, etc.

[0045] The machine learning model can continuously receive state information during the manufacturing process, so that the machine learning model can responsively generate information that can be used to modify the performance of the manufacturing process in situ. Unlike relying on process engineers to set multiple deposition cycles for each station before performing a deposition process, the technology described herein can allow for almost real-time in-situ control to allow greater consistency between wafers that are processed in different stations of a multi-station manufacturing tool. In addition, in some embodiments, the machine learning model can be updated iteratively (e.g., periodically) over time, allowing the machine learning model to adapt to system drift. For example, in some embodiments, the machine learning model can be updated using post-processing metrology results performed on a subset of wafers. Such an update can be regarded as an "online" update of the machine learning model. In some embodiments, the trained machine learning model can be used as the initial starting point of another machine learning model, for example, using various transfer learning techniques. For example, transfer learning can be used to establish and / or train another machine learning model used in combination with similar manufacturing tools for different applications.

[0046] Certain implementations may be used in conjunction with a plurality of wafer fabrication processes, such as various plasma enhanced atomic layer deposition (ALD) processes, various plasma enhanced chemical vapor deposition (CVD) processes, or may be used on-the-fly during a single deposition process. In certain implementations, an RF power generator having multiple output ports may be used at any signal frequency, such as a frequency between about 300 kHz and about 60 MHz, which may include frequencies of about 400 kHz, about 1 MHz, about 2 MHz, about 13.56 MHz, and / or about 27.12 MHz. However, in other implementations, an RF power generator having multiple output ports may operate at any signal frequency, which may include, for example, relatively low frequencies between about 50 kHz and about 300 kHz, as well as higher signal frequencies between about 60 MHz and about 100 MHz.

[0047] It should be noted that while the specific implementations described herein may show and / or describe a multi-station semiconductor fabrication chamber having 4 (four) processing stations, these implementations are intended to encompass multi-station integrated circuit fabrication chambers having or utilizing any number of processing stations. Thus, in certain implementations, individual output ports of an RF power generator having multiple output ports may be assigned to a processing station in a multi-station fabrication chamber having, for example, 2 processing stations or 3 processing stations. In other implementations, individual output ports of an RF power generator having multiple output ports may be assigned to a processing station of a multi-station integrated circuit fabrication chamber having a large number of processing stations (e.g., 5 processing stations, 6 processing stations, 8 processing stations, 10 processing stations, or any other number of processing stations). Furthermore, embodiments of the present disclosure are applicable to chambers having only a single processing station. Furthermore, while particular implementations described herein may show and / or describe the use of a single, relatively low frequency RF signal (e.g., a frequency between about 300 kHz and about 2 MHz), as well as a single, relatively high frequency RF signal (e.g., a frequency between about 2 MHz and about 100 MHz), the disclosed implementations are intended to encompass the use of any number of frequencies below about 2 MHz and any number of frequencies above about 2 MHz.

[0048] Figure 1 Substrate processing apparatus 100 is shown for depositing films on or over a semiconductor substrate utilizing any number of processes according to various implementations. Figure 1 The processing apparatus 100 may use a single processing station 102 of a processing chamber having a single substrate holder 108 (e.g., a susceptor) in an interior volume that may be maintained under vacuum by a vacuum pump 118. A showerhead 106 and a gas delivery system 101 (which may be fluidly coupled to the processing chamber) may allow for delivery of film precursors, as well as delivery of, for example, carrier and / or purge gases and / or process gases, auxiliary reactants, and the like. Figure 1 Also shown is the apparatus used to generate the plasma within the processing chamber. Figure 1 The apparatus schematically shown in FIG. 1 may be particularly suitable for performing plasma enhanced CVD.

[0049] In some embodiments, the gas delivery system 101 may include various components for carrying out the process chemistry (e.g., a mixing vessel for mixing and / or conditioning the process gas delivered to the showerhead 106). Certain reactants may be stored in liquid form and delivered to the process station 102 of the process chamber after evaporation. The gas delivery system may include components for evaporating the liquid reactants. In some implementations, a liquid flow controller may be provided to control the mass flow rate of the liquid used for evaporation and delivered to the process station 102.

[0050] The showerhead 106 is operable to distribute process gases and / or reactants (e.g., film precursors) toward the substrate 112 at the processing station, and the flow of the process gases and / or reactants may be controlled by one or more valves upstream of the showerhead. Figure 1 In the illustrated implementation, substrate 112 is shown below showerhead 106 and is shown sitting on a single substrate holder 108. Showerhead 106 may include any suitable shape and may include any suitable number and arrangement of ports to distribute process gases to substrate 112. In some implementations involving two or more stations, gas delivery system 101 includes valves or other flow control structures upstream of the showerhead that can independently control the flow of process gases and / or reactants to each station, thereby allowing gas flow to one station while simultaneously prohibiting gas flow to a second station. In addition, gas delivery system 101 can be configured to independently control the process gases and / or reactants delivered to each station in a multi-station arrangement, such that the gas composition provided to different stations is different; for example, at the same point in time, the partial pressures of gas components between different stations may vary.

[0051] exist Figure 1 106. In some implementations, the gas volume 107 is shown as being located below the showerhead 106. In some implementations, the single substrate holder 108 can be raised or lowered to expose the substrate 112 to the gas volume 107 and / or change the size of the gas volume 107. Optionally, the single substrate holder 108 can be lowered and / or raised during portions of the deposition process to adjust the process pressure, reactant concentration, etc. within the gas volume 107. The showerhead 106 and the single substrate holder 108 are shown as being electrically coupled to the RF signal generator 114 and the matching network 116 to couple power to the plasma generator. Thus, the showerhead 106 can function as an electrode to couple RF power into the process station 102. In some embodiments, the plasma energy is controlled by controlling one or more of the process station pressure, gas concentration, power output of the RF signal generator, etc. (e.g., by a system controller having appropriate machine-readable instructions and / or control logic). For example, the RF signal generator 114 and the matching network 116 can be operated at any suitable RF power level and can be operated to form a plasma having a desired radical species composition. In addition, the RF signal generator 114 can provide RF power having more than one frequency component, such as a low frequency component (e.g., less than about 2 MHz) and a high frequency component (e.g., greater than about 2 MHz).

[0052] In some implementations, plasma ignition and maintenance conditions are controlled by suitable hardware and / or suitable machine-readable instructions in a system controller, which can provide control instructions via a sequence of input / output control instructions. In one example, instructions for initiating ignition or maintaining plasma are provided in the form of a plasma activation portion of a process recipe. In some cases, the process recipe can be arranged in sequence so that at least some instructions for the process can be executed simultaneously. In some implementations, instructions for setting one or more plasma parameters can be included in a recipe before the plasma ignition process. For example, a first recipe may include instructions for setting the flow rate of an inert gas (e.g., helium) and / or a reactant gas, instructions for setting a plasma generator to a power set point, and time delay instructions for the first recipe. A subsequent second recipe may include instructions for enabling a plasma generator, and time delay instructions for the second recipe. A third recipe may include instructions for deactivating a plasma generator, and time delay instructions for the third recipe. It should be understood that these recipes can be further subdivided and / or iterated in any suitable manner within the scope of the present disclosure. In some deposition processes, the duration of plasma ignition may correspond to a duration of several seconds, such as about 3 seconds to about 15 seconds, or may involve longer durations, such as up to about 30 seconds. In certain implementations described herein, shorter plasma ignitions may be applied during a process cycle. These plasma ignition durations may fall on the order of less than about 50 milliseconds, with about 25 milliseconds being used in a particular example.

[0053] In some embodiments, the instructions used by the controller 150 may be provided via input / output control (IOC) sequence instructions. In one example, instructions for setting conditions for a process stage may be included in a corresponding recipe stage of a process recipe. In some cases, the process recipe stages may be arranged in sequence so that at least some instructions for the process may be executed simultaneously. In some embodiments, instructions for setting one or more reactor parameters may be included in a process recipe. For example, a first recipe stage may include instructions for setting the flow rate of an inert gas and / or a reactant gas (e.g., a first precursor), instructions for setting the flow rate of a carrier gas (e.g., argon), instructions for a first RF power level, and time delay instructions for the first recipe stage. A subsequent second recipe stage may include instructions for adjusting or stopping the flow rate of an inert gas and / or a reactant gas, instructions for adjusting the flow rate of a carrier gas or a purge gas, instructions for a second RF power level, and time delay instructions for the second recipe stage. The third recipe stage may include instructions for adjusting the flow rate of the second reactant gas, instructions for adjusting the duration of the flow of the second reactant gas, instructions for adjusting the flow rate of the carrier gas or the purge gas, instructions for a third RF power level, and time delay instructions for the third recipe stage. A subsequent fourth recipe stage may include instructions for adjusting or stopping the flow rate of the inert gas and / or the reactant gas, instructions for adjusting the flow rate of the carrier gas or the purge gas, instructions for a fourth RF power level, and time delay instructions for the fourth recipe stage. It should be understood that these recipes may be further subdivided and / or iterated in any suitable manner within the scope of the disclosed embodiments.

[0054] As described above, one or more processing stations may be included in a multi-station processing tool. Figure 2A schematic diagram of an embodiment of a multi-station processing tool 200 is shown, the multi-station processing tool 200 having an inbound load lock 202 and an outbound load lock 204, one or both of which may include a remote plasma source. A robot 206 at atmospheric pressure is configured to pass a wafer from a wafer boat loaded by a transfer box 208 into the inbound load lock 202 through an atmospheric port 210. The wafer is placed on a pedestal 212 in the inbound load lock 202 by the robot 206, the atmospheric port 210 is closed and the load lock is evacuated. In the case where the inbound load lock 202 includes a remote plasma source, the wafer can be exposed to a remote plasma treatment within the load lock before the wafer is introduced into the processing chamber 214. In addition, the wafer can also be heated in the inbound load lock 202 to, for example, remove moisture and adsorbed gases. Next, the chamber transfer port 216 to the processing chamber 214 is opened and another robot (not shown) places the wafer into the reactor and onto the substrate holder at the first station shown in the reactor for processing. Figure 2 The embodiment shown in includes a load lock, but it will be appreciated that in some embodiments the substrate may be provided directly into a processing station.

[0055] The illustrated processing chamber 214 includes four processing stations, Figure 2 , are numbered from 1 to 4 in the embodiment shown. Each station has a heated pedestal (shown as 218 for station 1), and a gas line inlet. It should be understood that in some embodiments, each processing station may have different or multiple uses. For example, in some embodiments, the processing station can be switched between an ALD processing mode and a plasma enhanced ALD processing mode. Additionally or alternatively, in some embodiments, the processing chamber 214 may include one or more matched pairs of ALD processing stations and plasma enhanced ALD processing stations. Although the processing chamber 214 shown includes four stations, it should be understood that a processing chamber according to the present disclosure may have any suitable number of stations. For example, in some embodiments, the processing chamber may have five or more stations; while in other embodiments, the processing chamber may have three or fewer stations.

[0056] It should be understood that various references to RF power settings in the present disclosure are generally intended to refer to RF power settings for each wafer unless otherwise specified. In embodiments involving multiple processing stations in a multi-station processing tool, one or more RF power sources may be provided to serve multiple processing stations (e.g., simultaneously and / or sequentially). In embodiments where a single RF power source serves the processing stations, the per-wafer power setting of the RF power source may be multiplied by the number of processing stations to which the plasma of the desired power level is simultaneously provided. In other words, when the present disclosure describes an RF power setting of 300 watts, it should be understood that the RF power setting reflects a value of 300 watts per wafer, and in a multi-station processing tool, the actual RF power setting of the RF power source may be the power setting per wafer multiplied by the number of stations.

[0057] The multi-station processing tool 200 may include a wafer handling system to transfer wafers within the processing chamber 214. In some embodiments, the wafer handling system may transfer wafers between various processing stations, and / or between a processing station and a load lock. It should be understood that any suitable wafer handling system may be used. Non-limiting examples include a wafer carousel and a wafer handling robot. Figure 2 Also shown is an embodiment of a system controller 250 for controlling processing conditions and hardware states of the multi-station processing tool 200. The system controller 250 may include one or more memory devices 256, one or more mass storage devices 254, and one or more processors 252. The processor 252 may include a CPU or computer, analog and / or digital input / output connections, a stepper motor controller board, etc.

[0058] In some embodiments, the system controller 250 controls all activities of the multi-station processing tool 200. The system controller 250 executes system control software 258, which is stored in the mass storage device 254, loaded into the memory device 256, and executed on the processor 252. Alternatively, the control logic can be hard-coded into the controller 250. Application specific integrated circuits, programmable logic devices (e.g., field programmable gate arrays or FPGAs), etc. can be used for these purposes. In the following discussion, wherever "software" or "code" is used, functionally equivalent hard-coded logic can be used there. The system control software 258 may include a plurality of instructions for controlling: time, gas mixture, gas flow rate, chamber and / or station pressure, chamber and / or station temperature, wafer temperature, target power level, RF power level, substrate holder, chuck and / or susceptor position, and other parameters of the specific process performed by the multi-station processing tool 200. The system control software 258 can be configured in any suitable manner. For example, subroutines or control objects for various process tool components may be written to control the operation of the process tool components required to perform various process tool processes according to the disclosed cleaning methods. The system control software 258 may be encoded in any suitable computer readable programming language.

[0059] In some embodiments, the system control software 258 may include input / output control (IOC) sequence instructions for controlling the various parameters described above. In some embodiments, other computer software and / or programs stored on a mass storage device 254 and / or a memory device 256 associated with the system controller 250 may be used. Examples of programs or portions of programs for this purpose include substrate positioning programs, process gas control programs, pressure control programs, heater control programs, and plasma control programs.

[0060] The substrate positioning program may include program code used by processing tool components to load a substrate onto the substrate holder 218 and to control the spacing between the substrate and other components of the multi-station processing tool 200 .

[0061] The process gas control program may include code for controlling gas composition (e.g., iodine-containing silicon precursors as described herein, and nitrogen-containing gases, carrier gases, and sweep gases) and flow rates, and optionally for flowing gases into one or more process stations prior to deposition to stabilize the pressure within the process station. The pressure control program may include code for controlling the pressure within the process station, for example, by adjusting a throttle valve in an exhaust system of the process station, gas flow into the process station, etc.

[0062] The heater control program may include code for controlling the flow of current to a heating element used to heat the substrate. Alternatively, the heater control program may control the delivery of a heat transfer gas (eg, helium) to the substrate.

[0063] A plasma control program may include code for setting RF power levels applied to process electrodes within one or more process stations according to embodiments herein.

[0064] A pressure control program may include code for maintaining pressure within a reaction chamber according to embodiments herein.

[0065] In some embodiments, there may be a user interface associated with the system controller 250. The user interface may include a display screen, a graphical software display of device and / or processing conditions, and user input devices such as a pointing device, keyboard, touch screen, microphone, etc.

[0066] In some embodiments, the parameters adjusted by the system controller 250 may be related to process conditions. Non-limiting examples include composition and flow rates of process gases, temperature, pressure, plasma conditions (e.g., RF bias power level), etc. These parameters may be provided to the user in the form of a recipe that may be input using a user interface.

[0067] A number of signals for monitoring the process from sensors of the various process tools may be provided through analog and / or digital input connections of the system controller 250. Such signals for controlling the process may be output on analog and digital output connections of the multi-station process tool 200. Non-limiting examples of process tool sensors that may be monitored include mass flow controllers, pressure sensors (e.g., manometers), thermocouples, etc. Appropriately programmed feedback and control algorithms may be used with the data from these sensors to maintain process conditions.

[0068] The controller 250 may provide program instructions for implementing the above-described deposition process. The program instructions may control various process parameters, such as DC power level, RF bias power level, pressure, temperature, etc. The instructions may control these parameters to operate the in-situ deposition of the film stack according to the various embodiments described herein.

[0069] The system controller 250 will typically include one or more memory devices and one or more processors configured to execute instructions so that the apparatus will perform methods consistent with the present embodiments. A machine-readable medium containing instructions for controlling processing operations consistent with the present embodiments may be coupled to the system controller 250.

[0070] In some implementations, the system controller 250 is part of a system, which can be part of the above examples. Such a system can include a semiconductor processing device, which includes one or more processing tools, one or more chambers, one or more platforms for processing, and / or specific processing components (wafer holders, gas flow systems, etc.). These systems can be integrated with electronic devices for controlling their operation before, during, and after the processing of semiconductor wafers or substrates. The electronic device can be referred to as a "controller", which can control various components or subcomponents of one or more systems. Depending on the processing conditions and / or system type, the system controller 250 can be programmed to control any of the processes disclosed herein, including the delivery of process gases, temperature settings (e.g., heating and / or cooling), pressure settings, vacuum settings, power settings, radio frequency (RF) generator settings, RF matching circuit settings, frequency settings, flow rate settings, fluid delivery settings, position and operation settings, wafer transfer in and out of tools connected to or connected through an interface with a specific system and other transfer tools and / or load locks.

[0071] In general, the system controller 250 may be defined as an electronic device having various integrated circuits, logic, memory, and / or software to receive instructions, issue instructions, control operations, enable cleaning operations, enable endpoint measurements, and the like. The integrated circuits may include chips in the form of firmware that store program instructions, digital signal processors (DSPs), chips defined as application specific integrated circuits (ASICs), and / or one or more microprocessors or microcontrollers that execute program instructions (e.g., software). The program instructions may be instructions sent to the system controller 250 in the form of various individual settings (or program files) that define operating parameters for performing a particular process on or for a semiconductor wafer or system. In some embodiments, the operating parameters may be part of a recipe defined by a process engineer to accomplish one or more process steps during the manufacture of one or more layers, materials, metals, oxides, silicon, silicon dioxide, surfaces, circuits, and / or dies of a wafer.

[0072] In some implementations, the system controller 250 may be part of or coupled to a computer that is integrated with the system, coupled to the system, otherwise networked to the system, or a combination thereof. For example, the system controller 250 may be in the "cloud" or may be all or part of a wafer fab host system that may allow remote access to wafer processing. The computer may enable remote access to the system to monitor the current progress of a manufacturing operation, examine the history of past manufacturing operations, examine trends or performance criteria for multiple manufacturing operations, to change parameters of a current process, set processing steps to follow a current process, or start a new process. In some examples, a remote computer (e.g., a server) may provide a processing recipe to the system via a network (which may include a local network or the Internet). The remote computer may include a user interface that enables input or programming of parameters and / or settings, which are then sent from the remote computer to the system. In some examples, the system controller 250 receives instructions in the form of data that specify parameters for each processing step to be performed during one or more operations. It should be understood that the parameters may be specific to the type of process to be performed and the type of tool that the system controller 250 is configured to interface with or control. Thus, as described above, the system controller 250 can be distributed, for example, by including one or more discrete controllers networked together and working toward a common purpose (e.g., processing and control as described herein). An example of a distributed controller for such a purpose is one or more integrated circuits on a chamber that communicate with one or more integrated circuits remotely (e.g., at a platform level or as part of a remote computer), which combine to control processing on the chamber.

[0073] Exemplary systems may include, but are not limited to, plasma etch chambers or modules, deposition chambers or modules, spin rinse chambers or modules, metal plating chambers or modules, cleaning chambers or modules, chamfer edge etch chambers or modules, physical vapor deposition (PVD) chambers or modules, chemical vapor deposition (CVD) chambers or modules, PEALD chambers or modules, atomic layer etch (ALE) chambers or modules, ion implantation chambers or modules, track chambers or modules, and any other semiconductor processing system that may be associated with or used in the manufacture and / or preparation of semiconductor wafers.

[0074] As described above, depending on one or more processing steps to be performed by the tool, the system controller 250 can communicate with one or more other tool circuits or modules, other tool components, cluster tools, other tool interfaces, adjacent tools, neighboring tools, tools located throughout the factory, a host computer, another controller, or tools used in the transport of materials for transporting wafer containers to and from tool locations and / or load ports in a semiconductor manufacturing facility.

[0075] It should be noted that Figure 2Only one example of a multi-station processing chamber that can be used in some embodiments of the technology, systems and methods described herein is shown. In some implementations, the multi-station processing chamber may include multiple chambers or reactors (e.g., two, four, six, eight, etc.), wherein the multiple chambers or reactors are modular in nature and clustered. For example, the multiple chambers or reactors may be clustered around one or more shared components (e.g., one or more wafer handling systems, system controllers, etc.). The multiple chambers or reactors may be in a common vacuum environment. In some embodiments, the vacuum environment and the multiple modules it contains, as well as the shared wafer handling resources are collectively referred to as a "cluster tool".

[0076] In some embodiments, the stations may be operably coupled to components (e.g., one or more manifolds each coupled to a gas source, an RF generator, etc.). In the case where the processing chamber is a multi-station processing chamber, some components may be shared in common. For example, each station may be connected to a common gas source via one or more manifolds. As another example, the RF generator may be shared by all stations. In some embodiments, the stations may be operably coupled to common components via independently controllable components. For example, the stations may be operably coupled to the manifold via a gas valve. More specifically, a first station may be operably coupled to the manifold via a first gas valve, and a second station may be operably coupled to the manifold via a second gas valve, wherein the first gas valve and the second gas valve may be independently controlled and / or actuated. As described above, these independently controllable components may be actuated based on the output of the machine learning model so as to achieve greater wafer-to-wafer consistency between the stations of the multi-station apparatus. Below and in conjunction with Figure 3 What is shown and described is an example of an independently controllable valve, which can be used as a control target based on the output of a machine learning model. In addition, the measurement results associated with the valve can be used as inputs to the machine learning model. For example, valve opening and closing timing information can be used to determine the amount of gas flowing to a particular station. The machine learning model can take the amount of gas flowing to the station as input and generate a predicted wafer characteristic at a given time based on the amount of gas flowing to the station within a time margin (time window) corresponding to the given time.

[0077] In some embodiments, a multi-station processing chamber is associated with one or more manifolds, wherein each manifold may be coupled to a different gas source. Different manifolds may be used in association with different manufacturing processes. For example, a first manifold may be used for gas flow during a deposition process. As another example, a second manifold may be used for gas flow during an etching process and / or during an inhibition process. As used herein, an inhibition process refers to adjusting the growth rate within a feature during, for example, an ALD process. For example, an inhibition process may be used to prevent growth at the top of a feature while growing at the bottom of the feature. As another example, a third manifold may be used during an oxidation step, and a fourth manifold may be used during a reduction step. In a more specific example, a third manifold and / or a fourth manifold may be used during a passivation process. As used herein, a passivation process may be used to change the surface composition of a film or substrate, for example, to prevent etching of the sidewalls of a feature. As another example, a manifold may be used to provide a gas associated with a sweep step (e.g., a non-reactive gas, such as argon), wherein the gas is used to remove volatile byproducts. Such a manifold may be shared between multiple stations in a processing chamber.

[0078] In some embodiments, corresponding valves may have a naming convention, wherein the naming convention indicates that the corresponding valve operably couples a particular manifold to a different station. For example, valve X01 may operably couple station X to a particular manifold. As a more specific example, in some embodiments, a multi-station process chamber including four stations may include valves x101, x201, x301, and x401, wherein valve x101 operably couples station 1 to a manifold, valve x201 operably couples station 2 to a manifold, and so on. It should be understood that a process chamber may include any appropriate number of independently controllable valves (e.g., four, eight, sixteen, twenty, etc.).

[0079] Figure 3Schematic diagrams showing example couplings of various manifolds for a single station of a multi-station process chamber according to some embodiments. As shown, multiple manifolds are coupled to a showerhead 301. For example, manifold 1 is operably coupled to showerhead 301 via valve 302, manifold 2 is operably coupled to showerhead 301 via valve 304, manifold 3 is operably coupled to showerhead 301 via valve 306, and manifold 4 is operably coupled to showerhead 301 via valve 308. As described below, in some embodiments, each of valves 302, 304, 306, and 308 can be separated and independently activated for corresponding valves associated with other stations of the multi-station process chamber. In some embodiments, each manifold can be operably coupled to a different gas source. For example, manifold 1 can be used to provide a first gas to a station via a first gas source, and manifold 2 can be used to provide a second gas to a station via a second gas source. In some embodiments, different manifolds can be used in conjunction with different manufacturing processes. Additionally or alternatively, in some embodiments, multiple manifolds may be manufactured during the execution of a single manufacturing process. Figure 3 Four manifolds are shown in FIG. 1 , but it should be understood that the stations may be operably coupled to any suitable number of manifolds. Figure 3 Only four valves are shown in FIG. 1 , but it should be understood that the process chamber may include any suitable number of valves (eg, four, eight, sixteen, twenty, etc.).

[0080] In some implementations, a trained machine learning model may be used to take as input tool control information measured from a running manufacturing tool (e.g., during the execution of a manufacturing operation (e.g., a deposition operation)). The tool control information may indicate the state of one or more components of the manufacturing tool (e.g., of a multi-station processing chamber). For example, the one or more components may include valves associated with one or more manifolds configured to deliver various gases to the stations. In some embodiments, each valve may be associated with a particular manifold and station. It should be noted that in some embodiments, one or more components of the manufacturing tool may include multiple valves, so that data (e.g., measurements) related to multiple manifolds and / or multiple stations of the processing chamber are used as input to the trained machine learning model. As a more specific example, the measurements may include open and close times (e.g., time points corresponding to the times when the valves are opened and closed). The open and close times may be used to determine the amount of gas delivered to a given station via the corresponding manifold based on the duration of valve opening. It should be noted that in this case, the input to the trained machine learning model may include information derived from valve opening and closing times, such as the amount of gas delivered to the station calculated based on valve timing information. In this case, the amount of gas can be determined based on known gas flow rate and valve timing information. In some implementations, the amount of gas can be determined based on valve timing information, gas flow rate, and other information (e.g., ampoule pressure, partial pressure, or any combination thereof). As another example, the one or more components may include an RF switch that couples RF power from an RF generator to each station. In some embodiments, the measurement may include a measurement from one or more sensors associated with a processing chamber. These sensors may include one or more camera sensors, one or more pressure sensors, one or more temperature sensors, one or more voltage sensors, one or more current sensors, one or more spectral sensors, or any combination thereof.

[0081] In some embodiments, the trained machine learning model can be configured to produce as an output an adjustment to one or more control components of the manufacturing tool. It should be noted that the one or more control components can be the same as (or include one or more of) the one or more components associated with the measurement values ​​used as input to the machine learning model, or the one or more control components can be different from the one or more components. For example, in the case where the one or more components used to provide input to the machine learning model include a first valve associated with a first manifold (e.g., which provides a first gas to a particular station), the one or more control components that can be adjusted can include the first valve, can include a second valve (e.g., it is associated with a second manifold, the second manifold provides a second gas to the station, or provides a second gas to a different station other than the station where the model input measurement has been made), an RF switch, and / or any suitable combination thereof. As a more specific example, in some implementations, the one or more control components can include one or more diverter valves associated with a particular manifold and diverting gas from a given station. Adjustment of the one or more control components may thus produce a change in the manufacturing process in a first station of the multi-station processing chamber relative to a second station of the multi-station processing chamber. For example, adjustment of the one or more control components may cause the RF power of a particular station to change (e.g., increase or decrease), thereby affecting the rate of the deposition process in that station; may cause precursor gas to be diverted away from a particular station, thereby stopping the deposition process in that station; may cause a change (e.g., increase or decrease) in the gas flow rate to a particular station, thereby affecting the rate of the deposition process in that station, etc. It should be noted that in the case where the manufacturing apparatus is a single-station module, adjustment of the one or more control components may allow for granular control of the individual processing steps performed in the station, for example by adjusting the gas flow of a particular species, adaptively changing the number or time of pulses assigned to a particular step, etc., rather than by controlling variability from station to station.

[0082] It should be noted that the trained machine learning model can be configured to directly generate adjustments to one or more control components as output. Additionally or alternatively, the trained machine learning model can be configured to generate predicted wafer characteristics based on measured state information associated with one or more components. Continuing with this example, the one or more control components can be adjusted based on the predicted wafer characteristics. The predicted wafer characteristics can include a predicted wafer thickness at a given time based on measurements associated with the one or more components. For example, the wafer thickness of a wafer in a given station can be predicted based on the amount of gas delivered to the station, which in turn can be determined based on the valve timing of a valve associated with a manifold that delivers the gas to the station. Additionally, it should be noted that the machine learning model can be configured to take into account or take as additional input other information related to the processing chamber, such as chamber pressure, gas pressure, ampoule temperature, non-ampoule gas feed information, carrier gas flow rate, camera data from one or more cameras that obtain visual information from a station in the processing chamber, data from one or more sensors (e.g., temperature sensors, voltage sensors, current sensors, spectral sensors, etc.), and the like.

[0083] The machine learning model may be of any suitable type and / or architecture. Example types of machine learning models that may be used include regression models (e.g., linear regression models, logistic regression models, etc.), neural networks, deep neural networks, convolutional neural network (CNN) regressors, networks using Bayesian optimization, etc., etc. In addition, it should be noted that in some implementations, the machine learning model may be updated periodically, such as as an "online" update of the model. For example, the machine learning model may be updated based on post-processing metrology obtained using a subset of processed wafers. As a more specific example, a training set constructed using known component information (e.g., known valve timing information, known RF power information, etc.) and chip thickness measured using various metrology techniques may be used to retrain the machine learning model. The metrology techniques may include one or more of: the quality of a thin film grown for at least one wafer; a Fourier transform infrared (FTIR) spectral peak; the thickness of a film deposited on the at least one wafer; a refractive index; stress at a local location on the at least one wafer; stress associated with wafer bow of the at least one wafer; particle information; and / or material dielectric constant information. It should be noted that in some implementations, the machine learning model can be trained to predict any of these metrology measurements. Such predictions (generally referred to herein as predicted wafer characteristics) can be considered "virtual metrology" measurements.

[0084] Figure 44 is a schematic diagram of a system 400 for automatically controlling process chamber components according to some implementations. As shown, the system 400 includes a manufacturing tool 402, a trained machine learning model 404, and a manufacturing tool adjustment component 406. As described above, the manufacturing tool 402 can be a multi-station process chamber configured to perform various deposition and / or etching processes. Figure 4 As shown, the trained machine learning model 404 can be configured to take the measured tool control information as input, where such information can include valve timing information, RF power information, RF switch information, etc. In some embodiments, the trained machine learning model 404 can generate as output a predicted wafer characteristic at the current time (e.g., based on the tool control information measured at the current time). The predicted wafer characteristic can include a predicted wafer thickness of wafers in each station of a processing chamber (e.g., manufacturing tool 402). In such an implementation, the output of the trained machine learning model 404 can be provided to a manufacturing tool adjustment component 406, which can be configured to determine adjustments to one or more control components of the manufacturing tool 402 to produce consistency between multiple wafers processed in the manufacturing tool 402. For example, as described above, the adjustment can include: actuating one or more diverter valves for diverting gas from one or more stations; modifying the gas flow rate of gas flowing to one or more stations; modifying the RF power provided to one or more stations, etc. Alternatively, in some embodiments, the machine learning model can directly determine one or more adjustments. In such an embodiment, the manufacturing tool adjustment component 406 may be omitted.

[0085] In addition, if Figure 4 As shown, the post-processing metrology system 408 can be configured to determine metrology information for a subset of wafers processed by the manufacturing tool 402. The metrology results can then be used to update the trained machine learning model 404, for example, by retraining the trained machine learning model 404. Thus, the trained machine learning model 404 can be periodically updated using post-processing metrology data.

[0086] Figure 5is a flow chart of an exemplary process for automatically controlling components of a processing chamber according to some embodiments. In some implementations, the blocks of process 500 may be performed by a processor or controller (e.g., a controller associated with a processing chamber). In some embodiments, process 500 may be performed by an independent edge node controller configured to communicate with a tool controller or processor. It should be understood that although some of the embodiments described below in conjunction with process 500 are related to a multi-station processing chamber, the blocks of process 500 may be performed in conjunction with a single-station module. In some embodiments, two or more blocks of process 500 may be performed substantially in parallel. In some implementations, one or more blocks of process 500 may be omitted. In some implementations, the blocks of process 500 may be performed in accordance with Figure 5 The order shown may be executed in a different order.

[0087] Process 500 may begin at 502 by obtaining information related to the status of a component of a manufacturing tool currently performing a manufacturing process on one or more wafers. As described above, the information may include valve timing information. For example, the valve timing information may include valve opening and closing times of one or more valves of a multi-station process chamber. As a more specific example, the one or more valves may be valves associated with one or more manifolds, wherein each manifold delivers a particular gas to a station of the multi-station process chamber (e.g., the above combined valve timing information). Figure 3 As a specific example, the one or more valves may be valves associated with a manifold configured to deliver precursor gases to the various stations. It should be understood that in some embodiments, multiple sets of valves may be considered, with each set of valves being associated with a manifold configured to deliver a particular type of gas (e.g., precursor gas, gas for an oxidation step, gas for an inhibition step, gas for a passivation step, etc.). Where the information includes valve timing information, the valve timing information may be acquired using any suitable rate, such as 10 Hz, 50 Hz, 100 Hz, 200 Hz, 800 Hz, 1000 Hz, 1500 Hz, 2000 Hz, etc. In some embodiments, the information about the component status may additionally or alternatively include pressure information (e.g., chamber pressure information, relative ampoule pressure information, gas pressure information, etc.), RF power information (e.g., RF switch timing information, RF power delivered to the various stations, etc.), temperature information, or any combination thereof.

[0088] At 504, process 500 may provide the obtained information to a trained machine learning model. The obtained information or a system of information derived from the obtained information may be provided as an input to the trained machine learning model. For example, where the obtained information includes valve timing information, the information provided to the machine learning model may include the amount of a particular gas provided to a given station over a duration corresponding to the time at which the information was obtained at box 502. In some implementations, the machine learning model may generate as an output a predicted wafer thickness of a wafer in a given station. For example, for a processing chamber comprising four stations, each station having a wafer undergoing a deposition process, the machine learning model may output a predicted wafer thickness of each wafer. In some embodiments, the machine learning model may generate as an output a numerical value of one or more control components and / or a change to the current setting of the one or more control components, thereby generating consistency in wafer thickness between multiple wafers in multiple stations of the processing chamber. For example, the output may indicate that air flow to a particular station is to be reduced and / or diverted because the predicted thickness of wafers in that station exceeds the predicted thickness of wafers in other stations; and indicate that RF power to a particular station is to be reduced because the predicted thickness of wafers in that station exceeds the predicted thickness of wafers in other stations; or the like.

[0089] At 506, the process 500 can determine whether to adjust the control components of the manufacturing tool. For example, in some embodiments, the process 500 can determine that the adjustment is to be made in response to the output of the machine learning model (e.g., the output generated at block 504), where the output indicates that the difference in wafer thickness at the current time between two wafers in two stations is predicted to exceed a predetermined threshold. As another example, in some embodiments, the process 500 can determine whether to adjust the control components of the manufacturing tool directly based on the output of the machine learning model, where the machine learning model directly generates parameters used by the control components that can provide consistency across wafers.

[0090] If at 506, the process 500 determines that no adjustment is to be made to the control components ("No" at 506), the process 500 may proceed to 510. Conversely, if at 506, the process 500 determines that an adjustment is to be made to the control components ("Yes" at 506), the process 500 may proceed to block 508 and instructions may be transmitted to a controller of the manufacturing tool to adjust one or more control components of the manufacturing tool. As described above, the adjustment may result in: diverting gas from a particular station (e.g., where the manufacturing tool includes a common mass flow controller for all stations); modifying (e.g., decreasing or increasing) the gas flow rate where individual mass flow controllers are used; modifying the RF power delivered to a particular station; or any combination thereof.

[0091] At 510, process 500 can determine whether the manufacturing process has been completed. For example, for a deposition process, process 500 can determine whether a maximum number of deposition cycles (e.g., as set in a recipe being implemented) has been performed, whether the predicted wafer thickness for all wafers being manufactured has reached a target thickness, etc. It should be noted that in the event that process 500 determines that a specified maximum number of deposition cycles has been performed, but the predicted wafer thickness has not reached the target thickness, process 500 can present or send an alert that flags an error condition associated with the manufacturing tool.

[0092] If at 510, process 500 determines that the manufacturing process has been completed ("yes" at 510), process 500 can end. Conversely, if at 510, process 500 determines that the manufacturing process has not been completed ("no" at 510), process 500 can loop back to 502 and updated information at the next time step can be obtained. Process 500 can loop through blocks 502 to 510 until the manufacturing process has been completed.

[0093] In some embodiments, the techniques described herein can be used to determine the degradation of at least one component of a multi-station processing chamber. Then, the determined degradation can be used to perform predictive maintenance, so that the at least one component can be replaced before failure, but other components that have not been degraded are not replaced. For example, using conventional techniques, it may not be possible to determine which valve in a set of valves used in a tool has failed. Therefore, traditionally, all valves can be replaced according to a periodic and / or predetermined schedule (for example, after a set period of time, after running a set number of processes, etc.), regardless of whether a given valve is degraded. However, using the techniques described herein, a specific valve can be identified as experiencing more than a predetermined amount of degradation and / or the specific valve may be about to fail, and the identified valve can be replaced while retaining other valves. This can allow for significant cost savings because only the components identified as being degraded are replaced, rather than replacing all components regardless of the current state. In some implementations, components can be identified as experiencing degradation based on the output of the machine learning model. For example, in some embodiments, a specific station of a multi-station processing chamber can be identified as having a component that has experienced degradation based on the output of the machine learning model.

[0094] Figure 7Example timing diagrams for control chamber components according to some embodiments are shown. Referring to panel 702, timing diagrams for supply valves and diverter valves at different stations are shown. Specifically, curve 704 shows the timing of the supply valve at the first station, while curve 706 shows the timing of the diverter valve at the first station. Note that curves 704 and 706 are reversed relative to each other because when the supply valve is open, the diverter valve is closed, and vice versa. Similarly, curve 708 shows the timing of the supply valve at the second station, while curve 710 shows the timing of the diverter valve at the second station. Panel 707 represents a single cycle. It should be noted that during the cycle highlighted in panel 707, the supply valves of both stations are open, while the diverter valve is closed. Referring to panel 711, curve 704 shows that the supply valve at the first station is closed (e.g., because the target deposition thickness has been reached), while curve 706 shows that the diverter valve at the first station is open. However, during the same time period, curves 708 and 710 show that even when a deposition cycle in the first station has been completed, the supply valve of the second station and the diverter valve of the second station continue to cycle between open and closed, thereby allowing additional deposition cycles to be performed.

[0095] The timing diagram shown in panel 720 illustrates a similar concept. Curves 722 and 724 show timing diagrams of the RF switch used to switch the RF power on and off for the first station and the second station, respectively. As shown, during the first time period 725, the RF power is switched on and off for both the first station and the second station within a set of cycles. During the second time period 727, the RF switch of the first station remains closed (as shown by curve 722), for example, because the wafer being processed in the first station reaches the target thickness. However, during the second time period 727, as the deposition cycle continues in the second station, the RF power for the second station continues to cycle on and off (as shown by curve 724).

[0096] It should be noted that although the above Figure 7 The timing diagram shown is described as corresponding to the first station and the second station, but in some embodiments, multiple stations can follow the same timing. For example, referring to curve 722, stations 1, 2 and 3 can follow the timing shown in curve 722, while station 4 can follow the timing shown in curve 724.

[0097] As mentioned above, for example, combined Figure 1 , Figure 5 and Figure 7In some embodiments, for one or more stations that have reached the target deposition thickness, the precursor supply valve can be closed and the diverter valve can be opened, while other stations continue the deposition cycle. In some embodiments, the diverted precursor gas (and / or any other gas) can be recovered, filtered and recycled. By recycling and reusing the diverted precursor gas (or any other suitable gas), significant cost savings can be achieved. In some embodiments, the diverted precursor gas can be diverted to the precursor recovery tank by opening the recovery valve, which allows the diverted precursor gas to flow from the diverter line to the precursor recovery tank. The recovery valve can be automatically opened (e.g., without user input) in response to determining that one or more stations have completed the processing step. The diverted precursor gas can be filtered to separate the carrier gas from the precursor gas, for example, by a filter.

[0098] Figure 8 An exemplary schematic diagram of a system that can be used for precursor gas recovery is shown. Figure 8 In the system shown, when processing is completed in one or more stations, the precursor gas can be diverted via the diverting line. In response to determining that processing has been completed in one or more stations and processing is being performed in at least one other station, valve R1 can be opened to allow the precursor gas to flow from the diverting line to the precursor recovery tank. After processing in all stations is completed, valve R3 can be closed and recovery valve R4 can be opened to allow the diverted precursor gas to flow to the membrane filter. The filter can be configured to filter the carrier gas to the foreline and recycle the precursor gas to the ampoule. When valve A1 is closed and valve A2 is opened to bypass the ampoule, valve R2 can be opened and valve R1 can be closed to prevent the carrier gas without precursor from entering the recovery tank. Background of the Disclosed Computing Implementations

[0099] A system including a manufacturing tool as described herein may include logic for automated control of components.

[0100] The analysis logic can be designed and implemented in any of a variety of ways. For example, the logic can be implemented in hardware and / or software. Examples are given in the controller section of this article. Hardware-implemented control logic can be provided in any of a variety of forms, including hard-coded logic in a digital signal processor, an application-specific integrated circuit, and other devices with algorithms implemented as hardware. The analysis logic can also be implemented as software or firmware instructions configured to execute on a general-purpose processor. The system control software can be provided by "programming" in a computer-readable programming language.

[0101] The computer program code for controlling the processes in the process series can be written in any conventional computer readable programming language: for example, assembly language, C, C++, Pascal, Fortran or other languages. The compiled object code or script is executed by the processor to perform the tasks identified in the program. As also indicated, the program code can be hard-coded.

[0102] Integrated circuits used in the logic may include chips in the form of firmware storing program instructions, digital signal processors (DSPs), chips defined as application specific integrated circuits (ASICs), and / or one or more microprocessors, or microcontrollers that execute program instructions (e.g., software). Program instructions may be instructions delivered in the form of various individual settings (or program files) defining operating parameters for performing a particular analysis or image analysis application.

[0103] In some implementations, the image analysis logic resides (and executes) on a computing resource on or closely associated with the manufacturing tool from which the camera image is captured. In some implementations, the image analysis logic is remote from the manufacturing tool from which the camera image is captured. For example, the analysis logic may be executed on a cloud-based resource.

[0104] Figure 6 6 is a block diagram of an example of a computing device 600 suitable for implementing some embodiments of the present disclosure. For example, device 600 may be suitable for implementing some or all of the functionality of the image analysis logic disclosed herein.

[0105] The computing device 600 may include a bus 602 that directly or indirectly couples the following devices: a memory 604, one or more central processing units (CPUs) 606, one or more graphics processing units (GPUs) 608, a communication interface 610, an input / output (I / O) port 612, an input / output component 614, a power supply 616, and one or more presentation components 618 (e.g., a display). In addition to the CPU 606 and the GPU 608, the computing device 600 may also include Figure 6 Additional logic devices not shown include, but are not limited to, an image signal processor (ISP), a digital signal processor (DSP), an ASIC, an FPGA, etc.

[0106] although Figure 6The various blocks of are shown as being connected by wires via bus 602, but this is not intended to be limiting and is for clarity only. For example, in some embodiments, presentation components 618 such as a display device may be considered I / O components 614 (e.g., if the display is a touch screen). As another example, CPU 606 and / or GPU 608 may include memory (e.g., memory 604 may represent a storage device in addition to the memory of GPU 608, CPU 606, and / or other components). In other words, Figure 6 The computing devices described in the present disclosure are illustrative only. No distinction is made between categories such as "workstations," "servers," "laptops," "desktops," "tablets," "client devices," "mobile devices," "handheld devices," "electronic control units (ECUs)," "virtual reality systems," and / or other device or system types, as all of these are contemplated under Figure 6 within the range of computing devices.

[0107] The bus 602 may represent one or more buses, such as an address bus, a data bus, a control bus, or a combination thereof. The bus 602 may include one or more bus types, such as an Industry Standard Architecture (ISA) bus, an Extended Industry Standard Architecture (EISA) bus, a Video Electronics Standards Association (VESA) bus, a Peripheral Component Interconnect (PCI) bus, a Peripheral Component Interconnect Express (PCIe) bus, and / or other types of buses.

[0108] Memory 604 may include any of a variety of computer-readable media. Computer-readable media may be any available media that can be accessed by computing device 600. Computer-readable media may include volatile and non-volatile media, and removable and non-removable media. By way of example and not limitation, computer-readable media may include computer storage media and / or communication media.

[0109] Computer storage media may include volatile and nonvolatile media and / or removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules, and / or other data types. For example, memory 604 may store computer readable instructions (e.g., representing programs and / or program elements, such as an operating system). Computer storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by computing device 600. As used herein, computer storage media itself does not include signals.

[0110] Communication media may embody computer-readable instructions, data structures, program modules, and / or other data types in a modulated data signal such as a carrier wave or other transport mechanism, and include any information delivery media. The term "modulated data signal" may refer to a signal having one or more of its characteristics that are set or changed in such a way as to encode information in the signal. By way of example and not limitation, communication media may include wired media such as a wired network or a direct wire connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above items should also be included within the scope of computer-readable media.

[0111] The CPU 606 may be configured to execute computer-readable instructions to control one or more components of the computing device 600 to perform one or more of the methods and / or processes described herein. The CPUs 606 may each include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) capable of processing multiple software threads simultaneously. The CPU 606 may include any type of processor and may include different types of processors, depending on the type of computing device 600 implemented (e.g., a processor with fewer cores for a mobile device and a processor with more cores for a server). For example, depending on the type of computing device 600, the processor may be an ARM processor implemented using reduced instruction set computing (RISC) or an x86 processor implemented using complex instruction set computing (CISC). In addition to one or more microprocessors or auxiliary coprocessors (e.g., math coprocessors), the computing device 600 may also include one or more CPUs 1306.

[0112] GPU 608 can be used by computing device 600 to render graphics (e.g., 3D graphics). GPU 608 may include many (e.g., tens, hundreds, or thousands) cores capable of processing many software threads simultaneously. GPU 608 may generate pixel data of an output image in response to a rendering command (e.g., a rendering command from CPU 606 received via a host interface). GPU 608 may include a graphics memory, such as a display memory, for storing pixel data. Display memory may be included as part of memory 604. GPU 608 may include two or more GPUs operating in parallel (e.g., via a link). When combined, each GPU 608 may generate pixel data of different parts of an output image or different output images (e.g., a first GPU for a first image, a second GPU for a second image). Each GPU may contain its own memory or may share memory with other GPUs.

[0113] In examples where computing device 600 does not include GPU 608 , CPU 606 may be used to render graphics.

[0114] The communication interface 610 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 600 to communicate with other computing devices via an electronic communication network (including wired and / or wireless communications). The communication interface 610 may include components and functionality that enable communication over any of a variety of different networks, such as wireless networks (e.g., Wi-Fi, Z-Wave, Bluetooth, Bluetooth LE, ZigBee, etc.), wired networks (e.g., communicating over Ethernet), low-power wide area networks (e.g., LoRaWAN, SigFox, etc.), and / or the Internet.

[0115] I / O ports 612 can enable computing device 600 to be logically coupled to other devices, including coupling to I / O components 614, presentation components 618, and / or other components, some of which can be built into (e.g., integrated into) computing device 600. Illustrative I / O components 614 include microphones, mice, keyboards, joysticks, tracking pads, satellite antennas, scanners, printers, wireless devices, etc. I / O components 614 can provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by users. In some cases, the input can be transmitted to appropriate network elements for further processing. NUI can implement any combination of speech recognition, stylus recognition, facial recognition, biometric recognition, gesture recognition on and near the screen, air gestures, head and eye tracking, and touch recognition (as described in more detail below) related to the display of computing device 600. Computing device 600 may include a depth camera, such as a stereo camera system, an infrared camera system, an RGB camera system, a touch screen technology, and a combination of these, to perform gesture detection and recognition. Additionally, computing device 600 may include an accelerometer or gyroscope that enables detection of motion (e.g., as part of an inertial measurement unit (IMU)). In some examples, computing device 600 may use the output of the accelerometer or gyroscope to render an immersive augmented reality or virtual reality.

[0116] The power supply 616 may include a hardwired power supply, a battery power supply, or a combination thereof. The power supply 616 may provide power to the computing device 600 to enable the components of the computing device 600 to operate.

[0117] The presentation component 618 may include a display (e.g., a monitor, a touch screen, a television screen, a head-up display (HUD), other display types, or combinations thereof), speakers, and / or other presentation components. The presentation component 618 may receive data from other components (e.g., GPU 608, CPU 606, etc.) and output the data (e.g., as images, video, sound, etc.).

[0118] The present disclosure may be described in the general context of computer code or machine-usable instructions, including computer-executable instructions (e.g., program modules) executed by a computer or other machine (e.g., a personal data assistant or other handheld device). In general, program modules, including routines, programs, objects, components, data structures, etc., refer to code that performs specific tasks or implements specific abstract data types. The present disclosure may be practiced in a variety of system configurations, including in handheld devices, consumer electronics, general-purpose computers, more specialized computing devices, etc. The present disclosure may also be practiced in distributed computing environments, where tasks are performed by remote processing devices that are linked through a communications network. Other Considerations

[0119] Exemplary systems may include, but are not limited to, plasma etch chambers or modules, plasma assisted deposition chambers or modules, such as plasma assisted chemical vapor deposition (PECVD) chambers or modules or plasma assisted atomic layer deposition (PEALD) chambers or modules, atomic layer etch (ALE) chambers or modules, clean rooms or modules, physical vapor deposition (PVD) chambers or modules, ion implantation chambers or modules, and any other plasma assisted semiconductor processing system that may be associated with or used in the manufacture and / or fabrication of semiconductor wafers.

[0120] Unless otherwise specified, the plasma power levels and related parameters provided herein are suitable for processing 300 mm wafer substrates. It will be appreciated by those skilled in the art that these parameters may be adjusted as needed for substrates of other sizes.

[0121] The apparatus / processes described herein can be used in conjunction with photolithographic patterning tools or processes, e.g., for preparing or manufacturing electronic devices including semiconductor devices, displays, LEDs, photovoltaic panels, etc. Typically, though not necessarily, these tools / processes will be used or operated together in a common manufacturing facility. Photolithographic patterning of films typically includes some or all of the following operations, each of which enables multiple available tools: (1) coating a workpiece, i.e., substrate, with a photoresist using a spin coating or spray coating tool; (2) curing the photoresist using a hot plate or oven or UV curing tool; (3) exposing the photoresist to visible light or UV light or X-rays using a tool such as a wafer stepper; (4) developing the resist to selectively remove the resist and thereby pattern it using a tool such as a wet cleaning station; (5) transferring the resist pattern to an underlying film or workpiece using a dry or plasma assisted etching tool; and (6) removing the resist using a tool such as a radio frequency or microwave plasma resist stripper.

[0122] As used in this specification and the appended claims, the singular forms "a", "an", and "the" include plural referents unless the context and the like dictate otherwise. For example, reference to "a unit" includes a combination of two or more such units. Unless otherwise indicated, the "or" conjunction is used in its proper sense as a Boolean logical operator, encompassing both selection of feature alternatives (A or B, where the selection of A is mutually exclusive with B) and selection of feature combinations (A or B, where both A and B are selected).

[0123] It should be understood that the phrases "for each <item> in one or more <items>," "each <item> in one or more <items>," and the like, if used herein, include both single-item groups and multi-item groups, i.e., the phrase "for...each" is used in the sense that it is used in a programming language to refer to each item in any group of items referenced. For example, if the group of items referenced is a single item, then "each" will refer only to that single item (although dictionary definitions of "each" often define the term to mean "each of two or more things"), and does not mean that there must be at least two of those items. Similarly, the terms "set" or "subset" by themselves should not be taken to necessarily cover multiple items - it should be understood that a set or subset may cover only one member or multiple members (unless the context dictates otherwise).

[0124] The use of ordinal indicators such as (a), (b), (c) ... etc., if any, in the present disclosure and claims should be understood not to convey any particular order or sequence, except where such order or sequence is explicitly indicated. For example, if there are three steps labeled (i), (ii), and (iii), it should be understood that unless otherwise indicated, these steps may be performed in any order (or even simultaneously, if not otherwise contraindicated). For example, if step (ii) involves processing of an element created in step (i), then step (ii) may be considered to occur at some point after step (i). Similarly, if step (i) involves processing of an element created in step (ii), the opposite should be understood. It should also be understood that the use of the ordinal indicator "first" herein, such as "the first item," should not be understood to implicitly or inherently imply that there must be a "second" instance, such as "the second item."

[0125] It may be claimed that various computing components including processors, memories, instructions, routines, modules, or other components may be "configured to" perform a task or tasks. In such contexts, the phrase "configured to" is used to refer to a structure by a component including a structure (e.g., stored instructions, circuitry, etc.) that performs a task or tasks during operation. Thus, even when a particular component is not necessarily currently operational (e.g., not in an on state), it may be said that a unit / circuit / component is configured to perform a task.

[0126] The components using the term "configured to" may relate to hardware - for example, circuits, memories storing program instructions executable to perform an operation. In addition, "configured to" may refer to a general structure (such as a general circuit) that is manipulated by software and / or firmware (such as an FPGA or a general processor executing software) so as to be able to operate in a manner to perform the task(s). In addition, "configured to" may refer to one or more memories or memory elements storing computer-executable instructions for performing the task(s). Such memory elements may include memory on a computer chip with processing logic. In some contexts, "configured to" may also include adapting a manufacturing process (such as a semiconductor manufacturing facility) to manufacture equipment (such as an integrated circuit) that is suitable for implementing or performing one or more tasks.

[0127] Although the above embodiments have been described in some detail for the purpose of clarity of understanding, it is apparent that certain variations and modifications may be implemented within the scope of the appended claims. It should be noted that there are many alternative ways of implementing the processes, systems, and devices of the embodiments of the present invention. Therefore, the embodiments of the present invention should be considered illustrative rather than restrictive, and the embodiments are not limited to the details given herein.

Claims

1. A method for deposition control in a process chamber, the method comprising: (a) obtaining information at a current point in time, the information indicating a state of one or more components of the process chamber during a deposition process performed on one or more wafers, wherein the deposition process includes a plurality of deposition cycles performed in the process chamber; (b) determining whether to adjust one or more control components of the process chamber by providing an input based on the information obtained to a trained machine learning model configured to determine an adjustment as an output, wherein the adjustment to the one or more control components results in a change in the deposition process; (c) in response to determining that the one or more control components are to be adjusted, transmitting instructions to a controller of the process chamber to cause the adjustment to be made to the one or more control components; as well as (d) Repeat (a) to (c) until the deposition process is completed.

2. The method according to claim 1, wherein: The processing chamber is a multi-station processing chamber.

3. The method according to claim 2, wherein: The adjustment of the one or more control components causes a change in the deposition process that occurs in a first station of the multi-station processing chamber relative to a second station of the multi-station processing chamber.

4. The method according to claim 2, wherein: In (d), (a) to (c) are repeated until the deposition is completed in each station of the multi-station process chamber.

5. The method according to claim 1, wherein: The processing chamber is a single-station processing chamber.

6. The method according to claim 1, wherein: The machine learning model is configured to determine the adjustment based at least in part on a determination of a plurality of predicted wafer characteristics of the one or more wafers undergoing the deposition process at the current point in time.

7. The method according to claim 6, wherein: The predicted characteristics of the one or more wafers include a deposition thickness for each of the one or more wafers.

8. The method according to any one of claims 1 to 7, wherein: Determining whether to make adjustments to the one or more control components of the processing chamber includes comparing the predicted characteristic for a given wafer to a target characteristic for the given wafer.

9. The method according to claim 8, wherein: The predicted characteristics include virtual metrology measurements.

10. The method according to any one of claims 1 to 7, wherein: The one or more components of the process chamber include one or more valves associated with one or more manifolds of the process chamber, each manifold configured to flow a gas to a station of the process chamber.

11. The method according to claim 10, wherein: The information indicative of the status includes a duration that each of the one or more valves has been open.

12. The method according to claim 11, wherein: The input based on the obtained information provided to the trained machine learning model includes an amount of gas provided to a given station operably coupled to a manifold of the one or more manifolds based on the duration that a corresponding valve has been open.

13. The method of claim 12, further comprising determining the amount of gas based on the duration that the corresponding valve has been open and a gas flow rate.

14. The method according to any one of claims 1 to 7, wherein: The trained machine learning model is configured to take as input at least one of: chamber pressure information; gas pressure information; ampoule temperature; non-ampoule gas feed; or carrier gas flow rate.

15. The method according to any one of claims 1 to 7, wherein: The information obtained in (a) is performed at a sampling rate greater than about 100 Hz.

16. The method according to any one of claims 1 to 7, wherein: The one or more control components include one or more diverter valves that divert gas flowing through a manifold from a station of the process chamber.

17. The method of claim 16, further comprising causing the diverted precursor gas to flow to a precursor recovery tank by causing a recycle valve to be opened in response to determining that the one or more diverter valves are open.

18. The method according to claim 17, further comprising: causing the diverted precursor gas to be filtered; as well as The filtered diverted precursor gas is reused for use in subsequent processing steps performed in the process chamber.

19. The method according to any one of claims 1 to 7, wherein: The processing chamber is a multi-station processing chamber, and wherein the adjustment of the one or more control components includes reducing the flow rate of the gas to the first station relative to the flow rate of the gas to the second station.

20. The method according to any one of claims 1 to 7, wherein: The processing chamber is a multi-station processing chamber including at least a first station and a second station, and wherein the adjustment of the one or more control components includes changing a radio frequency (RF) power used to generate a plasma associated with the deposition process for the first station relative to the second station.

21. The method according to any one of claims 1 to 7, further comprising: (e) obtaining post-process metrology data for at least one wafer of the one or more wafers; as well as (f) using the processed metrology data to update the trained machine learning model.

22. The method according to claim 21, wherein: The post-processing metrology data includes at least one of: resistivity data; quality of thin films grown for the at least one wafer; Fourier transform infrared (FTIR) spectrum peaks; thickness of films deposited on the at least one wafer; refractive index; stress at a local location on the at least one wafer; stress associated with wafer bow of the at least one wafer; particle information; or material dielectric constant information.

23. The method according to any one of claims 1 to 7, further comprising: (e) determining a degradation state of at least one of the one or more components of the process chamber based on the predicted characteristic.

24. A computer program product comprising a non-transitory computer readable medium on which computer executable instructions are provided, the computer executable instructions for causing a computing system to perform a deposition control method for a processing chamber, wherein the instructions include instructions for: (a) obtaining information at a current point in time, the information indicating a state of one or more components of the process chamber during a deposition process performed on one or more wafers, wherein the deposition process includes a plurality of deposition cycles performed in the process chamber; (b) determining whether to adjust one or more control components of the process chamber by providing an input based on the information obtained to a trained machine learning model configured to determine an adjustment as an output, wherein the adjustment to the one or more control components results in a change in the deposition process; (c) in response to determining that the one or more control components are to be adjusted, transmitting instructions to a controller of the processing chamber to cause the adjustment to be made to the one or more control components; as well as (d) Repeat (a) to (c) until the deposition process is completed.

25. The computer program product of claim 24, wherein: The processing chamber is a multi-station processing chamber.

26. The computer program product of claim 25, wherein: The adjustment of the one or more control components causes a change in the deposition process that occurs in a first station of the multi-station processing chamber relative to a second station of the multi-station processing chamber.

27. The computer program product of claim 25, wherein: In (d), (a) to (c) are repeated until the deposition process has been completed in each station of the multi-station processing chamber.

28. The computer program product of claim 24, wherein: The processing chamber is a single-station processing chamber.

29. The computer program product of claim 24, wherein: The machine learning model is configured to determine the adjustment based at least in part on a determination of a plurality of predicted wafer characteristics of the one or more wafers undergoing the deposition process at the current point in time.

30. The computer program product of claim 29, wherein: The predicted characteristics of the one or more wafers include a deposition thickness for each of the one or more wafers.

31. A computer program product according to any one of claims 24 to 30, wherein: Determining whether to make adjustments to the one or more control components of the processing chamber includes comparing the predicted characteristic for a given wafer to a target characteristic for the given wafer.

32. The computer program product of claim 31, wherein: The predicted characteristics include virtual metrology measurements.

33. A computer program product according to any one of claims 24 to 30, wherein: The one or more components of the process chamber include one or more valves associated with one or more manifolds of the process chamber, each manifold configured to flow a gas to a station of the process chamber.

34. The computer program product of claim 33, wherein: The information indicative of the status includes a duration that each of the one or more valves has been open.

35. The computer program product of claim 34, wherein: The input based on the obtained information provided to the trained machine learning model includes an amount of gas provided to a given station operably coupled to a manifold of the one or more manifolds based on the duration that a corresponding valve has been open.

36. The computer program product of claim 35, wherein: The instructions also include instructions for determining the amount of gas based on the duration that the corresponding valve has been open and a gas flow rate.

37. A computer program product according to any one of claims 24 to 30, wherein: The trained machine learning model is configured to take as input at least one of: chamber pressure information; gas pressure information; ampoule temperature; non-ampoule gas feed; or carrier gas flow rate.

38. A computer program product according to any one of claims 24 to 30, wherein: The information obtained in (a) is performed at a sampling rate greater than about 100 Hz.

39. A computer program product according to any one of claims 24 to 30, wherein: The one or more control components include one or more diverter valves that divert gas flowing through a manifold from a station of the process chamber.

40. A computer program product according to any one of claims 24 to 30, wherein: The processing chamber is a multi-station processing chamber, and wherein the adjustment of the one or more control components includes reducing the flow rate of the gas to the first station relative to the flow rate of the gas to the second station.

41. A computer program product according to any one of claims 24 to 30, wherein: The processing chamber is a multi-station processing chamber including at least a first station and a second station, and wherein the adjustment of the one or more control components includes changing a radio frequency (RF) power used to generate a plasma associated with the deposition process for the first station relative to the second station.

42. The computer program product of any one of claims 24 to 30, wherein the instructions further comprise instructions for: (e) obtaining post-process metrology data for at least one wafer of the one or more wafers; and (f) using the processed metrology data to update the trained machine learning model.

43. The computer program product of claim 42, wherein: The post-processing metrology data includes at least one of: resistivity data; quality of thin films grown for the at least one wafer; Fourier transform infrared (FTIR) spectrum peaks; thickness of films deposited on the at least one wafer; refractive index; stress at a local location on the at least one wafer; stress associated with wafer bow of the at least one wafer; particle information; or material dielectric constant information.

44. The computer program product of any one of claims 24 to 30, wherein the instructions further comprise instructions for: (e) determining a degradation state of at least one of the one or more components of the process chamber based on the predicted characteristic.

Images