Residual thickness compensation

Abstract

A method includes identifying a material associated with a substrate processing operation of a recipe. The method further includes determining a predicted total residual thickness value after the substrate processing operation. The method further includes determining a predicted material thickness value for the material associated with the substrate processing operation based on a predicted total residual thickness value. The method further includes updating the recipe based on the material and the predicted material thickness value of the material to generate an updated recipe. The method further includes treating a substrate based on the updated recipe.

Description

Technical Field

[0001] The present disclosure relates to compensation, and more particularly, to residual thickness compensation during substrate manufacturing. Background Art

[0002] Products can be produced by performing one or more manufacturing processes using manufacturing equipment. For example, substrate processing equipment can be used to produce substrates through substrate processing operations. Summary of the Invention

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

[0004] One aspect of the present disclosure includes a method that includes identifying a material associated with a substrate processing operation of a recipe. The method further includes determining an expected total residual thickness value after the substrate processing operation. The method further includes determining an expected material thickness value for the material associated with the substrate processing operation based on the expected total residual thickness value. The method further includes updating the recipe based on the material and the expected material thickness value of the material to generate an updated recipe. The method further includes processing a substrate based on the updated recipe.

[0005] Another aspect of the present disclosure includes a non-transitory computer-readable storage medium comprising instructions that, when executed by a processing device operatively coupled to a memory, perform operations. The operations include identifying a material associated with a substrate processing operation of a recipe. The operations further include determining an expected total residual thickness value after the substrate processing operation. The operations further include determining an expected material thickness value for the material associated with the substrate processing operation based on the expected total residual thickness value. The operations further include updating the recipe based on the material and the expected material thickness value of the material to generate an updated recipe. The operations further include processing a substrate based on the updated recipe.

[0006] Another aspect of the present disclosure includes a system comprising a memory and a processing device coupled to the memory. The processing device identifies a material associated with a substrate processing operation of a recipe. The processing device further determines an expected total residual thickness value after the substrate processing operation. The processing device further determines an expected material thickness value for the material associated with the substrate processing operation based on the expected total residual thickness value. The processing device further updates the recipe based on the material and the expected material thickness value of the material to generate an updated recipe. The processing device further causes the substrate to be processed based on the updated recipe. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0008] Figure 1 is a block diagram illustrating an example system architecture in accordance with certain embodiments.

[0009] Figure 2 A dataset generator associated with residual thickness compensation according to certain embodiments is illustrated.

[0010] Figure 3 is a block diagram illustrating determining prediction data associated with residual thickness compensation according to certain embodiments.

[0011] Figures 4A to 4C is a flow chart of a method associated with residual thickness compensation according to certain embodiments.

[0012] Figure 5 is a block diagram of an example computer system in accordance with certain embodiments. DETAILED DESCRIPTION

[0013] Described herein are techniques for residual thickness compensation (e.g., adjusting film deposition parameters based on residual thickness during substrate fabrication, a scalable runtime software approach to compensating for deposition thickness drift based on chamber residual thickness, and cycle time adjustment based on chamber residual).

[0014] Depending on the process recipe, the manufacturing equipment may deposit a film on the surface of the substrate. The manufacturing equipment may deposit a multilayer film on the surface of the substrate, and may perform an etching process to form a pattern in the deposited film. For example, the manufacturing equipment may perform a chemical vapor deposition (CVD) process to deposit alternating layers on the substrate. The film may include one or more material layers formed during the deposition process, and each layer may include a specific thickness gradient (e.g., a change in thickness along the layer of the deposited film). For example, a first layer may be formed directly on the surface of the substrate (referred to as the proximal layer or proximal end of the film) and have a first thickness. After the first layer is formed on the surface of the substrate, a second layer having a second thickness may be formed on the first layer. The process continues until the deposition process is completed and the final layer of the film (referred to as the distal layer or distal end of the film) is formed.

[0015] The film can be subjected to, for example, an etching process to form a pattern on the surface of the substrate, a chemical-mechanical polishing (CMP) process to smooth the surface of the film, or any other process to produce a finished substrate. The etching process can include exposing the sample surface to a high-energy process gas (e.g., plasma) to decompose material at the surface, which can then be removed by a vacuum system.

[0016] Each substrate manufacturing process (e.g., a deposition process, an etching process, a polishing process, etc.) can be performed in a processing chamber according to a process recipe. A process recipe defines a specific set of operations to be performed on a substrate during a process and can include one or more settings associated with each operation. For example, a deposition process recipe can include a temperature setting for the processing chamber, a pressure setting for the processing chamber, a flow rate setting for a precursor of a material included in a film deposited on the substrate surface, and the like. Consequently, the thickness of each film layer can be associated with these processing chamber settings.

[0017] The film may include alternating layers of different materials. For example, the film may include alternating layers of oxide and nitride layers (oxide-nitride-oxide-nitride stack or ONON stack), alternating oxide and polysilicon layers (oxide-polysilicon-oxide-polysilicon stack or OPOP stack), etc. Each set of alternating layers may be referred to as a cycle. For example, the film may include 40 cycles (e.g., 40 sets of oxide-nitride layers), where the thickness of the layers in some cycles may be different from the thickness of the layers in other cycles. For example, the film stack may include 40 oxide-nitride cycles (e.g., 80 layers, alternating 40 layers of oxide and 40 layers of nitride), where the first cycle of the film stack has an oxide layer of a first thickness and a nitride layer of a second thickness, then the next 9 cycles of the film stack have an oxide layer of a third thickness and a nitride layer of a fourth thickness, and the last 30 cycles of the film stack have an oxide layer of a fifth thickness and a nitride layer of a sixth thickness.

[0018] The film may also include layers of different materials in varying (non-alternating) patterns (e.g., oxide-nitride-nitride-oxide stacks or ONO, oxide-nitride-oxide-oxide stacks or ONOO, etc.). During the substrate fabrication process, the thickness may vary (drift) from cycle to cycle due to continuous changes in deposition parameters and variations in process chamber conditions (e.g., buildup of contaminants, residual thickness on the walls of the process chamber, corrosion on certain components, etc.). Variations in layer thickness may cause the gas distribution plate to move closer or further away from the surface of the substrate, thereby affecting plasma flow and / or temperature and causing further deformation of the film.

[0019] In some manufacturing systems, these variations are compensated for by manually increasing or decreasing the deposition time of subsequent cycles to maintain the desired total thickness of the film stack. For example, if the first cycle has a thickness greater than that required by the process recipe, a technician can manually decrease the deposition time of the second cycle in the process recipe to produce a cycle thinner than that required by the process recipe. A processing chamber may have multiple substrate processing areas or slots. After a deposition operation, deposition drift (e.g., caused in part by different residual thicknesses on the walls of each substrate processing area) may vary for each individual substrate processing area (e.g., a slot) within a single processing chamber. These variations between substrate processing areas are also typically compensated for by manually increasing or decreasing the deposition time of subsequent layers to maintain the desired total thickness of the film stack. However, such processes are not scalable, prone to errors, and time-consuming. Such processes may also result in delays in manufacturing, loss of throughput, film defects, inconsistent and abnormal products, unscheduled user time or downtime, and defective products. Such processes also result in a significant increase in the time required to perform optimization of the process recipe parameters.

[0020] Aspects and implementations of the present disclosure address these and other shortcomings of the prior art by performing residual thickness compensation, such as by adjusting film deposition parameters based on residual thickness during substrate fabrication (eg, compensating for deposition thickness drift based on chamber residual thickness).

[0021] The processing device identifies a material associated with a substrate processing operation of the recipe. In some examples, the processing device determines that the substrate processing operation will deposit a nitride on the substrate. In some examples, the processing device determines that the substrate processing operation will deposit an oxide on the substrate. In some examples, the processing device determines that the substrate processing operation will deposit a conductor (e.g., copper, tungsten, etc.) on the substrate.

[0022] The processing device determines an expected total residual thickness value after the substrate processing operation. The expected total residual thickness value may be an expected total residual thickness value on the substrate and / or on the walls of the processing chamber until the end of the current substrate processing operation in response to the substrate processing operation of the recipe.

[0023] The processing device determines an expected material thickness value for a material associated with a substrate processing operation (e.g., a material thickness value for a current substrate processing operation) based on the expected total residual thickness value. In some embodiments, the expected material thickness value is the difference between a desired material thickness value and an actual material thickness value. In some embodiments, to determine the expected material thickness value, the processing device provides the expected total residual thickness value as an input to a trained machine learning model. In some embodiments, to determine the expected material thickness value, the processing device receives an output associated with prediction data from the trained machine learning model, the material thickness value associated with the predicted data. In some embodiments, the trained machine learning model is trained using data inputs including historical total residual thickness values ​​and a target output of historical material thickness values.

[0024] The processing device updates a recipe based on the material and the expected material thickness value of the material to generate an updated recipe. In some embodiments, to update the recipe, the processing device determines an updated time value associated with the substrate processing operation. In some embodiments, the processing device determines the updated time value associated with the substrate processing operation based on the time value associated with the substrate processing operation, the expected total residual thickness value, and the expected material thickness value. In some embodiments, updating the recipe includes determining at least one of an updated radio frequency (RF) power value for the substrate processing operation, an updated spacing value for the substrate processing operation, an updated gas flow rate value for the substrate processing operation, or an updated chamber pressure value for the substrate processing operation.

[0025] The processing device processes the substrate based on the updated recipe.

[0026] Aspects of the present disclosure bring technical advantages. Aspects of the present disclosure avoid the time-consuming and error-prone process of manually calculating the "operation time offset" for each cycle and feeding the offset into a table. Aspects of the present disclosure are scalable. Aspects of the present disclosure avoid delays, loss of throughput and / or defects in the film generated in manufacturing. Aspects of the present disclosure result in a significant reduction in the time required to optimize the parameters of the execution process recipe. The present disclosure can also result in generating diagnostic data and performing corrective actions to avoid inconsistent and abnormal products, as well as unscheduled user time or downtime. Aspects of the present disclosure allow for deposition drift compensation based on material type.

[0027] Figure 1 is a block diagram of an exemplary system 100 (an exemplary system architecture) according to certain embodiments. The system 100 (eg, via the corrective action component 122 and / or the prediction component 114) can perform the methods described herein (eg, Figures 4A to 4C System 100 includes client device 120, manufacturing equipment 124, sensor 126, metrology equipment 128, prediction server 112, and data repository 140. In some embodiments, prediction server 112 is part of prediction system 110. In some embodiments, prediction system 110 further includes server machines 170 and 180.

[0028] In some embodiments, one or more of client device 120, manufacturing equipment 124, sensor 126, metrology equipment 128, prediction server 112, data repository 140, server machine 170, and / or server machine 180 are coupled to each other via network 130 for generating prediction data 160 for performing residual-based adjustments to film deposition parameters during substrate fabrication. In some embodiments, network 130 is a public network that provides client device 120 with access to prediction server 112, data repository 140, and other publicly available computing devices. In some embodiments, network 130 is a private network that provides client device 120 with access to manufacturing equipment 124, sensor 126, metrology equipment 128, data repository 140, and other dedicated computing devices. In some embodiments, the network 130 includes one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), routers, hubs, switches, server computers, cloud computing networks, and / or combinations thereof.

[0029] In some embodiments, client device 120 includes a computing device, such as a personal computer (PC), laptop computer, mobile phone, smartphone, tablet computer, small notebook computer, etc. In some embodiments, client device 120 includes a corrective action component 122. In some embodiments, corrective action component 122 may also be included in prediction system 110 (e.g., a machine learning processing system). In some embodiments, corrective action component 122 is instead included in prediction system 110 (e.g., rather than in client device 120). Client device 120 includes an operating system that allows a user to perform one or more of merging, generating, viewing, or editing data, providing instructions to prediction system 110 (e.g., a machine learning processing system), etc.

[0030] In some embodiments, corrective action component 122 receives one or more of user input (e.g., via a graphical user interface (GUI) displayed by client device 120), property data 142, performance data 152, and the like. In some embodiments, property data 142 may be a predicted total residual thickness value, a predicted material thickness value, and the like. In some embodiments, corrective action component 122 transmits data (e.g., user input, property data 142, performance data 152, and the like) to prediction system 110, receives prediction data 160 from prediction system 110, determines a corrective action based on prediction data 160, and causes the corrective action to be implemented. In some embodiments, corrective action component 122 stores the data (e.g., user input, property data 142, performance data 152, and the like) in data repository 140, and prediction server 112 retrieves the data from data repository 140. In some embodiments, prediction server 112 stores the output of trained machine learning model 190 (e.g., prediction data 160) in data repository 140, and client device 120 retrieves the output from data repository 140. In some embodiments, corrective action component 122 receives an indication of a corrective action from prediction system 110 (eg, based on prediction data 160 ) and causes the corrective action to be performed.

[0031] Manufacturing equipment 124 can produce products such as substrates, wafers, semiconductors, electronic components, etc., following a recipe or by executing multiple operations over a period of time. Manufacturing equipment 124 may include processing chambers. Processing chambers can be used to perform any number of processes on substrates. The same or different substrate processing operations can be performed in each processing chamber or substrate processing area. Substrate processing operations may include atomic layer deposition (ALD), physical vapor deposition (PVD), chemical vapor deposition (CVD), etching, annealing, curing, pre-cleaning, metal or metal oxide removal, etc. Other processes may be performed on the substrates therein. The processing chambers may include one or more sensors configured to capture data about the substrate before, after, or during the substrate processing operations. For example, one or more sensors may be configured to capture spectral data and / or non-spectral data of a portion of the substrate during the substrate processing operations. In other or similar embodiments, one or more sensors may be configured to capture data related to the environment within the processing chamber before, after, or during the substrate processing operations. For example, one or more sensors may be configured to capture data associated with the temperature, pressure, gas concentration, etc., of the environment within the processing chamber during substrate processing operations.

[0032] The processing chamber can be used for a process of depositing a material on a substrate. For example, as previously described, the processing chamber can be a chamber for a deposition process. In some embodiments, the processing chamber includes a chamber body and a showerhead enclosing an internal volume. The showerhead may include a showerhead base and a showerhead gas distribution plate. Alternatively, in some embodiments, the showerhead may be replaced by a cover and a nozzle, or in other embodiments, by a plurality of pie-shaped showerhead compartments and a plasma generation unit. The chamber body can be made of aluminum, stainless steel or other suitable materials such as titanium (Ti). The chamber body generally includes side walls and a bottom. An exhaust port can be defined in the chamber body, and the internal volume can be coupled to a pump system. The pump system may include one or more pumps and a throttle valve for evacuating and regulating the pressure of the internal volume of the processing chamber.

[0033] The showerhead may be supported on a sidewall of the chamber body. The showerhead (or lid) may be opened to allow access to the interior volume of the processing chamber and may provide a seal for the processing chamber when closed. A gas panel may be coupled to the processing chamber to provide processing and / or cleaning gases to the interior volume via the showerhead or lid and nozzles (e.g., via holes in the showerhead or lid and nozzles). For example, the gas panel may provide precursors for film materials deposited on the surface of the substrate. In some embodiments, the precursors may include silicon-based precursors or boron-based precursors. The showerhead may include a gas distribution plate (GDP) and may have a plurality of gas delivery holes (also referred to as channels) extending through the GDP. A substrate support assembly is positioned below the showerhead in the interior volume of the processing chamber. The substrate support assembly holds the substrate during processing (e.g., during deposition) using, for example, an electrostatic chuck.

[0034] In some embodiments, the processing chamber may include metrology equipment (e.g., metrology equipment 128) and / or sensors (e.g., sensor 126) configured to generate in-situ metrology measurements (e.g., metrology data) and / or sensor measurements (e.g., sensor data) during a process performed in the processing chamber. In some embodiments, the metrology measurements and / or sensor measurements may be a subset of property data 142 and / or performance data 152. The metrology equipment and / or sensors may be operably coupled to a system controller. In some embodiments, the metrology equipment may be configured to generate metrology measurements (e.g., thickness) of the film during a specific instance of the deposition process. In some embodiments, the sensors may be configured to generate sensor measurements (e.g., thickness) of the film during a specific instance of the deposition process. The system controller may generate a thickness profile of the film based on the metrology measurements received from the metrology equipment. The system controller may generate a thickness profile of the film based on the sensor measurements received from the sensors. In other or similar embodiments, the processing chamber does not include metrology equipment. In such embodiments, the system controller may receive one or more metrology measurements of the film after the deposition process is completed at the processing chamber. The system controller may determine a deposition rate based on the one or more metrology measurements and may correlate a thickness profile of the resulting film based on the determined concentration gradient and the determined deposition rate of the deposition process.

[0035] The manufacturing equipment 124 can perform processes on substrates (e.g., wafers, etc.) at a processing chamber. Examples of substrate processes include deposition processes for depositing one or more film layers on a substrate surface, etching processes for forming patterns on a substrate surface, and the like. The manufacturing equipment 124 can perform each process according to a process recipe. A process recipe defines a set of specific operations to be performed on a substrate during a process and can include one or more settings associated with each operation. For example, a deposition process recipe can include a temperature setting for the processing chamber, a pressure setting for the processing chamber, a flow rate setting for a precursor of a material included in a film deposited on the substrate surface, and the like.

[0036] In some embodiments, the manufacturing equipment 124 includes sensors 126 configured to generate data related to substrates processed at the manufacturing system 100. For example, a processing chamber may include one or more sensors configured to generate a residual thickness profile associated with the processing chamber (e.g., the thickness of material on the processing chamber walls) before, during, and / or after a process (e.g., a deposition process). For example, the processing chamber may include one or more sensors configured to generate spectral or non-spectral data associated with the substrate before, during, and / or after a process (e.g., a deposition process) is performed on the substrate. In some embodiments, the spectral data generated by the sensors 126 may indicate the concentration of one or more materials deposited on the substrate surface. Sensors 126 configured to generate spectral data associated with the substrate may include reflectometer sensors, ellipsometer sensors, thermal spectral sensors, capacitance sensors, and the like. Sensors 126 configured to generate non-spectral data associated with the substrate may include residual thickness sensors, temperature sensors, pressure sensors, flow rate sensors, voltage sensors, and the like.

[0037] The metrology equipment 128 may provide metrology data related to substrates processed by the manufacturing equipment 124. In some embodiments, the metrology data may be a subset of the property data 142 and / or the performance data 152. The metrology data may include values ​​for film property data (e.g., wafer-space film properties), dimensions (e.g., thickness, height, etc.), dielectric constant, dopant concentration, density, defects, etc. In some embodiments, the metrology data may further include values ​​for one or more types of surface profile property data (e.g., etch rate, etch rate uniformity, critical dimension of one or more features included on the substrate surface, critical dimension uniformity across the entire substrate surface, edge position error, etc.). The metrology data may be for finished or semi-finished products. The metrology data may be different for each substrate. The metrology data may be generated using, for example, reflectometry techniques, ellipsometry techniques, transmission electron microscopy (TEM), etc.

[0038] In some embodiments, the prediction server 112, the server machine 170, and the server machine 180 each include one or more computing devices, such as a rack server, a router computer, a server computer, a personal computer, a mainframe computer, a laptop computer, a tablet computer, a desktop computer, a graphics processing unit (GPU), an accelerator application-specific integrated circuit (ASIC) (e.g., a tensor processing unit (TPU)), etc.

[0039] The prediction server 112 includes a prediction component 114. In some embodiments, the prediction component 114 identifies (e.g., receives from the client device 120, retrieves from the data repository 140) property data 142 (e.g., expected total residual thickness value, expected material thickness value, etc.) and generates prediction data 160 associated with performing corrective actions (e.g., updating a recipe, updating deposition operating parameters, updating process operating parameters, etc.). In some embodiments, the prediction component 114 determines the prediction data 160 using one or more trained machine learning models 190. In some embodiments, the trained machine learning models 190 are trained using historical property data 144 and historical performance data 154.

[0040] For example, the deposition operation parameters may include deposition times for each layer and / or cycle of the process recipe, a temperature setting for the processing chamber, a pressure setting for the processing chamber, a flow rate setting for a precursor of a material included in a film deposited on the substrate surface, a showerhead height, etc. In some embodiments, updating the recipe includes updating the deposition operation parameters. For example, the deposition operation parameters may include a time value, an updated time value, a recipe, an updated recipe, an RF power for a substrate processing operation, a spacing value for a substrate processing operation, a gas flow value for a substrate processing operation, or a chamber pressure value for a substrate processing operation, an updated RF power for a substrate processing operation, an updated spacing value for a substrate processing operation, an updated gas flow value for a substrate processing operation, an updated chamber pressure value for a substrate processing operation, etc.

[0041] In some embodiments, the prediction system 110 (e.g., prediction server 112, prediction component 114) generates the prediction data 160 using supervised machine learning (e.g., a supervised dataset, historical property data 144 labeled with historical performance data 154, etc.). In some embodiments, the prediction system 110 generates the prediction data 160 using semi-supervised learning (e.g., a semi-supervised dataset, performance data 152 being a predicted percentage, etc.). In some embodiments, the prediction system 110 generates the prediction data 160 using unsupervised machine learning (e.g., an unsupervised dataset, clustering, clustering based on historical property data 144, etc.).

[0042] In some embodiments, the fabrication equipment 124 (e.g., a deposition chamber, a cluster tool, a wafer back grinding system, a wafer saw system, a die attach machine, a wire bonder, a die encapsulation system, a molding system, a sealing system, a can welder, a deflash / trim / form / singulation (DTFS) machine, a branding system, a wire trim system, etc.) is part of a substrate processing system (e.g., an integrated processing system). The fabrication equipment 124 includes one or more of a controller, a housing system (e.g., a substrate carrier, a front opening unified pod (FOUP), a self-learning FOUP, a process kit housing system, a substrate housing system, a cassette, etc.), a side storage pod (SSP), an aligner device (e.g., an alignment chamber), a factory interface (e.g., an equipment front end module (EFEM)), a load lock, a transfer chamber, one or more process chambers, a robotic arm (e.g., disposed in the transfer chamber, disposed in the front interface, etc.), and / or the like. The housing system, SSP, and load lock mounted to the factory interface, and a robotic arm positioned in the factory interface are used to transfer contents (e.g., substrates, process kit rings, carriers, verification wafers, etc.) between the housing system, SSP, load lock, and the factory interface. An aligner device is positioned in the factory interface to align the contents. The load lock and process chamber mounted to the transfer chamber, and the robotic arm positioned in the transfer chamber transfer contents (e.g., substrates, process kit rings, carriers, verification wafers, etc.) between the load lock, the process chamber, and the transfer chamber. In some embodiments, the manufacturing equipment 124 includes components of a substrate processing system. In some embodiments, property data 142 for a process chamber or substrate is generated by the process chamber or substrate undergoing one or more processes (e.g., deposition, etching, heating, cooling, transfer, treatment, flow, etc.) performed by the components of the manufacturing equipment 124.

[0043] In some embodiments, the sensors 126 provide property data 142 (eg, sensor values, such as historical and current sensor values) of a process chamber (eg, expected total residual thickness) or substrate (eg, expected material thickness) processed by the fabrication equipment 124 .

[0044] In some embodiments, the sensors 126 include one or more of the following: metrology tools such as an ellipsometer (for determining the properties and surface of thin films by measuring material properties such as layer thickness, optical constants, surface roughness, composition, and optical anisotropy), an ion mill (for preparing heterogeneous bulk materials when uniformly thinning large-area materials), a capacitance-voltage (CV) system (for measuring CV and capacitance-time (Ct) characteristics of semiconductor devices), an interferometer (for measuring distance with respect to wavelength and determining the wavelength of a specific light source), a source measure unit (SME), a magnetometer, optical and imaging systems, a surface brightness meter, a wafer prober (for testing semiconductor wafers before they are separated into individual dies or chips), an imaging station, a critical-dimension scanning electron microscope (CD-SEM) (for ensuring the stability of the manufacturing process by measuring the critical dimensions of the substrate), a reflectometer (for measuring the reflectivity and emissivity of a surface), a resistance probe (for measuring the resistivity of a thin film), a resistance high-energy electron diffraction (REED) system, and a surface probing system. diffraction; RHEED) system (used to measure or monitor the crystal structure or crystal orientation of epitaxial thin films of silicon or other materials), X-ray diffractometer (used to clearly determine the crystal structure, crystal orientation, film thickness and residual stress in silicon wafers, epitaxial films or other substrates), etc.

[0045] In some embodiments, property data 142 is used for equipment health and / or product health (eg, product quality). In some embodiments, property data 142 is received over a period of time.

[0046] In some embodiments, the sensor 126 and / or the metrology equipment 128 provides property data 142 including one or more of: morphological data, dimensional property data, dimensional property data, image data, scanning electron microscope (SEM) images, energy dispersive x-ray (EDX) images, defect distribution data, spatial location data, elemental analysis data, wafer feature data, chip layers, chip layout data, edge data, grayscale data, signal-to-noise data, temperature data, spacing data, current data, power data, voltage data, and / or the like.

[0047] In some embodiments, the property data includes morphological data (e.g., data related to the substrate's topography, such as deposited layer thickness or residual thickness). In some embodiments, the property data 142 includes size property data (e.g., data describing the size of a substrate property). In some embodiments, the property data 142 includes dimensional property data (e.g., data describing the dimension of a substrate property). In some embodiments, the property data 142 includes SEM images (e.g., images captured by a scanning electron microscope that uses a focused electron beam to scan a substrate surface to generate high-resolution images). In some embodiments, the property data 142 includes EDX images (e.g., images generated from data collected using x-ray technology to identify the elemental composition of a material). In some embodiments, the property data 142 includes defect distribution data (e.g., data describing the distribution of defects on a substrate, such as in space, time, etc.). In some embodiments, the property data 142 includes spatial location data (e.g., data describing the spatial location of a substrate property, defect, element, etc.). In some embodiments, the property data 142 includes elemental analysis data (e.g., data describing the elemental composition of a substrate). In some embodiments, the property data 142 includes wafer characterization data (e.g., data describing the distribution of defects on a substrate wafer resulting from a single manufacturing issue). In some embodiments, the property data 142 includes chip layer data (e.g., associated with a layer or operation in a substrate manufacturing process). In some embodiments, the property data 142 includes chip layout data (e.g., data describing the substrate layout with respect to a planar geometry). In some embodiments, the property data 142 includes edge data (e.g., data describing the edge of a wafer, such as chip edge, wafer edge thickness, wafer bow and / or warpage, etc.). In some embodiments, the property data 142 includes grayscale data (e.g., data describing pixel brightness of a substrate image) and signal-to-noise ratio data (e.g., data describing the signal-to-noise ratio of a substrate as measured, for example, using spectrometric equipment).

[0048] In some embodiments, property data 142 (e.g., historical property data 144, current property data 146, etc.) is processed (e.g., by client device 120 and / or by prediction server 112). In some embodiments, processing of property data 142 includes generating features. In some embodiments, a feature is a pattern in property data 142 (e.g., slope, width, height, peak value, etc.) or a combination of values ​​from property data 142 (e.g., power derived from voltage and current, etc.). In some embodiments, property data 142 includes features that are used by prediction component 114 to obtain prediction data 160.

[0049] In some embodiments, the metrology equipment 128 can be included as part of the manufacturing equipment 124. For example, the metrology equipment 128 can be included within or coupled to a processing chamber and configured to generate metrology data (e.g., property data 142, performance data 152, etc.) of the processing chamber or substrate before, during, and / or after a process (e.g., a deposition process, an etching process, etc.) while the substrate remains in the processing chamber. In some cases, the metrology equipment 128 can be referred to as in-situ metrology equipment. In another example, the metrology equipment 128 can be coupled to another station of the manufacturing equipment 124. For example, the metrology equipment can be coupled to a transfer chamber, a load lock, or a factory interface.

[0050] In some embodiments, the sensor 126 may be included as part of the manufacturing equipment 124. For example, the sensor 126 may be included within or coupled to a processing chamber and configured to generate sensor data of the interior of the processing chamber or the substrate while the substrate remains in the processing chamber before, during, and / or after a process (e.g., a deposition process, an etching process, etc.). In some cases, the sensor 126 may be referred to as an in-situ sensor. In another example, the sensor 126 may be coupled to another station of the manufacturing equipment 124. For example, the sensor may be coupled to a transfer chamber, a load lock, or a factory interface.

[0051] In some embodiments, metrology equipment 128 (e.g., ellipsometer equipment, imaging equipment, spectroscopic equipment, etc.) is used to determine metrology data (e.g., inspection data, image data, spectroscopic data, ellipsometer data, material composition, optical or structural data, etc.) corresponding to the interior (e.g., surface) of a processing chamber or corresponding to a substrate produced by fabrication equipment 124 (e.g., substrate processing equipment). In some examples, after a substrate is processed by fabrication equipment 124, metrology equipment 128 is used to inspect a portion (e.g., a layer) of the substrate and / or the interior of the processing chamber. In some embodiments, metrology equipment 128 performs scanning acoustic microscopy (SAM), ultrasonic inspection, x-ray inspection, and / or computed tomography (CT) inspection. In some examples, after fabrication equipment 124 deposits one or more layers on a substrate, metrology equipment 128 is used to determine the quality of the processed substrate (e.g., layer thickness, layer uniformity, interlayer spacing, etc.). In some embodiments, metrology equipment 128 includes an imaging device (e.g., SAM equipment, ultrasonic equipment, x-ray equipment, CT equipment, etc.). In some embodiments, property data 142 includes sensor data from sensors 126 and / or metrology data from metrology equipment 128. In some embodiments, property data 142 includes sensor data from sensors 126 and / or metrology data from metrology equipment 128 located in situ (within the process chamber). In some embodiments, performance data 152 includes user input via client device 120 and / or metrology data from metrology equipment 128. Property data 142 may include metrology data from a first subset of metrology equipment 128, and performance data 152 may include metrology data from a second subset of metrology equipment 128.

[0052] In some embodiments, performance data 152 may be associated with the performance of a recipe (e.g., a deposition recipe, an updated recipe, etc.). For example, performance data 152 may be data for substrates or processing chambers that have undergone the recipe and / or processing operations of the recipe.

[0053] In some embodiments, the property data 142 may be derived from metrology data and / or sensor data. Metrology data may be data describing the metrology of the substrate. Sensor data may be data describing the conditions and characteristics within the processing chamber. In some embodiments, the property data may include a deposition thickness value (e.g., the actual or expected amount of material deposited on the substrate). In some embodiments, the property data may include a total residual thickness value (e.g., the actual or expected amount of material remaining deposited on the chamber walls). In some embodiments, the deposition thickness value is derived from a deposition operation recipe (e.g., the deposition thickness is equal to the expected deposition thickness for the deposition operation).

[0054] In some embodiments, the actual deposited thickness value or the actual residual thickness value may refer to the amount of material actually deposited on the substrate or remaining deposited on the walls of the processing chamber after the deposition operation. The actual deposited thickness value or the actual total residual thickness value may be measured using metrology equipment or sensors.

[0055] In some embodiments, the expected deposited thickness value or the expected residual thickness value may refer to the amount of material expected to be deposited on the substrate or remaining deposited on the walls of the processing chamber after a deposition operation. In some embodiments, the expected deposited thickness value or the actual total residual thickness value may be derived from a process recipe (e.g., deposition operation parameters). In some embodiments, the expected deposited thickness value or the actual residual thickness value may be derived from historical actual thickness values ​​corresponding to the same deposition operation (e.g., historical actual thickness values ​​on the substrate and / or the walls of the processing chamber).

[0056] In some embodiments, data repository 140 is a memory (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or other type of component or device capable of storing data. In some embodiments, data repository 140 includes multiple storage components (e.g., multiple drives or multiple databases) across multiple computing devices (e.g., multiple server computers). In some embodiments, data repository 140 stores one or more of property data 142, performance data 152, and / or prediction data 160.

[0057] The property data 142 may include an expected total residual thickness value, an expected material thickness value, an actual total residual thickness value, an actual material thickness value, an expected total residual thickness value, and a desired material thickness value, etc. The performance data 152 may include an expected total residual thickness value, an expected material thickness value, an actual total residual thickness value, an actual material thickness value, an expected total residual thickness value, and a desired material thickness value, etc.

[0058] In some embodiments, the data repository 140 may store expected total residual thickness values ​​and expected material thickness values. The expected total residual thickness value may include one or more data points associated with an expected residual film profile expected to be generated by a particular process recipe. In some embodiments, the expected total residual thickness value may include an expected film thickness, an expected thickness of one or more film layers, and / or an expected thickness of one or more film cycles, etc. The expected material thickness value may include one or more data points associated with the current film thickness generated by the manufacturing equipment 124. For example, the expected material thickness value may include an expected thickness of the film, an expected thickness of one or more film layers, and / or an expected thickness of one or more film cycles, etc. For example, the expected material thickness value may include a measured thickness of the film, a measured thickness of one or more film layers, and / or a measured thickness of one or more film cycles, etc. The expected material thickness value may be measured using the metrology equipment 128. The updated recipe may include one or more adjustments or offsets to be applied to parameters of the processing chamber or process recipe. For example, an updated recipe may include adjustments to the deposition time of a film layer and / or cycle, a temperature setting for a processing chamber, a pressure setting for a processing chamber, a flow rate setting for a precursor of a material included in a film deposited on a substrate surface, a power supplied to a processing chamber, a ratio of two or more settings, etc. An updated recipe may be generated by comparing a predicted total residual thickness (e.g., a residual thickness predicted to be generated by a process recipe) and determining adjustments to be applied to the process recipe parameters using an algorithm, a library of known failure modes, etc. The updated recipe may be applied to steps associated with a deposition process, an etching process, etc.

[0059] In some embodiments, the data repository 140 may be configured to store data associated with known failure modes. A failure mode may be one or more values ​​(e.g., vectors, scalars, etc.) associated with one or more problems or failures associated with a processing chamber subsystem. In some embodiments, a failure mode may be associated with a corrective action. For example, a failure mode may include parameter adjustment steps to correct the problem or failure indicated by the failure mode. For example, the prediction system may compare a determined failure mode with a library of known failure modes to determine the type of failure experienced by the subsystem, the cause of the failure, a recommended corrective action to correct the failure, and the like.

[0060] In some embodiments, the corrective action can be, for example, updating a recipe (e.g., a process recipe, a deposition recipe), determining a predicted material thickness value, processing the substrate based on the updated recipe, training a machine learning model using data inputs including historical residual thickness values ​​and target outputs including historical performance data, receiving an output associated with the predicted data using the trained machine learning model, wherein the predicted material thickness value is associated with the predicted data, and / or the like. In some embodiments, updating the process recipe can include updating process parameters (e.g., deposition time, flow rate, temperature, etc.). In some embodiments, the corrective action includes providing machine learning (e.g., resulting in a recipe update, a process / operating parameter update, etc. based on the predicted data 160).

[0061] In some embodiments, the prediction data 160 is associated with a corrective action. In some embodiments, the corrective action is associated with one or more of updating a recipe (e.g., a process recipe, a deposition recipe), determining an expected material thickness value, processing a substrate based on the updated recipe, training a machine learning model using data inputs including historical residual thickness values ​​and target outputs including historical property data, receiving outputs related to the prediction data using the trained machine learning model, repairing one or more pieces of fabrication equipment 124, replacing one or more pieces of fabrication equipment 124, computational process control (CPC), statistical process control (SPC) (e.g., SPC compared to a 3-sigma chart, etc.), advanced process control (APC), model-based process control, preventive operational maintenance, design optimization, manufacturing parameter updates, wafer recipe modifications, feedback control, machine learning modifications, etc.

[0062] In some embodiments, the data repository 140 may be configured to store data that is inaccessible to users of the manufacturing system. For example, a user of the manufacturing system (e.g., an operator) cannot access process data, spectral data, contextual data, etc. obtained for a substrate being processed in the manufacturing system. In some embodiments, all data stored in the data repository 140 is inaccessible to users of the manufacturing system. In some embodiments, a portion of the data stored in the data repository 140 is inaccessible to users of the manufacturing system, while another portion of the data stored in the data repository 140 is accessible to the user. In some embodiments, one or more portions of the data stored at the data repository 140 may be encrypted using an encryption mechanism unknown to the user (e.g., using a private encryption key to encrypt the data). In some embodiments, the data repository 140 may include multiple data storage devices, wherein data that is inaccessible to the user is stored in one or more first data repositories, and data that is accessible to the user is stored in one or more second data repositories.

[0063] The property data 142 includes historical property data 144 and current property data 146. In some embodiments, the property data 142 (e.g., sensor data) may include an expected total residual thickness value, an expected material thickness value, an updated RF power for a substrate processing operation, an updated spacing value for a substrate processing operation, an updated gas flow rate value for a substrate processing operation, an updated chamber pressure value for a substrate processing operation, deposited residual thickness data (actual), pressure data, temperature data, a temperature range, power data, a comparison parameter for comparing inspection data to threshold data, threshold data, cooling rate data, a cooling rate range, etc. In some embodiments, at least a portion of the property data 142 comes from the sensors 126 and / or the metrology equipment 128.

[0064] The performance data 152 includes historical performance data 154 and current performance data 156. The performance data 152 may indicate whether the substrate is correctly designed, correctly manufactured, consistent with other substrates, and / or operating properly. The performance data 152 may indicate whether the substrate processing operation is being performed accurately. For example, the performance data 152 may indicate the actual thickness deposited on the wafer or in the processing chamber during a deposition operation (e.g., material thickness value, total residual thickness value, historical material thickness value, deposition thickness on the processing chamber wall, etc.). The performance data 152 may indicate whether the substrate processing operation (e.g., deposition operation) is being performed effectively. For example, the performance data 152 may indicate deposition drift before, during, or after a deposition operation, and may indicate a deposition residual thickness value, a material thickness value, etc.

[0065] In some embodiments, at least a portion of the performance data 152 is associated with the quality of substrates produced by the manufacturing equipment 124. In some embodiments, at least a portion of the performance data 152 is based on metrology data from the metrology equipment 128 (e.g., historical performance data 154 includes metrology data indicating correctly processed substrates, substrate property data, yield, material thickness values, etc.) or sensors 126 (e.g., historical performance data 154 includes sensor data indicating correctly processed substrates, substrate property data, yield, material thickness values, etc.). In some embodiments, at least a portion of the performance data 152 is based on inspections of substrates or within the processing chamber (e.g., current performance data 156 based on actual inspections). In some embodiments, the performance data 152 includes user input (e.g., via the client device 120) indicating substrate quality or deposition drift within the processing chamber / substrate processing area or on the substrate. In some embodiments, the performance data 152 includes indications of absolute values ​​(e.g., inspection data for a substrate indicates missing threshold data by a calculated value, or a drift value is missing a calculated threshold drift value by a calculated value) or relative values ​​(e.g., inspection data for a film deposition indicates missing threshold data by 5%, or a drift value is missing a threshold drift value by 5%). In some embodiments, performance data 152 indicates that a threshold amount of error is met (e.g., deposition drift after a deposition operation is at least 5% error, production is at least 5% error, flow is at least 5% error, deformation is at least 5% error, specification limit).

[0066] In some embodiments, historical data includes one or more of historical property data 144 and / or historical performance data 154 (e.g., at least a portion of which is used to train machine learning model 190). Current data includes one or more of current property data 146 and / or current performance data 156 (e.g., at least a portion of which is input to trained machine learning model 190 after model 190 is trained using historical data). In some embodiments, current data is used to retrain trained machine learning model 190.

[0067] In some embodiments, the prediction data 160 is used to take corrective actions on a process recipe, deposition operating parameters of a recipe / operation, manufacturing equipment, substrate processing system, or substrate processing equipment components.

[0068] Performing multiple types of metrology on a multi-layer product or processing chamber to determine whether to take corrective action can be costly in terms of time, metrology equipment 128, energy, bandwidth required to transmit metrology data, and processor overhead for processing the metrology data. By providing property data 142 to model 190 and receiving prediction data 160 from model 190, system 100 has the technical advantage of avoiding the costly process of using metrology equipment 128 and / or sensors on a multi-layer product and processing chamber, thereby avoiding wasted time and discarded substrates.

[0069] Performing a manufacturing process (e.g., deposition) using manufacturing equipment 124 and / or manufacturing parameters (e.g., deposition operating parameters) that can produce defective products or damage the manufacturing equipment is costly in terms of time, energy, product, manufacturing equipment 124, and the cost of identifying corrective actions to avoid producing defective products. By providing property data 142 to model 190, receiving prediction data 160 from model 190, and generating corrective actions (e.g., updating a recipe) based on prediction data 160, system 100 has the technical advantage of eliminating the cost of producing, identifying, and discarding defective substrates.

[0070] In some embodiments, the prediction system 110 further includes a server machine 170 and a server machine 180. The server machine 170 includes a data set generator 172 that is capable of generating a data set (e.g., a set of data inputs and a set of target outputs) to train, validate, and / or test the machine learning model 190. The data set generator 172 has the functionality of data collection, compilation, simplification, and / or partitioning to put the data in a form suitable for machine learning. In some embodiments (e.g., for small data sets), partitioning (e.g., explicit partitioning) for post-training validation is not used. Repeated cross-validation (e.g., 5-fold cross-validation, leave-one-out cross-validation) can be used during training, where a given data set is actually repeatedly partitioned into different training and validation sets during training. A model (e.g., the best model, the model with the highest accuracy, etc.) is selected from the model vectors on the automatically separated combined subsets. In some embodiments, the dataset generator 172 may explicitly divide the historical data (e.g., the historical property data 144 and the corresponding historical performance data 154) into a training set (e.g., 60% of the historical data), a validation set (e.g., 20% of the historical data), and a test set (e.g., 20% of the historical data). Figure 2Some operations of the dataset generator 172 are described in detail. In some embodiments, the prediction system 110 (e.g., via the prediction component 114) generates multiple sets of features (e.g., training features). In some examples, a first set of features corresponds to a first set of property data types (e.g., from a first set of sensors, a first value combination from the first set of sensors, a first value pattern from the first set of sensors), the first set of property data types corresponding to each dataset (e.g., a training set, a validation set, and a test set), and a second set of features corresponds to a second set of property data types (e.g., from a second set of sensors different from the first set of sensors, a second value combination different from the first value combination, a second pattern different from the first pattern), the second set of property data types corresponding to each dataset.

[0071] The server machine 180 includes a training engine 182, a validation engine 184, a selection engine 185, and / or a testing engine 186. In some embodiments, the engines (e.g., the training engine 182, the validation engine 184, the selection engine 185, and the testing engine 186) represent hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, a processing device, etc.), software (e.g., instructions running on a processing device, a general-purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. The training engine 182 is capable of training a machine learning model 190 using one or more sets of features associated with a training set from the dataset generator 172. In some embodiments, the training engine 182 generates multiple trained machine learning models 190, each of which corresponds to a different set of parameters (e.g., property data 142) and corresponding responses (e.g., performance data 152) of the training set. In some embodiments, multiple models are trained on the same parameters with different objectives for the purpose of modeling multiple effects. In some examples, a first training machine learning model is trained using property data 142 from all sensors 126 (e.g., sensors 1 through 5), a second training machine learning model is trained using a first subset of the property data (e.g., from sensors 1, 2, and 4), and a third training machine learning model is trained using a second subset of the property data that partially overlaps with the first feature subset (e.g., from sensors 1, 3, 4, and 5).

[0072] The validation engine 184 can validate the trained machine learning models 190 using the corresponding feature set of the validation set from the dataset generator 172. For example, a first trained machine learning model 190 trained using the first set of features of the training set is validated using the first set of features of the validation set. The validation engine 184 determines the accuracy of each trained machine learning model 190 based on the corresponding feature set of the validation set. The validation engine 184 evaluates and marks (e.g., to be discarded) trained machine learning models 190 whose accuracy does not meet the threshold accuracy. In some embodiments, the selection engine 185 can select one or more trained machine learning models 190 that have an accuracy that meets the threshold accuracy. In some embodiments, the selection engine 185 can select the trained machine learning model 190 with the highest accuracy among the trained machine learning models 190.

[0073] The testing engine 186 can test the trained machine learning models 190 using the corresponding feature set of the test set from the dataset generator 172. For example, the first trained machine learning model 190 trained using the first set of features of the training set is tested using the first set of features of the test set. The testing engine 186 determines the trained machine learning model 190 with the highest accuracy among all the trained machine learning models based on the test set.

[0074] In some embodiments, the machine learning model 190 (e.g., for classification) refers to a model artifact generated by the training engine 182 using a training set comprising data inputs and corresponding target outputs (e.g., correctly classifying the condition or order of the corresponding training input). Patterns in the dataset that map the data inputs to the target outputs (correct classification or level) can be found, and images capturing these patterns are provided to the machine learning model 190. In some embodiments, the machine learning model 190 uses one or more of Gaussian Process Regression (GPR), Gaussian Process Classification (GPC), Bayesian neural networks, neural network Gaussian processes, deep belief networks, Gaussian mixture models, or other probabilistic learning methods. Non-probabilistic methods may also be used, including one or more of support vector machines (SVM), radial basis functions (RBF), clustering, nearest neighbor algorithms (k-NN), linear regression, random forests, neural networks (e.g., artificial neural networks), etc. In some embodiments, the machine learning model 190 is a multivariate analysis (MVA) regression model.

[0075] The prediction component 114 provides the current property data 146 to the trained machine learning model 190 (e.g., as an input) and runs the trained machine learning model 190 (e.g., runs it on the input to obtain one or more outputs). The prediction component 114 can determine (e.g., extract) prediction data 160 from the trained machine learning model 190 and determine (e.g., extract) uncertainty data indicating a confidence level that the prediction data 160 corresponds to the current performance data 156. In some embodiments, the prediction component 114 or the corrective action component 122 uses the uncertainty data (e.g., an uncertainty function or a derived function derived from the uncertainty function) to determine whether to use the prediction data 160 to perform a corrective action or whether to further train the model 190.

[0076] For purposes of illustration and not limitation, aspects of the present disclosure describe the training of one or more machine learning models 190 using historical data (i.e., prior data, historical property data 144, and historical performance data 154) and providing current property data 146 to one or more trained probabilistic machine learning models 190 to determine prediction data 160. In other implementations, heuristic models or rule-based models are used to determine prediction data 160 (e.g., without using trained machine learning models). In other implementations, non-probabilistic machine learning models may be used. The prediction component 114 monitors the historical property data 144 and the historical performance data 154. In some embodiments, the prediction component 114 monitors the historical property data 144 and the historical performance data 154. Figure 2 The data inputs 210 describe any information being monitored or otherwise used in a heuristic or rule-based model.

[0077] In some embodiments, the functionality of client device 120, prediction server 112, server machine 170, and server machine 180 is provided by a smaller number of machines. For example, in some embodiments, server machines 170 and 180 are integrated into a single machine, while in some other embodiments, server machine 170, server machine 180, and prediction server 112 are integrated into a single machine. In some embodiments, client device 120 and prediction server 112 are integrated into a single machine.

[0078] In general, the functions described as being performed by client device 120, prediction server 112, server machine 170, and server machine 180 in one embodiment may also be performed on prediction server 112 in other embodiments, as desired. Furthermore, the functions attributed to a particular component may be performed by different or multiple components operating together. For example, in some embodiments, prediction server 112 determines a corrective action based on prediction data 160. In another example, client device 120 determines prediction data 160 based on data received from a trained machine learning model.

[0079] In addition, the functionality of a particular component may be performed by different or multiple components operating together. In some embodiments, one or more of prediction server 112, server machine 170, or server machine 180 is accessed as a service provided to other systems or devices via an appropriate application programming interface (API).

[0080] In some embodiments, a "user" is represented as a single individual. However, other embodiments of the present disclosure encompass a "user" that is an entity controlled by multiple users and / or automation sources. In some examples, a group of individual users united into a group of administrators is considered a "user."

[0081] While embodiments of the present disclosure are discussed with respect to determining predicted data 160 for residual thickness compensation (e.g., updating a recipe based on thickness values) during substrate fabrication in a fabrication facility (e.g., a substrate processing facility), in some embodiments, the present disclosure may also be generally applicable to corrective actions in a fabrication facility. Embodiments may generally be applicable to determining part quality based on different types of data.

[0082] Figure 2 1 illustrates a method for providing a machine learning model (e.g., associated with residual thickness compensation, methods 400A-400C, etc.) (e.g., Figure 1 The dataset generator 272 (e.g., Figure 1 In some embodiments, the dataset generator 272 is Figure 1 Part of the server machine 170. Figure 2 The dataset generator 272 generates a dataset that can be used to train a machine learning model (e.g., see Figure 4B ) to cause the corrective action to be performed (see, for example, Figure 4C ).

[0083] Dataset generator 272 (e.g., Figure 1 The dataset generator 172 of Figure 1 The dataset generator 272 uses the historical property data 244 (e.g., Figure 1 Historical property data 144) and historical performance data 254 (e.g., Figure 1 The historical performance data 154) is used to create a data set. Figure 2 The system 200 illustrates a data set generator 272, a data input 210, and a target output 220 (eg, target data).

[0084] In some embodiments, the dataset generator 272 generates a dataset (e.g., a training set, a validation set, a test set) that includes one or more data inputs 210 (e.g., a training input, a validation input, a test input). In some embodiments, the dataset generator 272 does not generate a target output (e.g., for unsupervised learning). In some embodiments, the dataset generator generates one or more target outputs 220 corresponding to the data inputs 210 (e.g., for supervised learning). The dataset may also include mapping data that maps the data inputs 210 to the target outputs 220. The data inputs 210 are also referred to as “features,” “attributes,” or “information.” In some embodiments, the dataset generator 272 provides the dataset to the training engine 182, the validation engine 184, or the testing engine 186, where the dataset is used to train, validate, or test the machine learning model 190 (e.g., in connection with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A to 400C, etc.).

[0085] In some embodiments, the data set generator 272 generates the data input 210 and the target output 220. In some embodiments, the data input 210 includes one or more sets of historical property data 244 (e.g., total residual thickness values, material thickness values, etc.) (e.g., associated with adjustments to film deposition parameters based on residual thickness during substrate fabrication, methods 400A to 400C, etc.). In some embodiments, the historical property data 244 includes one or more of the following: property data from one or more types of sensors and / or metrology equipment, combinations of property data from one or more types of sensors and / or metrology equipment, patterns of property data from one or more types of sensors and / or metrology equipment, etc.

[0086] In some embodiments, dataset generator 272 generates a first data input corresponding to a first set of historical property data 244A to train, validate, or test a first machine learning model, and dataset generator 272 generates a second data input corresponding to a second set of historical property data 244B to train, validate, or test a second machine learning model (e.g., associated with adjustment of film deposition parameters based on residual thickness during substrate manufacturing, methods 400A to 400C, etc.).

[0087] In some embodiments, the dataset generator 272 discretizes (e.g., segments) one or more of the data input 210 or the target output 220 (e.g., for a classification algorithm for a regression problem). Discretization of the data input 210 or the target output 220 (e.g., segmentation via a sliding window) transforms the continuous values ​​of the variable into discrete values. In some embodiments, the discrete values ​​of the data input 210 indicate discrete historical property data 144 to obtain the target output 220 (e.g., discrete historical performance data 154).

[0088] The data input 210 and target output 220 used to train, validate, or test the machine learning model include information about a particular facility (e.g., a particular substrate manufacturing facility). In some examples, the historical property data 244 and the historical performance data 254 are for the same manufacturing facility (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate manufacturing, methods 400A to 400C, etc.).

[0089] In some embodiments, the information used to train the machine learning model is from a particular type of fabrication equipment 124 at a fabrication facility having particular characteristics and allows the trained machine learning model (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A-400C, etc.) to determine the effect of a particular set of fabrication equipment 124 based on input of current parameters (e.g., current property data 146) associated with one or more components that share a particular set of characteristics. In some embodiments, the information used to train the machine learning model is for components from two or more fabrication facilities and allows the trained machine learning model to determine the effect of a component based on input from one fabrication facility.

[0090] In some embodiments, after a data set is generated and used to train, validate, or test the machine learning model 190, the machine learning model 190 (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A-400C, etc.) is further trained, validated, or tested (e.g., Figure 1 current performance data 156) or adjustments (e.g., adjusting weights associated with input data to the machine learning model 190, such as connection weights in a neural network).

[0091] The machine learning model processes inputs to generate outputs (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A to 400C, etc.). The artificial neural network includes an input layer consisting of values ​​in the data points. The next layer is called a hidden layer, and the nodes at the hidden layer each receive one or more of the input values. Each node contains parameters (e.g., weights) to be applied to the input values. Thus, each node essentially inputs the input values ​​into a multivariate function (e.g., a nonlinear mathematical transformation) to produce an output value. The next layer can be another hidden layer or an output layer. In either case, the nodes at the next layer receive output values ​​from the nodes at the previous layer, and each node applies weights to those values ​​and then generates its own output value. This can be performed at each layer. The last layer is the output layer, where there is one node for each class, prediction, and / or output that the machine learning model can produce.

[0092] Thus, the output may include one or more predictions or inferences (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A to 400C, etc.). For example, the output predictions or inferences may include one or more predictions of deposition drift, film buildup on chamber components, corrosion of chamber components, predicted failure of chamber components, predicted failure of deposition operations, etc. Processing logic determines an error (i.e., a classification error) based on the difference between the output (e.g., prediction or inference) of the machine learning model and the target label associated with the input training data. Processing logic adjusts the weights of one or more nodes in the machine learning model based on the error. An error term or delta may be determined for each node in the artificial neural network. Based on the error, the artificial neural network adjusts one or more of its parameters (weights of one or more inputs to the node) for one or more of its nodes. Parameters may be updated in a back-propagation manner such that nodes at the highest layer are updated first, followed by nodes at the next layer, and so on. The artificial neural network comprises multiple layers of "neurons," where each layer receives input values ​​from neurons at the previous layer. The parameters of each neuron include weights associated with the values ​​received from each of the neurons at the previous layer. Thus, adjusting the parameters may include adjusting the weights assigned to each of the inputs of one or more neurons at one or more layers in the artificial neural network.

[0093] After one or more rounds of training, processing logic may determine whether a stopping criterion has been met. The stopping criterion may be a target level of accuracy, a target number of processed images from the training data set, a target amount of change in a parameter over one or more previous data points, a combination thereof, and / or other criteria. In some embodiments, the stopping criterion is met when at least a minimum number of data points have been processed and at least a threshold accuracy has been achieved. The threshold accuracy may be, for example, 70%, 80%, or 90% accuracy. In some embodiments, the stopping criterion is met if the accuracy of the machine learning model has stopped improving. If the stopping criterion is not met, further training is performed. If the stopping criterion has been met, training may be complete. Once the machine learning model is trained, a retained portion of the training data set may be used to test the model.

[0094] Figure 3 is an example method for generating prediction data 360 (e.g., Figure 1 The system 300 is a block diagram of a system 300 for predicting the residual thickness compensation of a sample using a trained machine learning model (e.g., associated with residual thickness compensation, methods 400A to 400C, etc.) (e.g., Figure 1 model 190) to determine the predicted data 360.

[0095] At block 310, the system 300 (e.g., Figure 1 The prediction system 110) uses historical data (e.g., Figure 1 The historical property data 344 and / or historical performance data 354 of the model 190 are performed data partitioning (e.g., via Figure 1 170 ) to generate a training set 302, a validation set 304, and a test set 306 (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A to 400C, etc.). In some examples, the training set is 60% of the historical data, the validation set is 20% of the historical data, and the test set is 20% of the historical data. The system 300 generates multiple sets of features for each of the training set, validation set, and test set. In some examples, if the historical data includes data from 20 sensors (e.g., Figure 1 For example, if the training set consists of features derived from sensors 126 of manufacturing equipment and / or metrology equipment and 100 products (e.g., each product corresponding to property data from 20 sensors), the first set of features is sensors 1 to 10, the second set of features is sensors 11 to 20, the training set is products 1 to 60, the validation set is products 61 to 80, and the test set is products 81 to 100. In this example, the first set of features for the training set would be the parameters of sensors 1 to 10 from products 1 to 60.

[0096] At block 312, the system 300 performs model training using the training set 302 (e.g., via Figure 1 4. In some embodiments, the system 300 trains multiple models using multiple sets of features from the training set 302 (e.g., a first set of features from the training set 302, a second set of features from the training set 302, etc.). For example, the system 300 trains machine learning models to generate a first trained machine learning model using a first set of features from the training set (e.g., property data from sensors 1 to 10 for products 1 to 60) and a second trained machine learning model using a second set of features from the training set (e.g., property data from sensors 11 to 20 for products 1 to 60). In some embodiments, the first trained machine learning model and the second trained machine learning model are combined to generate a third trained machine learning model (e.g., in some embodiments, the third trained machine learning model is a better predictor than the first trained machine learning model or the second trained machine learning model itself). In some embodiments, the sets of features are used to compare model overlap (e.g., the first set of features is from property data from sensors 1 to 15, and the second set of features is from property data from sensors 5 to 20). In some embodiments, hundreds of models are generated, including models with various feature arrangements and combinations of models.

[0097] At block 314, the system 300 performs model validation using the validation set 304 (e.g., via Figure 1Validation engine 184 of the validation set 304). The system 300 validates each of the trained models (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A-400C, etc.) using a corresponding set of features from the validation set 304. For example, the system 300 validates a first trained machine learning model using a first set of features from the validation set (e.g., parameters from sensors 1-10 of products 61-80) and validates a second trained machine learning model using a second set of features from the validation set (e.g., parameters from sensors 11-20 of products 61-80). In some embodiments, the system 300 validates hundreds of models generated at block 312 (e.g., models with various feature arrangements, combinations of models, etc.). At block 314, the system 300 determines the accuracy of each of the one or more trained models (e.g., via model validation), and determines whether one or more of the trained models has an accuracy that satisfies a threshold accuracy. In response to determining that none of the trained models have an accuracy that satisfies the threshold accuracy, the process returns to block 312, where the system 300 performs model training using a different set of features from the training set. In response to determining that one or more of the trained models have an accuracy that satisfies the threshold accuracy, the process continues to block 316. The system 300 discards trained machine learning models that have an accuracy (e.g., based on a validation set) that is lower than the threshold accuracy.

[0098] At block 316, the system 300 performs model selection (e.g., via Figure 1 The selection engine 185 of the system 300 determines which of the one or more trained models that meet the threshold accuracy has the highest accuracy (e.g., the selected model 308, based on the validation of block 314). In response to determining that two or more of the trained models that meet the threshold accuracy have the same accuracy, the process returns to block 312, where the system 300 performs model training using a further refined training set corresponding to the further refined set of features to determine the trained model with the highest accuracy.

[0099] At block 318, the system 300 performs model testing using the test set 306 (e.g., via Figure 1306 ). In response to the accuracy of the selected model 308 not meeting the threshold accuracy (e.g., the selected model 308 is overfitted to the training set 302 and / or the validation set 304 and is not applicable to other data sets, such as the test set 306), the process continues to block 312, where the system 300 performs model training (e.g., retraining) using a different training set corresponding to a different set of features (e.g., property data from different sensors). In response to determining that the selected model 308 has an accuracy that meets the threshold accuracy based on the test set 306, the process continues to block 320. At least in box 312, the model learns patterns in the historical data to make predictions, and in box 318, the system 300 applies the model to the remaining data (e.g., the test set 306) to test the predictions (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate manufacturing, methods 400A to 400C, etc.).

[0100] At block 320, the system 300 receives current property data 346 (eg, Figure 1 ) and determining (e.g., extracting) predicted data 360 (e.g., Figure 1 The current property data 346 may be used to perform corrective actions (e.g., updating a process recipe, modifying deposition operating parameters, processing a substrate based on the updated recipe, etc.) on the adjustment of film deposition parameters based on the residual thickness during substrate fabrication. In some embodiments, the current property data 346 may correspond to the same type of features in the historical property data 344. In some embodiments, the current property data 346 may correspond to the same type of features as a subset of these types of features in the historical property data 344 used to train the selected model 308 (e.g., associated with adjustment of film deposition parameters based on the residual thickness during substrate fabrication, methods 400A-400C, etc.).

[0101] In some embodiments, current data is received. In some embodiments, the current data includes current performance data 356 (e.g., Figure 1 In some embodiments, the current performance data 156 of the present invention and / or the current property data 346 (e.g., associated with adjusting film deposition parameters based on residual thickness during substrate fabrication, methods 400A-400C, etc.) are obtained from metrology equipment (e.g., Figure 1In some embodiments, the model is retrained based on the current data. In some embodiments, the new model is trained based on the current performance data 356 and the current property data 346.

[0102] In some embodiments, one or more of blocks 310 through 320 are performed in various orders and / or with other operations not presented herein. In some embodiments, one or more of blocks 310 through 320 are not performed. For example, in some embodiments, one or more of the data partitioning of block 310, the model validation of block 314, the model selection of block 316, and / or the model testing of block 318 are not performed.

[0103] Figures 4A to 4C is a flow chart of methods 400A to 400C associated with residual thickness compensation (e.g., residual-based adjustment of film deposition parameters during substrate fabrication) according to certain embodiments. In some embodiments, methods 400A to 400C are performed by processing logic comprising hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing devices, etc.), software (e.g., instructions executed on a processing device, a general-purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. In one implementation, method 400A may be executed by a computer system (e.g., Figure 1 In some embodiments, the method 400A is performed by a computer system architecture 100 (e.g., a computer system architecture 100). In other or similar implementations, one or more operations of method 400A may be performed by one or more other machines not depicted in the figure. In some embodiments, methods 400A to 400C are performed at least in part by prediction system 110. In some embodiments, method 400A is performed by client device 120 (e.g., corrective action component 122) and / or prediction system 110 (e.g., prediction component). In some embodiments, method 400B is performed by server machine 180 (e.g., training engine 182, etc.). In some embodiments, method 400C is performed by prediction server 112 (e.g., prediction component 114) and / or client device 120 (e.g., corrective action component 122). In some embodiments, a non-transitory storage medium stores instructions that, when executed by a processing device (e.g., of prediction system 110, server machine 180, prediction server 112, client device 120, etc.), cause the processing device to perform one or more of methods 400A to 400C.

[0104] To simplify explanation, methods 400A to 400C are depicted and described as a series of operations. However, operations according to the present disclosure may occur in various orders and / or concurrently with other operations not presented and described herein. Furthermore, in some embodiments, not all illustrated operations are performed to implement methods 400A to 400C according to the disclosed subject matter. Additionally, those skilled in the art will understand and appreciate that methods 400A to 400C may alternatively be represented as a series of related states via state diagrams or events.

[0105] Figure 4A is a flow chart of a method associated with adjusting film deposition parameters based on deposited thickness during substrate fabrication in accordance with aspects of the present disclosure.

[0106] refer to Figure 4A In some embodiments, at block 401, processing logic implementing method 400A identifies a material associated with a substrate processing operation of a recipe. In some embodiments, the identified material can be reused across cycles, and a cycle can have any number of materials in any order. In some embodiments, deposition drift compensation is based on the identified material and the substrate processing operation associated with the material. In some embodiments, the processing logic can receive user input identifying the material associated with the substrate processing operation of the recipe. In other embodiments, the processing logic can automatically select the material associated with the substrate processing operation of the recipe. For example, the processing logic can identify the material associated with the substrate processing operation of the recipe for which at least one process run has concluded and a material thickness value has been determined. In yet another embodiment, the processing logic can identify the material associated with the substrate processing operation of the recipe based on a currently executing process recipe. For example, the processing logic can perform a deposition process on a substrate according to the process recipe. The deposition process can be performed in one or more processing chambers. The process recipe can include one or more deposition parameters for the deposition process. For example, the deposition parameters may include the deposition time for each layer and / or cycle of the process recipe, the temperature setting of the processing chamber, the pressure setting of the processing chamber, the flow rate setting of the precursors of the materials included in the film deposited on the substrate surface, the height of the showerhead, etc. In some embodiments, the deposition parameters for one substrate processing area may be different from the deposition parameters for another substrate processing area. The deposition process may deposit multiple layers on the substrate. For example, the deposition process may deposit alternating layers of oxide and nitride layers, alternating layers of oxide and polysilicon, etc.

[0107] At block 402, processing logic determines an expected total residual thickness value after a substrate processing operation. For example, the expected total residual thickness value may include a measured thickness of the residual film, a measured thickness of one or more layers of the residual film, and / or a measured thickness of one or more cycles of the residual film, etc. The expected total residual thickness value may be measured using metrology equipment 128. In some embodiments, a thickness profile is retrieved from data repository 140. In some embodiments, the expected total residual thickness may be derived from a process recipe. For example, a process recipe may be designed to deposit a 200 angstrom film. In this case, the expected total residual thickness value may be 200 angstroms. In some embodiments, the expected total residual thickness value may be measured using sensor 126 to measure deposition on the walls of the processing chamber. In some embodiments, the expected total residual thickness may be determined based on at least one parameter of the deposition operation (e.g., deposition time, deposition rate, etc.).

[0108] In some embodiments, processing logic may determine property data (e.g., an expected total residual thickness value) using sensors that collect property data (e.g., by measuring the thickness of the deposition residue on the walls of the processing chamber). In some embodiments, processing logic may identify property data based on an expected thickness added to the substrate (e.g., thickness parameters of a deposition operation, measured thickness from a previous run of the same deposition operation, etc.). In some embodiments, processing logic may identify property data based on the material deposited and the amount of time the material was deposited (e.g., process recipe, deposition operation parameters, etc.). In some embodiments, processing logic may identify property data based on sensor data and / or metrology data (e.g., a total residual thickness value, a material thickness value, an RF power value for a substrate processing operation, a gap value for a substrate processing operation, a gas flow value for a substrate processing operation, or a chamber pressure value for a substrate processing operation, etc.) of a substrate undergoing a substrate processing operation (e.g., a deposition operation).

[0109] At block 403 , processing logic determines an expected material thickness value for a material associated with a substrate processing operation based on the expected total residual thickness value.

[0110] In some embodiments, processing logic may use one or more formulas or mathematical models to determine an expected material thickness value associated with a substrate processing operation. In some embodiments, a relationship may exist between the expected total residual thickness and the expected material thickness (e.g., based on the identified material associated with the substrate processing operation of the recipe). In some embodiments, the relationship may be a mathematical relationship, an equation, a function, etc. In some embodiments, the relationship may be derived using at least the expected material thickness (e.g., a data point representing the material thickness after a deposition operation), the expected total residual thickness (e.g., a data point representing the wall residual thickness, contamination, etc.), and the deposition time (a data point representing the duration of the deposition operation in the process recipe). In some embodiments, the input to the equation or function is the expected total residual thickness, and the output of the equation is the expected material thickness. In some embodiments, processing logic may appropriately scale the deposition operation time to compensate for drift.

[0111] In some embodiments, a first relationship associated with a material may correspond to a first substrate processing area (e.g., a slot), and a second relationship associated with the same material may correspond to a second substrate processing area (e.g., relationships based on the same material may be specific to a particular slot).

[0112] In some embodiments, the processing logic may generate an equation (e.g., polynomial, linear, logarithmic, etc.) associated with the identified material. In some embodiments, the equation may be reused across loops. In some embodiments, there may be a material-based equation specific to each substrate processing region (e.g., slot). For example, a third-order polynomial equation may be used, and the third-order polynomial equation may be expressed as: y = ax 3 +bx 2 +cx+d, where (x, y) are coordinates and a, b, c, and d are constants. For example, a polynomial equation of third order is used, and polynomial equations (of any order), linear, logarithmic, exponential, and the like may be used. In some embodiments, the equation may be generated using a set of predicted total residual thickness values ​​(e.g., predicted chamber wall residual thickness values) as the x-variable (e.g., chamber wall residual thickness after one cycle, chamber wall residual thickness after two cycles, chamber wall residual thickness after three cycles, etc.) and predicted material thickness values ​​as the y-coordinate (e.g., predicted deposition thickness at different chamber wall residual thicknesses). The (x, y) values ​​may be retrieved from the predicted profile. Using the set of (x, y) coordinates, the constants of the equation are determined. In some embodiments, the total residual thickness (e.g., chamber wall residual thickness) is the quaternization thickness plus the deposition thickness, where the deposition thickness is equal to the total thickness of all previous cycles. The quaternization thickness may include a layer (e.g., a silicon oxide layer) above the chamber walls before the substrate is introduced into the chamber for processing. The deposited quaternary layer reduces the likelihood that contaminants will interfere with subsequent processing steps.

[0113] In some embodiments, processing logic may use the equation and the total residual thickness to determine the (y) value for a particular cycle. Specifically, processing logic may receive input (e.g., based on user input, automatic input, etc.) indicating a layer or cycle of a deposition process. Processing logic may then input the actual material thickness of the cycle or layer (obtained from the thickness profile) into the equation to calculate (y) for the cycle or layer.

[0114] In some embodiments, the expected material thickness value may include an expected thickness of the film, an expected thickness of one or more layers of the film, and / or an expected thickness of one or more cycles of the film, etc. In some embodiments, the expected material thickness value is retrieved from the data repository 140. In some embodiments, the expected material thickness value may be the difference between the expected material thickness value and the actual material thickness value. In some embodiments, determining the expected material thickness value may include providing an expected total residual thickness value as an input to a trained machine learning model. In some embodiments, determining the expected material thickness value may include receiving an output associated with predicted data from the trained machine learning model, wherein the expected material thickness value is associated with the predicted data. In some embodiments, the trained machine learning model may be trained with data inputs including historical total residual thickness values ​​and a target output of historical material thickness values.

[0115] At block 404, processing logic updates the recipe based on the material and the expected material thickness value of the material to generate an updated recipe (e.g., having updated deposition operation parameters, such as a time value (for deposition), an updated time value, an RF power for the substrate processing operation, a spacing value for the substrate processing operation, a gas flow rate value for the substrate processing operation, a chamber pressure value for the substrate processing operation, a temperature of the chamber, etc.). In some embodiments, updating the recipe may include determining an updated time value associated with the substrate processing operation. In some embodiments, determining the updated time value associated with the substrate processing operation may be based on the time value associated with the substrate processing operation, the expected total residual thickness value, and the expected material thickness value. In some embodiments, updating the recipe may include determining at least one of an updated RF power for the substrate processing operation, an updated spacing value for the substrate processing operation, an updated gas flow rate value for the substrate processing operation, or an updated chamber pressure value for the substrate processing operation.

[0116] In some embodiments, the updated recipe may include one or more corrective actions (e.g., updating deposition operation parameters, updating the process recipe, etc.) to be applied to parameters of the process recipe (e.g., associated with the processing chamber during one or more operations of the process recipe). Specifically, the updated recipe may include adjustments to the deposition time for each of one or more layers and / or one or more cycles, a temperature setting for the processing chamber, a pressure setting for the processing chamber, a flow rate setting for a precursor of a material included in a film deposited on the substrate surface, a power supplied to the processing chamber, a ratio of two or more settings, and the like. For example, the updated recipe may include a deposition time adjustment for a cycle of the process recipe. In some embodiments, the updated recipe may include a set of parameter adjustments for each layer and / or cycle of the process recipe. For example, the updated recipe may include a deposition time adjustment for the first cycle, the second cycle, the third cycle, and so on, up to the final cycle. Each adjustment may be applied to the corresponding deposition step to adjust the thickness of one or more cycles or layers so that the film stack thickness is the same as the expected film thickness indicated by the expected total residual thickness. For example, the estimated total residual thickness may include the measured thickness of the film, the measured thickness of one or more layers of the film, and / or the measured thickness of one or more cycles of the film, etc. The thickness profile may be measured using metrology equipment 128 and / or one or more sensors 126. In some embodiments, the thickness profile is retrieved from data repository 140.

[0117] For example, if the expected film thickness after cycle 39 (e.g., the expected total residual thickness) is a first value (e.g., 30,000 nm), the expected film thickness after cycle 40 is a second value (e.g., 30,500 nm), and the actual film thickness (e.g., the actual material thickness) during the deposition run and after cycle 39 is a third value (e.g., 30,050), the updated recipe may indicate a correction to the deposition time of cycle 40 (e.g., reducing the deposition time of cycle 40 by a certain period of time) such that the actual film thickness after cycle 40 is equal to the expected film thickness (e.g., 30,500 nm).

[0118] In some embodiments, processing logic may generate an updated recipe using one or more formulas or mathematical models associated with materials associated with substrate processing operations of the recipe. For example, processing logic may generate a curve fit model using data values ​​from the expected total residual thickness and / or the expected material thickness, and then use the curve fit model to determine an offset time value for a particular step during the current deposition process.

[0119] For example, in some embodiments, a mathematical equation (eg, a polynomial equation having a third order, and the polynomial equation having a third order is expressed as: y=ax 3 +bx 2+cx+d, where (x, y) are coordinates and a, b, c, and d are constants) can be used to update a recipe (e.g., a recipe for a substrate processing area). For example, in some embodiments, the processing logic may use an equation and the total residual thickness to determine the (y) value for a particular cycle. Specifically, the processing logic may receive an input indicating a layer or cycle of a deposition process (e.g., based on user input, automatic input, etc.). The processing logic may then input the actual material thickness of the cycle or layer (obtained from the thickness profile) into the equation to calculate the (y) for the cycle or layer. In some embodiments, the processing logic generates an updated recipe based on the calculated (y) value. In some embodiments, the updated recipe may be generated based on the following equation. For example, updated recipe = (y[first cycle] / y[current number of cycles])*t 步骤 , where t 步骤 is the estimated time for the selected cycle. Although curve fitting methods are discussed for determining the correction profile (eg, updated recipe), other methods, formulas, and models may be used to generate the correction profile, including but not limited to regression analysis, least squares methods, and the like.

[0120] In some embodiments, processing logic may use a machine learning model (eg, machine learning model 190 ) or use an inference engine to generate an updated recipe.

[0121] At block 405, processing logic processes the substrate based on the updated recipe. For example, processing logic may deposit a first set of film layers on the substrate (e.g., perform a first set of cycles of substrate processing operations), determine an expected total residual thickness value for the deposited film, generate an updated recipe to correct any faults detected during deposition of the first set of film layers, apply the updated recipe to the process recipe, and deposit a second set of film layers on the substrate (e.g., perform a second set of cycles of substrate processing operations). Thus, the deposition process recipe may be adjusted in real time or near real time (e.g., substantially real time). This process may be repeated for each deposition step of the process recipe.

[0122] In some embodiments, the updated recipe of block 405 may still cause drift in the deposition of layers in the processing chamber and on the substrate. In some embodiments, after block 405, performance data associated with the updated recipe may be identified. In some embodiments, portions of method 400A may be updated based on the performance data (e.g., updating a predicted total residual thickness associated with the substrate processing operations of the recipe, updating a predicted material thickness value, updating deposition operation parameters, etc.). In some embodiments, the updated method 400A may be repeated to more accurately perform the substrate processing operations of the recipe. In some embodiments, processing logic may determine whether the drift (e.g., the material thickness value) meets a first threshold (e.g., is greater than a particular value), and may repeat method 400A until the drift meets a second threshold (e.g., is less than the particular value). Figure 4B is a method for training a machine learning model (e.g., Figure 1 The model 190) is used to determine prediction data associated with residual thickness compensation (e.g., Figure 1 Flowchart of a method for predicting data 160).

[0123] refer to Figure 4B At block 410 of method 400B, processing logic identifies historical property data for the substrate (e.g., historical total residual thickness values, historical materials associated with historical substrate processing operations of historical recipes, historical material thickness values, historical property data 144, etc.). The historical property data may include data from historical recipes, historical deposition operation parameters, historical substrates, historical processing chamber property data, etc.

[0124] In some embodiments, at block 412, processing logic identifies historical performance data (e.g., historical material thickness values, differences between actual material thickness values ​​and expected material thickness values, differences between actual total residual thickness values ​​and expected total residual thickness values, historical performance data) of historical deposition operating parameters, process recipes, substrates, processing chambers, and the like (e.g., historical performance data of substrates that meet a threshold, substrates that do not meet a threshold, recipes that meet a threshold, recipes that do not meet a threshold, etc.) Figure 1 The historical performance data 154 may include historical material thickness values ​​(e.g., the difference between an actual material thickness value and an expected material thickness value, the difference between an actual total residual thickness value and an expected total residual thickness value, etc.) and / or data from historical processing chambers (e.g., the accuracy of deposition operations, whether an expected total residual thickness value was met (e.g., after a deposition operation), etc.). The performance data (including historical property data) may include sensor data and / or metrology data (e.g., total residual thickness value, material thickness value, RF power value, spacing value, gas flow value, or chamber pressure value, etc.) or user input indicating the performance of a substrate in meeting specific parameters or achieving a specific performance level (e.g., the ability to pass a probe test that measures voltage). The performance data, including historical property data, may include sensor data and / or metrology data or user input indicating the performance of a processing chamber in meeting specific parameters or achieving a specific performance level (e.g., the ability to pass a deposition drift test). At least a portion of the historical property data and historical performance data may be associated with a new substrate processing equipment piece (e.g., for benchmarking). At least a portion of the historical property data and the historical performance data may be associated with a manufactured substrate.At least a portion of the historical property data and the historical performance data may be associated with a processing chamber.

[0125] At box 414, processing logic trains the machine learning model using data input including historical property data 144 and / or target output including historical performance data 154 to generate a trained machine learning model.

[0126] In some embodiments, the historical property data belongs to a historical substrate or processing chamber, and / or the historical performance data corresponds to a historical substrate or processing chamber. In some embodiments, the historical property data corresponds to a deposition operation or process recipe or corresponds to a substrate or processing chamber that has undergone a deposition operation or process recipe. In some embodiments, the historical property data includes historical measurements of historical substrates or processing chambers, and / or the historical performance data corresponds to a historical substrate or processing chamber. Historical performance data may be associated with substrate quality (such as metrology data of the substrate, substrate throughput, substrate defects, etc.). Historical performance data may be associated with the quality of a process recipe (e.g., a deposition operation) (such as the consistency of an actual total residual thickness value with an expected total residual thickness value). Historical performance data may be associated with the quality of a process recipe or deposition operation parameters (such as the ability to accurately deposit each layer and / or film, consistency with the expected total residual thickness of a deposition operation). Historical performance data may be associated with the quality of substrate processing equipment parts (such as test data, metrology data of the substrate, failure time of the substrate, etc.).

[0127] At box 414, processing logic trains the machine learning model using data inputs including historical property data 144 (e.g., historical total residual thickness values) and / or target outputs including historical performance data 154 (e.g., historical material thickness values) to generate a trained machine learning model.

[0128] In some embodiments, the historical property data pertains to a historical substrate or processing chamber, and / or the historical performance data corresponds to a historical substrate or processing chamber. In some embodiments, the historical property data comprises historical metrology of a historical substrate or processing chamber, and / or the historical performance data corresponds to a historical substrate or processing chamber. The historical performance data may be associated with substrate quality (e.g., metrology data of the substrate (e.g., total residual thickness value, material thickness value, etc.), substrate throughput, substrate defects, etc.). The historical performance data may be associated with the quality of a process recipe or deposition operation (e.g., the ability to accurately deposit an expected total residual thickness and / or an expected material thickness). The historical performance data may be associated with the quality of substrate processing equipment parts (e.g., test data, metrology data of the substrate, failure time of the substrate, etc.).

[0129] At block 414, the machine learning model may be trained using the historical property data 144 and / or the target output including the historical performance data 154 to generate a trained machine learning model that is configured to update a recipe, generate an updated recipe, and / or cause a corrective action to be taken (e.g., cause a substrate to be processed based on the updated recipe) based on the property data. In some embodiments, the trained machine learning model may be configured to generate an updated recipe based on the property data 142 (e.g., Figure 4A The estimated total residual thickness value of block 402, Figure 4A The predicted material thickness value of block 403 of FIG404 is used to predict performance data 152 (e.g., performance data, an updated process recipe, updated deposition operating parameters, performance data of a substrate processed with the updated process recipe, performance data of a substrate processed with the updated deposition operating parameters, etc.). In response to the predicted performance data meeting a threshold (e.g., deposition drift exceeding a particular value), the processing logic may cause corrective action to occur (e.g., updating deposition operating parameters, updating the process recipe, processing the substrate based on the updated recipe, etc.). In response to the predicted performance data not meeting the threshold, the process logic may cause corrective action not to occur (e.g., leaving the process recipe unchanged, leaving the deposition operating parameters unchanged, etc.).

[0130] In some embodiments, the deposition operation parameters may include time values, updated time values, recipes, updated recipes, RF power for substrate processing operations, spacing values ​​for substrate processing operations, gas flow values ​​for substrate processing operations, chamber pressure values ​​for substrate processing operations, updated RF power for substrate processing operations, updated spacing values ​​for substrate processing operations, updated gas flow values ​​for substrate processing operations, updated chamber pressure values ​​for substrate processing operations, etc.

[0131] Figure 4C is a method for using a trained machine learning model associated with adjusting film deposition parameters based on residual thickness during substrate fabrication (e.g., Figure 1 Model 190) and method 400C for performing corrective action.

[0132] refer to Figure 4C At block 420 of method 400C, processing logic identifies property data. In some embodiments, the property data of block 420 includes an expected total residual thickness value, an expected material thickness value, etc. In some embodiments, block 420 is similar to Figure 4A 401 and 402 of FIG.

[0133] At block 422, processing logic provides the property data as data input to a trained machine learning model (e.g., via Figure 4BIn some embodiments, the trained machine learning model may be associated with a variable relationship between the expected total residual thickness and the expected material thickness.

[0134] At block 424 , processing logic receives output from the trained machine learning model associated with the prediction data, wherein the updated recipe is based on the prediction data.

[0135] At block 426 , processing logic causes corrective action to be performed based on the predicted data.

[0136] In some embodiments, Figure 4A Block 403 includes training a machine learning model to determine an expected material thickness value for a material associated with a substrate processing operation based on the expected total residual thickness value. In some embodiments, Figure 4A Block 403 includes determining, using the trained machine learning model, an expected material thickness value for a material associated with the substrate processing operation based on the expected total residual thickness value.

[0137] In some embodiments, Figure 4A Block 404 includes training a machine learning model to update a recipe based on the material and the expected material thickness value of the material to generate an updated recipe. In some embodiments, Figure 4A Block 404 includes using the machine learning model to update the recipe based on the material and the expected material thickness value of the material to generate an updated recipe.

[0138] In some embodiments, the property data 142 is an expected total residual thickness value (e.g., an expected deposition thickness associated with a deposition process), and the trained machine learning model of box 422 is trained using data inputs including historical expected total residual thickness values ​​and / or historical actual total residual thickness values ​​and target outputs including historical performance data 154 (e.g., actual total residual thickness values, actual material thickness values, etc.).

[0139] In some embodiments, the property data 142 is cumulative thickness property data (e.g., an expected total residual thickness value, an expected material thickness value, etc.), and the trained machine learning model of block 422 is trained using data inputs including historical deposition thickness property data and target outputs including historical performance data 154, the historical performance data including historical deposition thickness property data for historical substrates or historical deposition operations. The predicted data 160 of block 424 can be associated with predicted performance data (e.g., performance data for a substrate or performance data for a deposition recipe or operation) based on the property data. In response to the predicted performance data meeting a threshold (e.g., deposition drift exceeding a certain level), the processing logic can cause corrective action to occur (e.g., updating a process recipe, updating deposition parameters, etc.). In response to the substrate not meeting the threshold, the processing logic can cause the corrective action to not occur (e.g., leaving the process recipe unchanged, leaving the deposition parameters unchanged, etc.).

[0140] Figure 5 is a block diagram of an example computer system 500 according to some embodiments. In some embodiments, computer system 500 is one or more of client device 120, prediction system 110, server machine 170, server machine 180, prediction server 112, etc.

[0141] In some embodiments, the computer system 500 is connected to other computer systems (e.g., via a network, such as a local area network (LAN), an intranet, an extranet, or the Internet). In some embodiments, the computer system 500 operates as a server or client computer in a client-server environment, or as a peer computer in a peer-to-peer or distributed network environment. In some embodiments, the computer system 500 is provided by a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a cellular phone, a web device, a server, a network router, a switch, or a bridge, or any device capable of executing a set (sequential or other) of instructions specifying the actions to be taken by the device. In addition, the term "computer" should include any computer collection that executes a set (or multiple sets) of instructions, alone or in combination, to perform any one or more of the methods described herein.

[0142] On the other hand, the computer system 500 includes a processing device 502, a volatile memory 504 (e.g., a random access memory (RAM)), a non-volatile memory 506 (e.g., a read-only memory (ROM) or an electrically erasable programmable ROM (EEPROM)), and a data storage device 518, which communicate with each other via a bus 508.

[0143] In some embodiments, the processing device 502 is provided by one or more processors, such as a general-purpose processor (such as, for example, a Complex Instruction Set Computing (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, a microprocessor that implements other types of instruction sets, or a microprocessor that implements a combination of various types of instruction sets) or a special-purpose processor (for example, an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), or a network processor).

[0144] In some embodiments, the computer system 500 further includes a network interface device 522 (e.g., coupled to the network 574). In some embodiments, the computer system 500 further includes a video display unit 510 (e.g., a liquid crystal display (LCD)), an alphanumeric input device 512 (e.g., a keyboard), a cursor control device 514 (e.g., a mouse), and a signal generating device 520.

[0145] In some implementations, the data storage device 518 includes a non-transitory computer-readable storage medium 524 having stored thereon instructions 526 encoding any one or more of the methods or functions described herein, including instructions for encoding Figure 1 Components (eg, corrective action component 122, prediction component 114, etc.) and instructions for implementing the methods described herein (eg, one or more of methods 400A to 400C).

[0146] In some embodiments, the instructions 526 also reside, in whole or in part, within the volatile memory 504 and / or within the processing device 502 during execution thereof by the computer system 500, and thus, in some embodiments, the volatile memory 504 and the processing device 502 also constitute machine-readable storage media.

[0147] Although computer-readable storage medium 524 is shown as a single medium in the illustrative example, the term "computer-readable storage medium" shall include a single medium or multiple media (e.g., a centralized or distributed database and / or associated caches and servers) that store one or more sets of executable instructions. The term "computer-readable storage medium" shall also include any tangible medium that can store or encode a set of instructions for execution by a computer, which instructions cause the computer to perform any one or more of the methodologies described herein. The term "computer-readable storage medium" shall include, but is not limited to, solid-state memories, optical media, and magnetic media.

[0148] The methods, components, and features described herein may be implemented by discrete hardware components or may be integrated into the functionality of other hardware components, such as application specific integrated circuits (ASICs), FPGAs, DSPs, or similar devices. Additionally, these methods, components, and features may be implemented by firmware modules or functional circuitry within a hardware device. Additionally, these methods, components, and features may be implemented in any combination of hardware elements and computer program components or in a computer program.

[0149] Unless otherwise specifically stated, terms such as "identify," "determine," "update," "cause," "provide," "receive," "execute," "obtain," "access," "add," "use," "train," and the like refer to actions and processes performed or implemented by a computer system that manipulate data represented as physical (electronic) quantities within computer system registers and memories and transform them into other data similarly represented as physical quantities within computer system memories or registers or other such information storage, transmission, or display devices. Furthermore, the terms "first," "second," "third," "fourth," and the like as used herein are intended as labels to distinguish between different elements and do not have ordinal meanings based on their numerical designations.

[0150] The examples described herein also relate to apparatus for performing the methods described herein. The apparatus may be specially constructed to perform the methods described herein, or it may comprise a general-purpose computer system selectively programmed by a computer program stored in the computer system. Such a computer program may be stored in a computer-readable tangible storage medium.

[0151] The methods and illustrative examples described herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may be used according to the teachings described herein, or it may prove convenient to construct more specialized apparatus to perform the methods described herein and / or each of their individual functions, routines, subroutines, or operations. Examples of the architecture of a variety of these systems are set forth in the foregoing description.

[0152] The above description is intended to be illustrative and not restrictive. Although the present disclosure has been described with reference to specific illustrative examples and implementations, it will be appreciated that the present disclosure is not limited to the described examples and implementations. The scope of the present disclosure should be determined with reference to the full scope of the appended claims and their equivalents.

Claims

1. A method comprising: identifying materials associated with the substrate processing operations of the recipe; determining an expected total residual thickness value after the substrate processing operation; determining an expected material thickness value for the material associated with the substrate processing operation based on the expected total residual thickness value; updating the recipe based on the material and the expected material thickness value of the material to generate an updated recipe; as well as The substrate is processed based on the updated recipe. 2 . The method of claim 1 , wherein the predicted material thickness value is a difference between a desired material thickness value and an actual material thickness value.

3. The method of claim 1, wherein said updating said recipe comprises determining an updated time value associated with said substrate processing operation.

4. The method of claim 3, wherein the determining the updated time value associated with the substrate processing operation is based on the time value associated with the substrate processing operation, the expected total residual thickness value, and the expected material thickness value.

5. The method of claim 1 , wherein updating the recipe comprises determining at least one of an updated radio frequency (RF) power for the substrate processing operation, an updated spacing value for the substrate processing operation, an updated gas flow value for the substrate processing operation, or an updated chamber pressure value for the substrate processing operation.

6. The method of claim 1 , wherein said determining said expected material thickness value comprises: providing the estimated total residual thickness value as input to a trained machine learning model; as well as An output associated with predicted data is received from the trained machine learning model, wherein the expected material thickness value is associated with the predicted data.

7. The method of claim 6, wherein the trained machine learning model is trained with data inputs comprising historical total residual thickness values ​​and target outputs of historical material thickness values.

8. A non-transitory computer-readable storage medium storing instructions that, when executed, cause a processing device to perform operations comprising: identifying materials associated with the substrate processing operations of the recipe; determining an expected total residual thickness value after the substrate processing operation; determining an expected material thickness value for the material associated with the substrate processing operation based on the expected total residual thickness value; updating the recipe based on the material and the expected material thickness value of the material to generate an updated recipe; as well as The substrate is processed based on the updated recipe. 9 . The non-transitory computer-readable storage medium of claim 8 , wherein the predicted material thickness value is a difference between an expected material thickness value and an actual material thickness value.

10. The non-transitory computer-readable storage medium of claim 8, wherein the updating the recipe comprises determining an updated time value associated with the substrate processing operation.

11. The non-transitory computer-readable storage medium of claim 10, wherein the determining the updated time value associated with the substrate processing operation is based on the time value associated with the substrate processing operation, the expected total residual thickness value, and the expected material thickness value.

12. The non-transitory computer-readable storage medium of claim 8, wherein the updating the recipe comprises determining at least one of an updated radio frequency (RF) power for the substrate processing operation, an updated spacing value for the substrate processing operation, an updated gas flow value for the substrate processing operation, or an updated chamber pressure value for the substrate processing operation.

13. The non-transitory computer-readable storage medium of claim 8, wherein said determining said expected material thickness value comprises: providing the estimated total residual thickness value as input to a trained machine learning model; as well as An output associated with predicted data is received from the trained machine learning model, wherein the expected material thickness value is associated with the predicted data.

14. The non-transitory computer-readable storage medium of claim 13, wherein the trained machine learning model is trained with data inputs comprising historical total residual thickness values ​​and target outputs of historical material thickness values.

15. A system comprising: Memory; as well as a processing device, the processing device being coupled to the memory, the processing device: identifying materials associated with the substrate processing operations of the recipe; determining an expected total residual thickness value after the substrate processing operation; determining an expected material thickness value for the material associated with the substrate processing operation based on the expected total residual thickness value; updating the recipe based on the material and the expected material thickness value of the material to generate an updated recipe; as well as The substrate is processed based on the updated recipe.

16. The system of claim 15, wherein the predicted material thickness value is a difference between a desired material thickness value and an actual material thickness value.

17. The system of claim 15, wherein to update the recipe, the processing device determines an updated time value associated with the substrate processing operation.

18. The system of claim 17, wherein the processing device determines the updated time value associated with the substrate processing operation based on the time value associated with the substrate processing operation, the expected total residual thickness value, and the expected material thickness value.

19. The system of claim 15, wherein to determine the expected material thickness value, the processing device: providing the predicted total residual thickness value as input to a trained machine learning model; and An output associated with predicted data is received from the trained machine learning model, wherein the expected material thickness value is associated with the predicted data.

20. The system of claim 19, wherein the trained machine learning model is trained with data inputs comprising historical total residual thickness values ​​and target outputs of historical material thickness values.

Images