In-situ chamber cleaning endpoint detection system and method using a computer vision system

Abstract

The present invention relates to an in-situ chamber cleaning endpoint detection system and method using a computer vision system. A system includes a camera mounted outside and adjacent to a window of a processing chamber configured to process semiconductor substrates. The window enables the camera to observe components within the processing chamber. The camera is configured to generate a video signal indicating the status of the component during a process being performed in the processing chamber. The system also includes a controller coupled to the processing chamber. The controller is configured to: control the camera; process the video signal from the camera; determine the status of the component based on the processing of the video signal; and determine whether to terminate the process based on the status of the component.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 62 / 575,190, filed on October 20, 2017. The entire disclosure of the above application is incorporated herein by reference. Technical Field

[0003] The present disclosure relates to substrate processing systems, and more particularly to detection systems and methods for detecting an endpoint of residual film removal using a computer vision system. Background Art

[0004] The background description provided here is for the purpose of generally presenting the context of the present disclosure. Work by the presently designated inventors to the extent described in this background section and in aspects of the description that could not be determined as prior art at the time of filing the application is neither explicitly nor implicitly admitted to be prior art against the present disclosure.

[0005] Substrate processing systems are used to perform processes such as depositing and etching films on substrates such as semiconductor wafers. For example, deposition can be performed using chemical vapor deposition (CVD), atomic layer deposition (ALD), or other deposition processes to deposit conductive films, dielectric films, or other types of films. During deposition, one or more precursor gases can be supplied to the process chamber during one or more process steps. Plasma can be used to initiate chemical reactions.

[0006] After the deposition is performed, the process gases are evacuated and the substrate is removed from the process chamber. As the film is deposited on the substrate, it is also deposited on the components located in the process chamber. Over time, residual film accumulates on these components and needs to be removed to prevent particle contamination, mechanical deformation, and / or substrate defects. Periodically, a chamber cleaning process is performed to remove residual film from the components in the process chamber.

[0007] Currently, fixed time cleaning processes are used. Verification of cleaning is performed manually using a chamber viewing port or using chemical sensors such as infrared absorption detectors and / or residual gas analyzers (RGA). Timing-based cleaning requires manual verification of the process chamber state to be clean. Another limitation of this approach is the inability to predict cleaning times when processes or hardware changes occur. Any modification to the process recipe or chamber configuration can result in different accumulation of residual film and different etch rates for the chamber cleaning process. This results in different cleaning times that require manual characterization.

[0008] Chemical sensors are often expensive and may have limited applications. For example, infrared absorption detectors are often limited to specific gas species. RGAs are also limited to the specific atomic masses that the detector can analyze. Summary of the invention

[0009] A system includes a camera mounted outside and adjacent to a window of a processing chamber configured to process semiconductor substrates. The window enables the camera to observe a component within the processing chamber. The camera is configured to generate a video signal indicating a state of the component during a process being performed within the processing chamber. The system also includes a controller coupled to the processing chamber. The controller is configured to: control the camera; process the video signal from the camera; determine a state of the component based on the processing of the video signal; and determine whether to terminate the process based on the state of the component.

[0010] In other features, in response to the process being a cleaning process performed to remove material deposited on the component by a previously performed process, the controller is configured to: compare a change in a property of a feature of the component observed through multiple frames of the video signal with a predetermined threshold; determine whether the material deposited on the component has been removed based on the comparison; and in response to determining that the material deposited on the component has been removed, terminate the cleaning process.

[0011] In other features, in response to the process being a cleaning process performed to remove material deposited on the component by a previously performed process, the controller is configured to: compare an image captured from the video signal with a predetermined image; determine whether the material deposited on the component has been removed based on the comparison; and in response to determining that the material deposited on the component has been removed, terminate the cleaning process.

[0012] In other features, the controller is configured to: receive data from one or more sensors in the process chamber; and generate a model based on the data received from the one or more sensors and a video signal received from the camera, the video signal indicating a state of the component when the process was previously performed in the process chamber. The controller is also configured to use the model to: process the video signal; determine the state of the component based on the processing of the video signal; and determine whether to terminate the process based on the determined state of the component.

[0013] In other features, the system further comprises an optical filter disposed between the camera and the window. The optical filter is configured to filter one or more wavelengths of light received from the component through the window and output a filtered signal to the camera. The controller is configured to determine the state of the component based on processing the filtered signal using optical interferometry.

[0014] In other features, the controller is configured to determine whether the process is performed on the entire component before terminating the process.

[0015] In other features, the controller is configured to determine whether the process is performed evenly across the component before terminating the process.

[0016] In other features, the controller is configured to determine a rate at which the process is performed at different locations on the component.

[0017] In other features, the component includes a semiconductor substrate, and the process includes a film removal process performed to remove a film from the semiconductor substrate. The controller is configured to determine whether the film is removed across the component before terminating the process.

[0018] In other features, the controller is configured to focus the camera at an edge of the component and determine whether to perform the process at the edge of the component before terminating the process.

[0019] In other features, the component includes a semiconductor substrate, and the process includes a deposition process. The system also includes a first optical filter disposed between the camera and the window. The first optical filter is configured to filter wavelengths of ultraviolet light received from the component through the window. The system also includes a second optical filter disposed between the camera and the window. The second optical filter is configured to filter wavelengths of infrared light received from the component through the window. The controller is configured to: determine a thickness of a material deposited at a plurality of locations on the component based on an output of the first optical filter; determine a temperature at a plurality of locations on the component based on an output of the second optical filter; correlate the determination of the thickness and the determination of the temperature; and determine a uniformity of deposition across the component based on the correlation.

[0020] In other features, the component includes a semiconductor substrate and the process includes a deposition process. The system also includes an optical sensor configured to view a bottom portion of the component. The controller is configured to: focus the camera on a top portion of the component; process images received from the optical sensor and the camera; and determine deposition uniformity across the top portion and across the bottom portion of the component based on the processing of the images.

[0021] In other features, the system further includes a second controller coupled to a second process chamber in which the same process is performed. The second controller is configured to control a second camera associated with the second process chamber. The system further includes a third controller configured to: analyze data from the controller and the second controller; compare a performance of the process in the process chamber with a performance of the process in the second process chamber based on the analyzed data; and determine, based on the comparison, whether the performance of the process in the process chamber matches the performance of the process in the second process chamber.

[0022] In other features, the system further includes a second controller coupled to a second process chamber in which the same process is performed on the same component. The second controller is configured to control a second camera associated with the second process chamber. The system further includes a third controller configured to: analyze data from the controller and the second controller; compare the performance of the process on the component in the process chamber with the performance of the process on the component in the second process chamber based on the analyzed data; and determine whether the performance of the process on the component in the process chamber matches the performance of the process on the component in the second process chamber based on the comparison.

[0023] In other features, in response to the process in the process chamber being completed earlier than the process in the second process chamber, the third controller is configured to terminate the process in the process chamber earlier than the process in the second process chamber.

[0024] In still other features, a method includes controlling a camera mounted outside and adjacent to a window of a processing chamber configured to process a semiconductor substrate. The window enables the camera to observe a component within the processing chamber. The method also includes: using the camera to generate a video signal indicating a state of the component during a process being performed within the processing chamber; determining the state of the component based on the video signal; and determining whether to terminate the process based on the state of the component.

[0025] In other features, the processing includes a cleaning process performed to remove material deposited on the component by a previously performed process, and the method further includes: comparing a change in a property of a feature of the component observed through multiple frames of the video signal to a predetermined threshold; determining whether the material deposited on the component has been removed based on the comparison; and terminating the cleaning process in response to determining that the material deposited on the component has been removed.

[0026] In other features, the processing includes a cleaning process performed to remove material deposited on the component by a previously performed processing, and the method further includes: comparing an image captured from the video signal with a predetermined image; determining whether the material deposited on the component has been removed based on the comparison; and terminating the cleaning process in response to determining that the material deposited on the component has been removed.

[0027] In other features, the method further includes: receiving data from one or more sensors in the process chamber; and generating a model based on the data received from the one or more sensors and a video signal received from the camera. The video signal indicates a state of the component when the process was previously performed in the process chamber. The method further includes using the model to: process the video signal; determine a state of the component based on the processing of the video signal; and determine whether to terminate the process based on the determined state of the component.

[0028] In other features, the method further comprises filtering one or more wavelengths of light received from the component through the window and determining the state of the component based on the filtering using optical interferometry.

[0029] In other features, the method further comprises determining whether the process is performed on the entire component before terminating the process.

[0030] In other features, the method further comprises determining whether the process is performed evenly across the component before terminating the process.

[0031] In other features, the method further comprises determining a rate at which the process is performed at different locations on the component.

[0032] In other features, the component includes a semiconductor substrate, and the process includes a film removal process performed to remove a film from the semiconductor substrate. The method also includes determining whether the film is removed throughout the component before terminating the process.

[0033] In other features, the method further comprises focusing the camera at an edge of the component and determining whether the processing is performed at the edge of the component before terminating the processing.

[0034] In other features, the component includes a semiconductor substrate and the process includes a deposition process. The method further includes: filtering wavelengths of ultraviolet light received from the component through the window; filtering wavelengths of infrared light received from the component through the window; determining a thickness of a deposited material at a plurality of locations on the component based on the filtering of the wavelengths of ultraviolet light; determining a temperature at a plurality of locations on the component based on the filtering of the wavelengths of infrared light; correlating the determination of the thickness with the determination of the temperature; and determining a deposition uniformity across the component based on the correlation.

[0035] In other features, the component includes a semiconductor substrate and the process includes a deposition process. The method also includes observing a bottom of the component using an optical sensor; focusing the camera on a top of the component; processing images received from the optical sensor and the camera; and determining deposition uniformity across the top and across the bottom of the component based on the processing of the images.

[0036] In other features, the method also includes receiving data from a second camera associated with the second processing chamber in which the same process is performed; analyzing data from the processing chamber and the second processing chamber; comparing the performance of the process in the processing chamber with the performance of the process in the second processing chamber based on the analyzed data; and determining whether the performance of the process in the processing chamber matches the performance of the process in the second processing chamber based on the comparison.

[0037] In other features, the method also includes receiving data from a second camera associated with the second processing chamber in which the same process is performed on the same component; analyzing data from the processing chamber and the second processing chamber; comparing the performance of the process on the component in the processing chamber with the performance of the process on the component in the second processing chamber based on the analyzed data; and determining whether the performance of the process on the component in the processing chamber matches the performance of the process on the component in the second processing chamber based on the comparison.

[0038] In other features, in response to the process in the process chamber being completed earlier than the process in the second process chamber, the method further includes terminating the process in the process chamber earlier than the process in the second process chamber.

[0039] Specifically, some aspects of the present invention can be described as follows:

[0040] 1. A system comprising:

[0041] a camera mounted externally and adjacent to a window of a processing chamber configured to process semiconductor substrates, the window enabling the camera to observe components within the processing chamber, the camera being configured to generate video signals indicative of a status of the components during processing being performed within the processing chamber; and

[0042] a controller coupled to the process chamber and configured to:

[0043] controlling the camera;

[0044] processing the video signal from the camera;

[0045] determining a status of the component based on the processing of the video signal; and

[0046] Whether to terminate the process is determined based on the status of the component.

[0047] 2. The system of clause 1, wherein, in response to the process being a cleaning process performed to remove material deposited on the component by a previously performed process, the controller is configured to:

[0048] comparing a change in a property of a feature of the component observed across a plurality of frames of the video signal to a predetermined threshold;

[0049] determining whether the material deposited on the component is removed based on the comparing; and

[0050] In response to determining that the material deposited on the component is removed, the cleaning process is terminated.

[0051] 3. The system of clause 1, wherein, in response to the process being a cleaning process performed to remove material deposited on the component by a previously performed process, the controller is configured to:

[0052] comparing an image captured from the video signal with a predetermined image;

[0053] determining whether the material deposited on the component is removed based on the comparing; and

[0054] In response to determining that the material deposited on the component is removed, the cleaning process is terminated.

[0055] 4. The system of clause 1, wherein the controller is configured to:

[0056] receiving data from one or more sensors in the process chamber;

[0057] generating a model based on the data received from the one or more sensors and a video signal received from the camera, the video signal indicating a state of the component when the process was previously performed in the process chamber; and

[0058] Use this model to:

[0059] processing the video signal;

[0060] determining a status of the component based on the processing of the video signal; and

[0061] Based on the determined status of the component, it is determined whether to terminate the process.

[0062] 5. The system according to clause 1, further comprising:

[0063] an optical filter disposed between the camera and the window, the optical filter being configured to filter one or more wavelengths of light received from the component through the window and output a filtered signal to the camera,

[0064] Wherein the controller is configured to determine the status of the component based on processing the filtered signal using optical interferometry.

[0065] 6. The system of clause 5, wherein the controller is configured to determine whether the process has been performed on the entire component before terminating the process.

[0066] 7. The system of clause 5, wherein the controller is configured to determine whether the process is performed evenly across the component before terminating the process.

[0067] 8. The system of clause 5, wherein the controller is configured to determine a rate at which the processing is performed at different locations on the component.

[0068] 9. A system according to clause 1, wherein the component includes a semiconductor substrate, and the processing includes a film removal processing performed to remove a film from the semiconductor substrate, and wherein the controller is configured to determine whether the film on the entire component is removed before terminating the processing.

[0069] 10. The system of clause 1, wherein the controller is configured to focus the camera at an edge of the component and determine whether to perform the process at the edge of the component before terminating the process.

[0070] 11. The system of clause 1, wherein the component comprises a semiconductor substrate and the process comprises a deposition process, the system further comprising:

[0071] a first optical filter disposed between the camera and the window, the first optical filter configured to filter wavelengths of ultraviolet light received from the component through the window; and

[0072] a second optical filter disposed between the camera and the window, the second optical filter configured to filter wavelengths of infrared light received from the component through the window;

[0073] Wherein, the controller is configured as:

[0074] determining a thickness of deposited material at a plurality of locations on the component based on an output of the first optical filter;

[0075] determining a temperature at a plurality of locations of the component based on an output of the second optical filter;

[0076] associating said determination of said thickness with said determination of said temperature; and

[0077] A deposition uniformity across the component is determined based on the correlation.

[0078] 12. The system of clause 1, wherein the component comprises a semiconductor substrate and the process comprises a deposition process, the system further comprising:

[0079] an optical sensor configured to observe a bottom portion of the component,

[0080] The controller is configured as follows:

[0081] focusing the camera on a top portion of the component;

[0082] processing images received from the optical sensor and the camera; and

[0083] Deposition uniformity across the top portion and across the bottom portion of the component is determined based on the processing of the images.

[0084] 13. The system of clause 1, further comprising:

[0085] a second controller coupled to a second processing chamber in which the same process is performed, the second controller being configured to control a second camera associated with the second processing chamber; and

[0086] A third controller is configured to:

[0087] analyzing data from the controller and the second controller;

[0088] comparing the performance of the process in the process chamber to the performance of the process in the second process chamber based on the analyzed data; and

[0089] Based on the comparison, it is determined whether the performance of the process in the process chamber matches the performance of the process in the second process chamber.

[0090] 14. The system of clause 1, further comprising:

[0091] a second controller coupled to a second processing chamber in which the same process is performed on the same component, the second controller being configured to control a second camera associated with the second processing chamber; and

[0092] A third controller is configured to:

[0093] analyzing data from the controller and the second controller;

[0094] comparing the performance of the process on the component in the process chamber with the performance of the process on the component in the second process chamber based on the analyzed data; and

[0095] Based on the comparison, it is determined whether the performance of the process on the component in the process chamber matches the performance of the process on the component in the second process chamber.

[0096] 15. The system of clause 14, wherein, in response to the processing in the processing chamber being completed earlier than the processing in the second processing chamber, the third controller is configured to terminate the processing in the processing chamber earlier than the processing in the second processing chamber.

[0097] Further scope of applicability of the present disclosure will become apparent from the detailed description, claims and drawings.The detailed description and specific examples are for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] The present invention will be more fully understood from the detailed description and accompanying drawings, in which:

[0099] Figure 1 is a functional block diagram of an example of a substrate processing system according to the present disclosure for generating a video signal and using a computer vision system to detect an endpoint of residual film removal based on the video signal;

[0100] Figure 2 is a diagram for operating according to the present disclosure Figure 1 A flowchart of an example of a method of a substrate processing system;

[0101] Figure 3 is a functional block diagram of an example of a substrate processing system according to the present disclosure for collecting sensor data to generate a learning model that is used by a computer vision system to identify an endpoint for residual film removal;

[0102] Figure 4-5 is a flow chart illustrating an example of a method for generating a learning model to identify an endpoint of residual film removal according to the present disclosure;

[0103] Figure 6 is a flow chart illustrating an example of a method for detecting an endpoint of residual film removal using a learning model according to the present disclosure;

[0104] Fig. 7A is a plan view illustrating an example of a structure located in a process chamber having markings to facilitate endpoint detection by a computer vision system according to the present disclosure;

[0105] Figure 7B is a side view illustrating an example of a structure located in a process chamber having markings to facilitate endpoint detection by a computer vision system according to the present disclosure;

[0106] Figure 8 is a functional block diagram of a system for optimizing processing conditions using a controller connected to multiple processing controllers including a computer vision system according to the present disclosure.

[0107] Fig. 9 is a flow chart illustrating an example of a method for comparing performance of processes performed within a plurality of chambers and matching process chambers based on the performance of the processes in the plurality of chambers according to the present disclosure;

[0108] Fig.10 is a flow chart illustrating an example of a method for matching process chambers at a component level according to the present disclosure;

[0109] Fig.11A is a schematic diagram of a system for determining uniformity of a film deposited on a substrate according to the present disclosure;

[0110] Fig. 11B is a schematic diagram of a system for determining whether a material is uniformly deposited on the top and bottom of a substrate according to the present disclosure;

[0111] Fig.12 is a flow chart illustrating an example of a method for determining whether a process (e.g., cleaning, etching, deposition, etc.) performed on a component has been performed uniformly across the component in accordance with the present disclosure;

[0112] Fig.13is a flow chart illustrating an example of a method for determining whether a process (eg, cleaning, etching, deposition, etc.) performed on a component has been performed to an edge of the component according to the present disclosure;

[0113] Fig.14 is a flow chart illustrating an example of a method for determining uniformity of a film deposited on a substrate according to the present disclosure; and

[0114] Fig.15 is a flow chart illustrating an example of a method for determining whether material is uniformly deposited on both the top and bottom of a substrate according to the present disclosure.

[0115] In the drawings, reference numbers may be reused to identify similar and / or identical elements. DETAILED DESCRIPTION

[0116] The present disclosure relates to systems and methods for detecting the endpoint of residual film removal using a computer vision system. The systems and methods utilize a camera mounted outside a processing chamber adjacent to a window within the processing chamber. The camera operates within a visible and / or infrared wavelength range and generates a video signal (as used herein, the term video signal also includes a series of still images taken at predetermined intervals). The video signal generated by the camera is processed and analyzed by a computer vision system to determine when the endpoint of residual film removal is reached.

[0117] In some examples, the endpoint is identified based on a detected change in a feature (e.g., shape or edge identification, contrast, color, brightness, or other feature) of a component in the process chamber. For example, the computer vision system can compare one or more attributes of one or more features frame by frame, differences over multiple frames, or differences over a rolling window of frames to one or more predetermined thresholds. In other examples, the endpoint is identified when a mark (made on the component below the residual film) is exposed with sufficient clarity.

[0118] In some examples, the computer vision system uses a learning model (such as a neural network or other deep learning model) trained using video signals and a chemical sensor such as an absorption sensor or an RGA sensor. During learning, the learning model identifies one or more attributes that are indicators of an endpoint. Once the learning model is generated, the chemical sensor is no longer required.

[0119] Reference now Figure 1, shows an example of a substrate processing system for performing substrate processing such as deposition. A substrate is arranged in a processing chamber and substrate processing such as deposition is performed. The substrate is removed and one or more additional substrates are processed. Over time, residual film or other material accumulates on components such as sidewalls, substrate supports, gas distribution devices, etc. located within the processing chamber. Cleaning is performed periodically to remove residual film or other material deposited on components in the processing chamber. In one example, nitrogen trifluoride (NF) is used during cleaning. 3 ) remote plasma. Although specific examples of substrate processing systems and processes are shown and described, other substrate processing systems and processes may be used. For example, throughout this disclosure, cleaning is described only as plasma processing. Instead, cleaning may be a chemical process. For example, instead of using plasma for cleaning, ClF 3 Cleaning is performed as described herein. The teachings of the present disclosure also apply when chemical processes are used instead of plasma for cleaning.

[0120] Figure 1 The substrate processing system 110 in FIG. 1 includes a processing chamber 112 that surrounds the other components of the substrate processing system 110 and contains an RF plasma (if used for a particular substrate process). The substrate processing system 110 includes a showerhead 114 or other gas distribution device and a substrate support assembly 116. A substrate 118 is disposed on the substrate support assembly 116. The showerhead 114 introduces and distributes process gases.

[0121] If plasma is used, the plasma used during substrate processing or cleaning can be a direct plasma or a remote plasma. In one example, the RF generation system 130 generates an RF voltage and outputs the RF voltage to the showerhead 114 or the substrate support assembly 116 (the other is DC grounded, AC grounded, or floating). By way of example only, the RF generation system 130 can include an RF voltage generator 132 that generates an RF voltage that is fed to the showerhead 114 or the substrate support assembly 116 by a matching network 134. In some examples, an in-situ plasma is used during substrate processing, while a remote plasma is delivered during cleaning.

[0122] The gas delivery system 140 includes one or more gas sources 142-1, 142-2, ... and 142-N (collectively referred to as gas sources 142), where N is an integer greater than zero. The gas source 142 provides one or more etching gas mixtures, precursor gas mixtures, cleaning gas mixtures, ashing gas mixtures, etc. to the processing chamber 112. Gasified precursors can also be used. The gas source 142 is connected to the manifold 148 through valves 144-1, 144-2, ... and 144-N (collectively referred to as valves 144) and mass flow controllers 146-1, 146-2, ... and 146-N (collectively referred to as mass flow controllers 146). The output of the manifold 148 is fed to the processing chamber 112. By way of example only, the output of the manifold 148 is fed to the showerhead 114.

[0123] The heater 150 may be connected to a heater coil (not shown) disposed in the substrate support assembly 116. The heater 150 may be used to control the temperature of the substrate support assembly 116 and the substrate 118. The valve 160 and the pump 162 may be used to exhaust reactants from the process chamber 112. The controller 170 may be used to control the components of the substrate processing system 110. By way of example only, the controller 170 may be used to control the flow of process gases, monitor process parameters such as temperature, pressure, power, etc., energize and extinguish plasma, remove reactants, etc.

[0124] The processing chamber 112 includes one or more windows 172 located on one or more surfaces of the processing chamber 112. One or more cameras 174 generate video signals of the substrate support assembly 116, the substrate, or other chamber components.

[0125] The controller 170 includes a computer vision system 180 that receives one or more video signals from the camera 174. The computer vision system 180 identifies the endpoint of the residual film removal based thereon. Although a single camera 174 is shown, one or more additional cameras 174 may be used. In some examples, the camera positioning device 182 adjusts the position of the camera 174 to change the field of view of the camera 174. In some examples, the controller 170 may be used to change the position of the camera 174 before, during, or after the endpoint of the residual film removal is detected. In some examples, one or more filters are arranged between the camera 174 and the processing chamber to filter one or more wavelengths or wavelength ranges.

[0126] In some examples, computer vision system 180 determines differences in visual characteristics of the video to identify the endpoint of residual film removal. For example, computer vision system 180 can compare the color, brightness, and / or contrast levels of different frames, differences across multiple frames, or differences across a rolling window of frames.

[0127] In other examples, the computer vision system 180 detects the endpoint of residual film removal by fully identifying a mark on one or more components located in the processing chamber. When the residual film is removed from one or more components during cleaning, characteristics such as the edges, colors, patterns, and / or overall shapes of the marks will become clearer. A correlation function can be used to compare the captured image with a predetermined image. When the calculated correlation value exceeds a predetermined threshold, a detection can be triggered. The endpoint can be determined when a predetermined event occurs, or a predetermined event can be used to start a timer, after which the endpoint can be declared.

[0128] In other examples, a chemical sensor such as an absorption sensor or an RGA sensor and a video signal can be used to train the learning model. Once trained, the learning model detects the endpoint based on the video signal and does not require the use of a chemical sensor, as will be further described below.

[0129] Reference now Figure 2 , a method 200 for detecting an endpoint of residual film removal using a video signal and a computer vision system is shown. At 202, if the substrate needs to be processed, the method continues at 203. At 203, the substrate is loaded into a processing chamber and substrate processing such as deposition is performed.

[0130] At 204, the substrate is removed from the process chamber after performing substrate processing. At 206, the method determines whether the chamber needs cleaning. If 206 is false, the method returns to 202. If 206 is true, the method continues at 208 and supplies a remote plasma (or supplies a gas mixture to the process chamber and energizes the plasma). If an in-situ plasma is used, a filter can be placed between the video camera and the process chamber.

[0131] At 210, a video signal is generated during the cleaning process. At 212, a computer vision system performs video analysis of the video signal. At 214, the computer vision system processes the video signal and detects an endpoint of film removal. At 216, the method determines whether the endpoint has been detected. If 216 is false, the method returns to 210. When 216 is true, the method continues at 218 and stops supplying the remote plasma (or extinguishes the plasma and turns off the plasma gas mixture). The method continues from 218 to 202.

[0132] Reference now Figure 3, a learning model can be generated and used to detect the endpoint of residual film removal. The modeling module 220 generates a learning model to be used by the computer vision system 180. The learning model is generated based on training data including video signals and the output of a chemical sensor 230 such as an absorption or RGA sensor. Although specific locations are shown, the chemical sensor can be arranged in other locations, such as at the discharge. Although the modeling module 220 is shown to be located in the controller 170, the modeling module 220 can be located in another computer (not shown) during model generation, and the model can be sent to the controller 170 and / or stored in the controller 170.

[0133] Modeling module 220 uses multiple sets of training data to identify features in the video signal that are indicators of the endpoint. In some examples, the multiple sets of training data are stored in a local or remote database 224. Once the learning model is generated, the learning model can be loaded into controller 170 (as shown in 228) and chemical sensor 230 is no longer required.

[0134] In some examples, the learning model includes a supervised learning model selected from the group consisting of a linear model, a support vector machine model, a decision tree model, a random forest model, and a Gaussian model. However, the present disclosure is not limited to supervised learning. Additionally or alternatively, unsupervised learning can be used. In other examples, the learning model uses principal component analysis (PCA), neural networking, autoencoding, regression analysis, and / or partial least squares (PLS). For example, the neural network can utilize a convolutional neural network (CNN), a recursive neural network (RNN), reinforcement learning (a reward-based model), and / or other methods.

[0135] Reference now Figure 4-5 , a method for generating a learning model for computer vision to identify the endpoint of residual film removal is shown. A method 400 for generating training data for generating a learning model (used by a computer vision system to detect the endpoint of residual film removal) is shown. At 404, the method determines whether the substrate needs to be processed. If 404 is true, the method continues at 408. At 408, the substrate is loaded into a processing chamber and substrate processing such as deposition is performed.

[0136] At 412, the substrate is removed from the processing chamber after performing the substrate processing. At 416, the method determines whether the chamber needs to be cleaned. If 416 is false, the method returns to 404. If 416 is true, the method supplies a remote plasma or supplies a gas mixture to the processing chamber and excites the plasma at 422. At 430, the method generates and stores a video signal of one or more structures in the processing chamber. At 434, the method generates and stores chemical sensor data generated by a chemical sensor. At 438, the endpoint of the residual film removal is detected using a chemical sensor. At 440, the method determines whether the residual film is removed. If 440 is false, the method returns to 430. If 440 is true, the method continues at 444 and stops supplying the remote plasma (or extinguishes the plasma and turns off the plasma gas mixture).

[0137] exist Figure 5 , which shows the Figure 4 The method 450 converts the training data generated in the above into a learning model. At 454, the method determines whether a sufficient number of training samples have been collected. At 458, the method generates a learning model that associates the video signal with the end point of residual film removal determined by the chemical sensor. At 462, the model is stored in the controller for use in the cleaning process. At 466, the controller uses the learning model based on the video (or a series of still pictures) to identify the end point of residual film removal without using a chemical sensor such as an absorption or RGA sensor.

[0138] Reference now Figure 6 , shows a method 500 that uses a video signal and a computer vision system with a learning model to detect an endpoint of residual film removal. At 502, if the substrate needs to be processed, the method continues at 503. At 503, the substrate is loaded into a processing chamber and substrate processing such as deposition is performed.

[0139] At 504, the substrate is removed from the processing chamber after performing the substrate processing. At 506, the method determines whether the chamber needs to be cleaned. If 506 is false, the method returns to 502. If 506 is true, the method continues at 508 and supplies a remote plasma (or supplies a gas mixture to the processing chamber and energizes the plasma). If an in-situ plasma is used, a filter can be placed between the camera and the processing chamber.

[0140] At 510, a video signal is generated during the cleaning process. At 512, the computer vision system performs video analysis of the video signal using the learning model. At 514, the computer vision system detects an endpoint of film removal using the learning model and the video signal. At 516, the method determines whether the endpoint has been detected. If 516 is false, the method returns to 510. When 516 is true, the method continues at 518 and stops supplying the remote plasma (or extinguishes the plasma and turns off the plasma gas mixture). The method continues from 518 to 502.

[0141] Reference now Figure 7A-7B , examples of markings are shown on structures arranged in a process chamber to facilitate detection of the endpoint of residual film removal. Fig. 7A In the example, structure 480 includes one or more marks 494-1, 494-2, ... and 494-Z, where Z is an integer greater than zero. In some examples, the marks can have different patterns, shades, colors, or other designs to improve accuracy during detection by a computer vision system. In some examples, structure 480 corresponds to the upper surface of a substrate support, such as Fig. 7A In other examples, such as Figure 7B As shown, structure 480 corresponds to the side of the substrate support. It will be appreciated that other components such as gas distribution devices, supports, sidewalls, etc. may also be labeled in a similar manner.

[0142] Reference now Figure 8 , the computer vision system and / or the learning model can be used to optimize process parameters, such as gas flow, gas type, gas composition, or other parameters. The optimization system 600 includes a controller 620, the controller 620 includes a process optimization module 624, and the process optimization module 624 includes a learning module 626. The controller 620 communicates with one or more processes 630-1, 630-2, ..., and 630-Y (where Y is an integer greater than zero). Each process includes a controller 638-1, 638-2, ..., and 638-Y that includes a computer vision system 640-1, 640-2, ..., and 640-Y, respectively. In some examples, some of the computer vision systems 640-1, 640-2, ..., and 640-Y may also include a learning model 650-1, 650-2, ..., and 650-Y. The process optimization module 624 receives a plurality of training sets including process conditions and results from the controllers 638-1, 638-2, ..., and 638-Y. The process optimization module 624 uses a learning model, neural network, or artificial intelligence implemented by the learning module 626 to optimize the process conditions for each of the controllers 638 - 1 , 638 - 2 , . . . , and 638 -Y.

[0143] The system and method for detecting the endpoint of residual film removal uses an inexpensive camera to identify features on the surface of one or more components in a process chamber to determine the cleaning state of the process chamber. After sufficient training data, a neural network (deep learning) model is used to change the cleaning process parameters (such as pressure, flow, pedestal position, etc.) to match the cleaning performance of different chambers.

[0144] The systems and methods of the present disclosure can be used to compare the performance of processes (e.g., cleaning processes, etching processes, deposition processes, etc.) in multiple process chambers in a single tool or several tools. Typically, process chambers used for specific processes (e.g., etching or deposition processes) have similar designs. Therefore, it is expected that the processes (e.g., cleaning, etching, deposition, etc.) performed in these process chambers have similar performance (e.g., similar run times). That is, two identical process chambers should perform the same process in the same way. In other words, the performance of the same process in two similarly designed process chambers should also be similar. When performing processes in these chambers, the systems and methods of the present disclosure can analyze data collected from process chambers of similar design. Then, the systems and methods can determine whether the performance of the process is similar between these process chambers based on the analysis (i.e., whether these process chambers are matched and whether the process is performed in the same way).

[0145] In addition, the analysis can help determine the performance of the process at the component level. For example, in the case of a cleaning process, the analysis can help determine whether the time taken by the cleaning process to clean a particular component is the same across multiple similarly designed process chambers. This determination can be in addition to determining whether the total run time of the cleaning process is the same for each process chamber.

[0146] A similar analysis can be performed to compare the performance of other processes such as deposition and etching processes between multiple process chambers. Based on the comparison, any anomalies in the execution of the process in a particular chamber can be detected, and the cause of the anomaly can be corrected. For example, one process chamber in a tool that includes multiple process chambers can be cleaned faster than other process chambers in the tool. The cleaning process for the process chamber should be stopped as early as possible to prevent damage to the process chamber. If the cleaning process is continued, the process chamber may be damaged. However, if multiple process chambers of multiple tools exhibit different cleaning times, the different cleaning times of these process chambers may indicate that design changes need to be made to these process chambers to achieve chamber matching. Similar analysis can be performed for other processes such as deposition and etching processes to achieve chamber matching.

[0147] The systems and methods of the present disclosure can be used to determine various performance criteria for processes such as deposition and etching processes. For example, performance criteria can include uniformity of etching across the wafer, uniformity of deposition across the wafer, uniformity of temperature across the wafer, and the like. For example, uniformity of film thickness can be determined during film deposition as well as film removal, as described below. Notably, the systems and methods can make these determinations in situ (i.e., without removing the wafer from the process chamber to a separate metrology chamber after processing to make these determinations) and in real time (i.e., while the process is being performed on the wafer).

[0148] For example, during the removal of a film from a wafer, the systems and methods can use techniques such as optical interferometry to observe the entirety of the film as it is thinned and determine whether the film is completely removed, whether the removal is uniform across the wafer, and so on. For example, consider removing a copper oxide film from a wafer. Copper changes color when oxidized. Therefore, during the removal of the copper oxide, the systems and methods of the present disclosure can observe changes in the wavelength of light reflected from the entire wafer as the copper oxide is removed and pure copper is exposed. The change in wavelength can indicate the state of the copper oxide removal. Using optical filters, the systems and methods can observe changes locally on the wafer as well as across the entire wafer.

[0149] Without the ability to observe the entire wafer, only a single point on the wafer may be observed, and an erroneous conclusion may be drawn about the removal of the film from the entire wafer based on the observation of the single point on the wafer. For example, suppose only the center of the wafer is observed and it is determined that the copper oxide is removed from the center. Without the ability to observe the entire wafer, an erroneous conclusion may be drawn that the copper oxide of the entire wafer has been removed, even though some of the copper oxide at the edge of the wafer has not been removed. In contrast, with the ability to observe the entire wafer, the systems and methods of the present disclosure can detect whether the film is removed faster from the center of the wafer than from the edge of the wafer, and can detect whether the film is removed from the entire wafer. Furthermore, these determinations are made in situ.

[0150] In another example, when a film is grown on a wafer, it typically takes time for the film to grow due to nucleation delays. Since the entire wafer can be observed, the systems and methods of the present disclosure can detect whether a film grows on a portion of a wafer before growing on another portion of the wafer. For example, an ultraviolet filter can be used to observe film growth (i.e., changes in film thickness due to growth) across the wafer. In addition, an infrared filter can be used to observe the temperature of the wafer and changes therein. The two observations (film thickness and wafer temperature) made using two filters (ultraviolet filter and infrared filter) can be correlated to determine the uniformity of film growth across the wafer. It is noteworthy that this determination is performed in real time in situ in the processing chamber without moving the wafer to a separate metrology chamber for performance.

[0151] In addition, the systems and methods of the present disclosure can monitor the edge of the wafer. For example, by focusing a camera on the top edge of the wafer, the systems and methods can determine whether the wafer has been processed (e.g., cleaned, etched, etc.) to the edge of the wafer. Again, this is done in situ, rather than after the wafer is moved to a separate metrology chamber. This ability to monitor the edge of the wafer while performing a process on the wafer can save processing time and processing materials by immediately interrupting the processing after determining that the processing has completed to the edge of the wafer based on observing the edge.

[0152] In addition, in some cases, deposition can be performed on the top and bottom surfaces of the wafer. The systems and methods of the present disclosure can be used to observe deposition processes on both the top and bottom surfaces of the wafer. For example, a fiber optic probe and / or a micro lens can be used in combination with the systems and methods of the present disclosure to observe a large area at the bottom of the wafer and obtain an image thereof. In some cases, one or more LEDs can be installed in the processing chamber to illuminate the bottom of the wafer. The captured images can be processed and analyzed to determine the uniformity of deposition across the bottom of the wafer. For example, the processed images can be associated with the actual measured deposition levels (which have been previously measured), and the correlation can then be used to infer the deposition levels and uniformity in situ.

[0153] The systems and methods of the present disclosure can also be used to measure the gap between a nozzle and a pedestal in a processing chamber. For example, two cameras can be focused on the nozzle and the pedestal, respectively, and the video signals from the two cameras can be used to determine the distance between the nozzle and the pedestal. The measurement results can then be used to control (change) the gap between the nozzle and the pedestal (e.g., by controlling a device that moves the pedestal relative to the nozzle).

[0154] In addition, in some processing chambers, parasitic plasma can linger in the processing chamber after the plasma is extinguished. Parasitic plasma is undesirable and difficult to detect. The camera of the system and method of the present disclosure can be used to detect parasitic plasma. Many other applications of the system and method of the present disclosure are contemplated and are within the scope of the present disclosure.

[0155] Fig. 9A method 700 for chamber matching according to the present disclosure is shown. For example, only a cleaning process is used. The teachings of method 700 can also be used with a deposition or etching process. At 702, method 700 starts a cleaning process to clean multiple chambers. At 704, method 700 collects data associated with cleaning multiple chambers. At 706, method 700 analyzes the collected data. At 708, method 700 determines whether the data from one of the multiple chambers is abnormal. For example, method 700 determines whether one of the multiple chambers has a shorter run time for the cleaning process than the rest of the multiple chambers (i.e., the cleaning is faster). At 710, if an abnormality is detected, method 700 determines one or more causes of the abnormality associated with one of the multiple chambers. At 712, method 700 corrects one or more causes of the abnormality. Thereafter, or if no abnormality is detected at 708, at 714, method 700 determines whether all chambers are clean or whether the cleaning process should continue (i.e., whether the cleaning process has not yet been completed). If all chambers have not been cleaned (ie, if the cleaning process cannot be terminated yet), method 700 returns to 702 .

[0156] Fig.10 A method 750 for matching process chambers at the component level according to the present disclosure is shown. Again, for example, only a cleaning process is used. The teachings of method 750 can also be used with a deposition or etching process. At 752, method 750 starts a cleaning process to clean multiple chambers. At 754, method 750 collects data associated with cleaning multiple chambers. At 756, method 750 analyzes the collected data. At 758, method 750 determines whether a component in one of the multiple chambers is cleaned before the same component is cleaned in other chambers (i.e., whether it is cleaned faster than the same component in other chambers). At 760, if a component in one of the multiple chambers is cleaned before the same component is cleaned in other chambers (i.e., faster than the same component in other chambers), method 750 stops the cleaning process in that chamber. Thereafter, if the components in one of the plurality of chambers are not cleaned before the same components are cleaned in the other chambers (i.e., not cleaned faster than the same components in the other chambers), then at 762, method 750 determines whether all chambers are clean or whether the cleaning process should continue (i.e., whether the cleaning process has not yet been completed). If all chambers have not yet been cleaned (i.e., if the cleaning process cannot yet be terminated), method 750 returns to 752.

[0157] Fig.11AA schematic diagram of a system 800 for determining uniformity of a film deposited on a substrate according to the present disclosure is shown. The system 800 includes a processing chamber 112 that processes (e.g., deposits a material on) a substrate 118. The processing chamber 112 includes a window 172. The system 800 includes a controller 170 that includes a computer vision system 180, a camera 174, and a camera positioning device 182.

[0158] System 800 also includes ultraviolet filter 806 and infrared filter 808. Ultraviolet filter 806 and infrared filter 808 are disposed between camera 174 and window 172. Ultraviolet filter 806 filters wavelengths of ultraviolet light received from substrate 118 through window 172. Infrared filter 808 filters wavelengths of infrared light received from substrate 118 through window 172.

[0159] The computer vision system 180 receives the output of the ultraviolet filter 806 and the infrared filter 808. The computer vision system 180 determines the thickness of the deposited material at multiple locations on the substrate 118 based on the output of the ultraviolet filter 806. The computer vision system 180 determines the temperature at multiple locations of the substrate 118 based on the output of the infrared filter 808. The computer vision system 180 correlates the determinations of thickness and temperature and determines the uniformity of deposition across the substrate 118 based on the correlation.

[0160] Fig. 11B A schematic diagram of a system 850 for determining whether a material is uniformly deposited on both the top and bottom of a substrate in accordance with the present disclosure is shown. The system 850 includes a processing chamber 999 that processes the upper and lower sides of a substrate 804 (e.g., deposits material thereon, as indicated by arrows). The processing chamber 802 includes a window 172. The details of depositing material on both sides of the substrate 804 (e.g., using CVD) are not relevant to the present disclosure and are therefore not shown and described. The system 850 includes a controller 170 that includes a computer vision system 180, a camera 174, and a camera positioning device 182.

[0161] The system 850 further includes an optical sensor 810 (eg, a fiber optic sensor or a micro lens) to observe the bottom of the substrate 804. A sensor positioning device 812 positions the optical sensor 810 to point to different locations at the bottom of the substrate 804.

[0162] The computer vision system 180 focuses the camera 174 on the top of the substrate 804. The computer vision system 180 receives an image of the top of the substrate 804 from the camera 174. The computer vision system 180 positions the optical sensor 810 to view the bottom of the substrate 804. The computer vision system 180 receives an image of the bottom of the substrate 804 from the optical sensor 810. The computer vision system 180 processes the images received from the camera 174 and the optical sensor 810. The computer vision system 180 determines the uniformity of the deposition across the top and across the bottom of the substrate 804 based on the processing of the images received from the camera 174 and the optical sensor 810.

[0163] Fig.12 A method 900 for determining whether a process (e.g., cleaning, etching, deposition, etc.) performed on a component has been performed uniformly across the component is shown in accordance with the present disclosure. At 902, the method 900 filters one or more wavelengths of light received from a component (e.g., a wafer) through a window provided on a processing chamber in which the component is being processed. At 904, the method 900 determines a state of the component based on filtering using optical interference. At 906, the method 900 determines whether the process has been performed uniformly across the component based on the state of the component. At 908, if the process has been performed uniformly across the component, the method 900 terminates the process. If the process has not been performed uniformly across the component, the method 900 returns to 902.

[0164] Fig.13 A method 920 for determining whether a process (e.g., cleaning, etching, deposition, etc.) performed on a component has been performed to the edge of the component is shown in accordance with the present disclosure. At 922, the method 920 focuses a camera on the edge of the component (e.g., a wafer). At 924, the method 920 determines whether the process has completed to the edge of the component. At 926, if the process has completed to the edge of the component, the method 920 terminates the process. If the process has not completed to the edge of the component, the method 920 returns to 922.

[0165] Fig.14 A method 940 for determining the uniformity of a film deposited on a substrate according to the present disclosure is shown. At 942, the method 940 measures the thickness of the film across the substrate using an ultraviolet filter, as described in reference Fig.11A At 944, method 940 measures the temperature across the substrate using an infrared filter, as described in reference Fig.11AAt 946, method 940 correlates the thickness and temperature measurements. At 948, method 940 determines whether the film has been uniformly deposited across the substrate based on the measurements. At 950, if the film has been uniformly deposited across the substrate, method 940 terminates the deposition process. If the film has not been uniformly deposited across the substrate, method 940 returns to 942.

[0166] Fig.15 is a method 960 for determining whether a material is uniformly deposited on both the top and bottom of a substrate according to the present disclosure. At 962, the method 960 uses the method described in reference Fig. 11B The camera observes the top of the substrate. At 964, method 960 uses the reference Fig. 11B The one or more optical sensors observe the bottom of the substrate. At 966, method 960 determines whether the film is uniformly deposited on the top and bottom of the substrate, as shown in reference Fig. 11B At 968, if the film has been uniformly deposited on the top and bottom of the entire substrate, the method 960 terminates the deposition process. If the film has not been uniformly deposited on the top and bottom of the entire substrate, the method 960 returns to 962.

[0167] The above methods are described separately for the sake of clarity only. One or more of the above methods can be combined in whole or in part and can be performed together.

[0168] The foregoing description is merely illustrative in nature and is not intended to limit the present disclosure, its application or use. The broad teachings of the present disclosure can be implemented in various forms. Therefore, although the present disclosure includes specific examples, the true scope of the present disclosure should not be so limited, because when studying the drawings, description and appended claims, other modifications will become apparent. It should be understood that one or more steps in the method can be performed in different orders (or simultaneously) without changing the principles of the present disclosure. In addition, although each embodiment is described above as having certain features, any one or more of those features described with respect to any embodiment of the present disclosure can be implemented in the features of any other embodiment and / or combined with the features of any other embodiment, even if the combination is not explicitly described. In other words, the described embodiments are not mutually exclusive, and the replacement of one or more embodiments with each other remains within the scope of the present disclosure.

[0169] Various terms including "connected," "engaged," "coupled," "adjacent," "near," "on," "above," "below," and "disposed" are used to describe spatial and functional relationships between elements (e.g., between modules, circuit elements, semiconductor layers, etc.). Unless explicitly described as "direct," when a relationship between a first and a second element is described in the above disclosure, the relationship may be a direct relationship in which no other intermediate elements exist between the first and second elements, but may also be an indirect relationship in which one or more intermediate elements exist (spatially or functionally) between the first and second elements. As used herein, the phrase "at least one of A, B, and C" should be interpreted to mean a logical (A or B or C), using a non-exclusive logical OR, and should not be interpreted to mean "at least one of A, at least one of B, and at least one of C."

[0170] In some implementations, the controller is part of a system, which can be part of the above examples. Such a system can include a semiconductor processing device, which includes one or more processing tools, one or more chambers, one or more platforms for processing, and / or specific processing components (wafer pedestals, gas flow systems, etc.). These systems can be integrated with electronic devices for controlling their operation before, during, and after the processing of semiconductor wafers or substrates. The electronic device can be referred to as a "controller", which can control various components or subcomponents of one or more systems. Depending on the processing requirements and / or system type, the controller can be programmed to control any of the processes disclosed herein, including the delivery of process gases, temperature settings (e.g., heating and / or cooling), pressure settings, vacuum settings, power settings, radio frequency (RF) generator settings, RF matching circuit settings, frequency settings, flow rate settings, fluid delivery settings, position and operation settings, wafer delivery in and out tools and other delivery tools and / or load locks connected to or interfaced with a specific system.

[0171] Broadly speaking, a controller can be defined as an electronic device having various integrated circuits, logic, memory, and / or software that receive instructions, issue instructions, control operations, enable cleaning operations, enable endpoint measurements, etc. The integrated circuits can include a chip in the form of firmware that stores program instructions, a digital signal processor (DSP), a chip defined as an application specific integrated circuit (ASIC), and / or one or more microprocessors, or a microcontroller that executes program instructions (e.g., software). The program instructions can be instructions transmitted to the controller in the form of various individual settings (or program files) that define operating parameters for performing specific processing on or for a semiconductor wafer or system. In some embodiments, the operating parameters can be part of a recipe defined by a process engineer to complete one or more processing steps in the preparation of one or more layers, materials, metals, oxides, silicon, silicon dioxide, surfaces, circuits, and / or dies of a wafer.

[0172] In some implementations, the controller may be part of or coupled to a computer that is integrated with the system, coupled to the system, otherwise networked to the system, or a combination of these. For example, the controller may be in the "cloud", or in all or part of a wafer fab host computer system that enables remote access to wafer processing. The computer may enable remote access to the system to monitor the current progress of a manufacturing operation, study the history of past manufacturing operations, study trends or performance criteria from multiple manufacturing operations, change parameters of a current process, set a processing step after the current process, or start a new process. In some examples, a remote computer (e.g., a server) may provide a process recipe to the system via a network (which may include a local network or the Internet). The remote computer may include a user interface that enables input or programming of parameters and / or settings, which are then transmitted from the remote computer to the system. In some examples, the controller receives instructions in the form of data that specify parameters for each processing step to be performed during one or more operations. It should be understood that the parameters may be specific to the type of processing to be performed and the type of tool that interfaces with or is controlled by the controller. Thus, as described above, the controller may be distributed, such as by including one or more discrete controllers networked together and working toward a common purpose (e.g., processing and control as described herein). An example of a distributed controller for such a purpose is one or more integrated circuits on a chamber that communicate with one or more integrated circuits located remotely (e.g., at a platform level or as part of a remote computer), which combine to control the processing on the chamber.

[0173] Example systems may include, but are not limited to, plasma etch chambers or modules, deposition chambers or modules, spin rinse chambers or modules, metal plating chambers or modules, cleaning chambers or modules, chamfer edge etch chambers or modules, physical vapor deposition (PVD) chambers or modules, chemical vapor deposition (CVD) chambers or modules, atomic layer deposition (ALD) chambers or modules, atomic layer etch (ALE) chambers or modules, ion implantation chambers or modules, track chambers or modules, and any other semiconductor processing system that may be associated with or used in the preparation and / or manufacture of semiconductor wafers.

[0174] As described above, depending on one or more process steps to be performed by the tool, the controller can communicate with one or more other tool circuits or modules, other tool components, cluster tools, other tool interfaces, adjacent tools, neighboring tools, tools located throughout the factory, a host computer, another controller, or tools used in material transport to transport wafer containers to and from tool locations and / or load ports in a semiconductor manufacturing facility.

Claims

1. A substrate processing system, comprising: a camera mounted externally and adjacent to a window of a processing chamber configured to process semiconductor substrates, the window enabling the camera to observe components within the processing chamber, the camera being configured to generate video signals indicative of a status of the components during a process being performed within the processing chamber; an optical filter disposed between the camera and the window, the optical filter configured to filter one or more wavelengths of light received from the component through the window and output a filtered signal to the camera; as well as a controller coupled to the process chamber and configured to: controlling the camera; processing the video signal from the camera; determining a state of the component based on processing the filtered signal using optical interferometry; as well as determining whether to terminate the processing based on the status of the component, Wherein the controller is configured to determine a rate at which the process is performed at different locations on the component.

2. The substrate processing system according to claim 1, wherein: In response to the process being a cleaning process performed to remove material deposited on the component by a previously performed process, the controller is configured to: comparing a change in a property of a feature of the component observed across a plurality of frames of the video signal to a predetermined threshold; determining whether the material deposited on the component is removed based on the comparing; as well as In response to determining that the material deposited on the component is removed, the cleaning process is terminated.

3. The substrate processing system according to claim 1, wherein: In response to the process being a cleaning process performed to remove material deposited on the component by a previously performed process, the controller is configured to: comparing an image captured from the video signal with a predetermined image; determining whether the material deposited on the component is removed based on the comparing; as well as In response to determining that the material deposited on the component is removed, the cleaning process is terminated.

4. The substrate processing system according to claim 1, wherein the controller is configured to: receiving data from one or more sensors in the process chamber; generating a model based on the data received from the one or more sensors and a video signal received from the camera, the video signal indicating a state of the component when the process was previously performed in the process chamber; as well as Use this model to: processing the video signal; determining a status of the component based on the processing of the video signal; as well as Based on the determined status of the component, it is determined whether to terminate the process.

5. The substrate processing system according to claim 1, wherein: The controller is configured to determine whether the process is performed on the entire component before terminating the process.

6. The substrate processing system according to claim 1, wherein: The controller is configured to determine whether the process is performed uniformly across the component before terminating the process.

7. The substrate processing system of claim 1 , wherein the component comprises a semiconductor substrate, and the processing comprises a film removal process performed to remove a film from the semiconductor substrate, and wherein the controller is configured to determine whether the film on the entire component is removed before terminating the processing.

8. The substrate processing system of claim 1, wherein the controller is configured to focus the camera at an edge of the component and determine whether the process is performed at the edge of the component before terminating the process.

9. The substrate processing system of claim 1 , wherein the component comprises a semiconductor substrate and the process comprises a deposition process, the system comprising: a first optical filter disposed between the camera and the window, the first optical filter configured to filter wavelengths of ultraviolet light received from the component through the window; and a second optical filter disposed between the camera and the window, the second optical filter configured to filter wavelengths of infrared light received from the component through the window; Wherein, the controller is configured as: determining a thickness of deposited material at a plurality of locations on the component based on an output of the first optical filter; determining a temperature at a plurality of locations of the component based on an output of the second optical filter; associating said determination of said thickness with said determination of said temperature; and A deposition uniformity across the component is determined based on the correlation.

10. The substrate processing system of claim 1, wherein the component comprises a semiconductor substrate and the process comprises a deposition process, the system further comprising: an optical sensor configured to observe a bottom portion of the component, The controller is configured as follows: focusing the camera on a top portion of the component; processing images received from the optical sensor and the camera; as well as Deposition uniformity across the top portion and across the bottom portion of the component is determined based on the processing of the images.

11. The substrate processing system according to claim 1, further comprising: a second controller coupled to a second process chamber in which the same process is performed, the second controller being configured to control a second camera associated with the second process chamber; and A third controller is configured to: analyzing data from the controller and the second controller; comparing the performance of the process in the process chamber to the performance of the process in the second process chamber based on the analyzed data; as well as Based on the comparison, it is determined whether the performance of the process in the process chamber matches the performance of the process in the second process chamber.

12. The substrate processing system according to claim 1, further comprising: a second controller coupled to a second processing chamber in which the same said process is performed on the same said component, the second controller being configured to control a second camera associated with the second processing chamber; and A third controller is configured to: analyzing data from the controller and the second controller; comparing the performance of the process on the component in the process chamber with the performance of the process on the component in the second process chamber based on the analyzed data; as well as Based on the comparison, it is determined whether the performance of the process on the component in the process chamber matches the performance of the process on the component in the second process chamber.

13. The substrate processing system according to claim 12, wherein: In response to the process in the process chamber being completed earlier than the process in the second process chamber, the third controller is configured to terminate the process in the process chamber earlier than the process in the second process chamber.

Images