Multi-sensor determination of state of semiconductor device

Abstract

Methods and apparatus are provided for multi-sensor determination of a state of a semiconductor device. In some embodiments disclosed herein, a semiconductor manufacturing apparatus may include: a plurality of sensors including one or more spatial sensors, one or more spectral sensors, and one or more temporal sensors disposed around the semiconductor manufacturing; and a controller communicatively coupled to the plurality of sensors, the controller configured such that: a set of signals is determined from the plurality of sensors for monitoring during a process performed by the semiconductor manufacturing apparatus; obtaining measurements relating to the set of signals from the plurality of sensors during the process; and determining an indication of a state of the semiconductor manufacturing apparatus based on a combination of data generated from the measurements related to the set of signals.

Description

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

[0001] Sensors associated with semiconductor equipment, such as multi-station manufacturing tools (including, for example, multi-station processing chambers), generally perform specific types of tasks. For example, automatic infrared endpoint detection (IR-EPD) is used to perform endpoint detection on all stations (e.g., all four stations of a multi-station processing chamber) during cleaning. Some sensors address specific types of problems. For example, one or more cameras are used alone to study the characteristics of a plasma. In addition, there may be ex-situ measurements performed using metrology tools, each of which can provide an indication of a specific state or characteristic associated with the device (film thickness, resistivity, refractive index (RI), stress, etc.). In general, sensors or metrology tools are generally used for specific tasks or specific types of problems.

[0002] The background and overview contained herein are provided only for the purpose of generally presenting the content of the present disclosure. Most of the present disclosure presents the work of the inventors, and just because such work is described in the background section or presented as content elsewhere in this article does not mean that it is admitted to be prior art. Summary of the invention

[0003] In one aspect of the present disclosure, a manufacturing system is disclosed. In some embodiments, the manufacturing system includes: semiconductor manufacturing equipment; a plurality of sensors, including one or more spatial sensors, one or more spectral sensors, and one or more time sensors disposed around the semiconductor manufacturing equipment; and a controller, which is communicatively coupled to the plurality of sensors, the controller being configured to: determine a signal group from the plurality of sensors for monitoring during a process performed by the semiconductor manufacturing equipment; obtain measurement results related to the signal group from the plurality of sensors during the process; and determine an indication of a state of the semiconductor manufacturing equipment based on a data combination generated from the measurement results related to the signal group.

[0004] In another aspect of the present disclosure, a method for multi-sensor determination of a state of a semiconductor device is disclosed. In some embodiments, the method includes: determining a signal group from a plurality of sensors of the semiconductor device for monitoring during a process to be performed by the semiconductor device, the plurality of sensors including at least one spatial sensor, at least one spectral sensor, and at least one temporal sensor; performing sensor measurements based on the process to be performed via two or more of: (i) the at least one spatial sensor, (ii) the at least one spectral sensor, or (iii) the at least one temporal sensor; and determining the state of the semiconductor device based on the sensor measurements.

[0005] In another aspect of the present disclosure, a multi-sensor measurement device is disclosed. In some embodiments, the multi-sensor measurement device includes: a housing having a major dimension of about 10 inches or less; a spectral sensor in the housing; a spatial sensor in the housing; a temporal sensor in the housing; and a physical interface whose shape allows the housing to be attached to a surface of the manufacturing tool and receive electromagnetic signals related to a process performed by the manufacturing tool through a window of the manufacturing tool.

[0006] In another aspect of the present disclosure, a multi-sensor determination method for a process chamber cleaning endpoint is disclosed.

[0007]

[0011] In another aspect of the present disclosure, a method of determining the presence of an undesirable species within a processing chamber is disclosed.

[0008] These and other features of the disclosed embodiments will be described in detail below with reference to the associated drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1A A manufacturing tool utilizing a plasma process to deposit or etch films on or over a substrate is shown; the tool includes a camera sensor.

[0010] Figure 1B A schematic diagram of an implementation of a multi-station processing tool is presented; the tool includes four camera sensors.

[0011] Figure 1C A top view of an electronic device manufacturing system having four multi-station manufacturing tools, one of which includes a camera sensor, is presented.

[0012] Figure 2 A processing chamber is schematically depicted with an exemplary camera oriented to capture images along a horizontal marker line leading into the chamber interior.

[0013] Figure 3 A processing chamber is schematically depicted with an exemplary camera oriented to capture images along a vertical line of sight to the interior of the chamber.

[0014] Figure 4 is a schematic diagram of an exemplary optical emission spectroscopy (OES) sensor according to some embodiments.

[0015] Figure 5 is a flow chart illustrating a method for determining a status of semiconductor manufacturing equipment, according to some embodiments.

[0016] Fig. 6A and Figure 6B is a flow chart illustrating a method for determining a status of semiconductor manufacturing equipment, according to some embodiments.

[0017] Fig. 7A and Figure 7B is a flow chart illustrating a method for determining a status of semiconductor manufacturing equipment, according to some embodiments.

[0018] Figure 8 is a block diagram showing a hardware configuration for a multi-sensor manufacturing tool (eg, a multi-station manufacturing tool) that utilizes multiple types of sensors to implement a combined sensor set for manufacturing equipment, according to some embodiments.

[0019] Fig. 9 A manufacturing tool (e.g., Figure 8 A cross-sectional view of an exemplary viewport of a multi-sensor manufacturing tool having a sensor group associated therewith.

[0020] Fig.10 An exterior perspective view of a chamber is shown showing an exemplary viewport for a manufacturing tool, according to some embodiments.

[0021] Fig.11 is a flow chart illustrating a multi-sensor determination method for a process chamber cleaning endpoint, according to some embodiments.

[0022] Fig. 12A and Fig. 12B is a flow chart illustrating a method for determining the presence of an undesired species within a processing chamber, according to some embodiments.

[0023] Fig.13 A schematic diagram of components of a computer device implemented in a computer system is shown according to some implementations. DETAILED DESCRIPTION

[0024] The present disclosure relates to the use of a multi-sensing system to characterize semiconductor equipment, such as a multi-station processing chamber. Current methods for sensing and monitoring the state of semiconductor equipment generally involve the use of sensors for corresponding tasks or task classes. The measurements obtained from each sensor are used individually to obtain or infer an indication of one or more characteristics associated with the equipment. In some cases, expert judgment is used to determine the characteristics based on experience or intuition, or based on results obtained from other sensors or technologies. That is, in some cases, the sensing technology may require a human factor to derive meaningful data.

[0025] In some common scenarios related to process engineering, process performance checks are performed by, for example, visually verifying how the plasma behaves in some way when viewed from a viewport based on the response of a sensor to one of hundreds of related channels, or (most commonly) by performance on a substrate (e.g., a 300 mm wafer) after the process is completed. These methods do not provide real-time responses. Engineers do not always have time to check the plasma for each recipe, and in rapidly changing systems, subtle system excursions may not be easily understood by visual inspection because plasma dynamics are on the order of microseconds (μs) to milliseconds (ms). Without an automated system, continuous monitoring of the relevant channels is not feasible, and subtle changes may not be captured with current sensor implementations, or subtle changes may be seen as noise in the system. The most reliable (and most commonly used) indication of process variation is the effect a specific recipe has after the process. Post-process metrology indicates whether the system state has changed using the response of characteristics on the wafer (deposited film thickness, refractive index (RI), etc.). However, not all wafers are measured after the process. These sensing inefficiencies may result in yield impact or wafer scrapping, and do not make real-time control feasible.

[0026] In some common scenarios regarding clean endpoint (i.e., endpoint of chamber clean), endpoint detection relies on timed cleans (which do not account for system or process variability and / or accumulation of variations based on different processes), or the use of IR-EPD. IR-EPD looks for specific voltage and signal slopes, and adds an over-etch step. The over-etch step may cause significant etching in some areas of the workstation, thereby shortening the life of the susceptor, for example due to aluminum fluoride (AlF 3 ) is caused by the formation of . Another method of endpoint detection in chamber cleaning involves the use of visual signals, such as using a camera. However, such vision-based detection is limited to the visible area, and accurately determining which area is the slowest or last part of cleaning requires extensive verification.

[0027] Therefore, there is a need for a sensing method that can acquire and provide continuous accurate signals in an automated manner to provide a better understanding of the status of the equipment system and provide opportunities for control.

[0028] The following terms are used throughout this manual:

[0029] "Fabrication equipment" refers to equipment in which a manufacturing process is performed. Fabrication equipment typically has a processing chamber in which a workpiece is located during processing. Typically, when in use, the fabrication equipment performs one or more semiconductor device fabrication operations. Examples of fabrication equipment used for semiconductor device fabrication include deposition reactors, such as electroplating units, physical vapor deposition reactors, chemical vapor deposition reactors, and atomic layer deposition reactors, and subtractive process reactors, such as dry etching reactors (e.g., chemical and / or physical etching reactors), wet etching reactors, and ashers. In some embodiments, the fabrication equipment can be a multi-station processing chamber having, for example, four stations.

[0030] As referred to herein, manufacturing equipment is sometimes referred to simply as a "processing chamber". In many embodiments, the processing chamber is typically a sealed enclosure in which the substrate is fixed during processing. The processing chamber may include components related to the delivery and removal of gases. It may also include components related to the generation of plasma and the control of plasma characteristics in the chamber. It may include components for controlling pressure (including evacuating the chamber). In the context of the present disclosure, the processing chamber may include a pedestal on which the substrate is located when the substrate is processed. The pedestal may be equipped with a chuck such as an ESC to position the substrate during processing.

[0031] As used herein, "semiconductor device manufacturing operations" are operations performed during the manufacture of semiconductor devices. As mentioned herein, such manufacturing operations are sometimes referred to as "processes" or "processes". Examples of processes include depositing materials on a substrate, selectively etching materials from a substrate, and ashing photoresists on a substrate. Typically, the entire manufacturing process includes a plurality of semiconductor device manufacturing operations, each of which is performed in its own semiconductor manufacturing tool, such as a plasma reactor, an electroplating unit, a chemical mechanical planarization tool, a wet etching tool, and the like. Categories of semiconductor device manufacturing operations include subtractive processes, such as etching processes and planarization processes, and additive processes, such as deposition processes (e.g., physical vapor deposition, chemical vapor deposition, atomic layer deposition, electrochemical deposition, electroless deposition). In the context of an etching process, a substrate etching process includes a process for etching a mask layer, or more generally, includes a process for etching any material layer previously deposited on and / or otherwise resting on a substrate surface. Such an etching process can etch a stack of layers in a substrate.

[0032] The terms "semiconductor wafer," "wafer," "substrate," "wafer substrate," and "partially fabricated integrated circuit" may be used interchangeably. One of ordinary skill in the art understands that the term "partially fabricated integrated circuit" may refer to a semiconductor wafer during any of the many stages of integrated circuit fabrication thereon. Wafers or substrates used in the semiconductor device industry typically have a diameter of 200 mm, or 300 mm, or 450 mm. In addition to semiconductor wafers, other workpieces that may utilize the disclosed embodiments include items such as printed circuit boards, magnetic recording media, magnetic recording sensors, mirrors, optical elements, display devices or components, such as backplanes for pixelated display devices, flat panel displays, micromechanical devices, and the like. The workpieces may have a variety of shapes, sizes, and materials.

[0033] Figure 1A A manufacturing tool is shown, which is represented as a substrate processing device 100. In various embodiments, the substrate processing device 100 can be configured to deposit a film on or above a semiconductor substrate using any number of processes. For example, the substrate processing device 100 can be configured to perform plasma enhanced chemical vapor deposition (PECVD) or plasma enhanced atomic layer deposition (PEALD). The substrate processing device 100 may include one or more sensors or sensor packages 117 located on the chamber wall. When implemented as a sensor package, the element 117 may include two or more different sensors, which may be different types of sensors. In some embodiments, the sensor includes two or three of the following sensor types: a spatial sensor, a spectral sensor, and a time sensor. The sensor or sensor package 117 can be configured to capture image data from the interior of the device 100. Note that although the sensor or sensor package 117 is shown as a single block, it represents an embodiment in which one, two or more sensors are close to each other, thereby optionally sharing a single viewport or other viewports. In some cases, the individual sensors within the block representing the sensor or sensor package 117 are trained on different components or fields of view in the interior of the chamber. In some cases, each sensor within block 117 may be configured to capture a different corresponding spectral range (far IR, near IR, visible light, UV, etc.).

[0034] Figure 1A The processing apparatus 100 may utilize a single processing station 102 of a processing chamber having a single substrate holder 108 (e.g., a pedestal) within an interior volume that may be maintained under vacuum by a vacuum pump 118. A showerhead 106 and a gas delivery system 101 (which is fluidly coupled to the processing chamber) may allow for delivery of, for example, film precursors, as well as carrier and / or sweep and / or process gases, secondary reactants, and the like.

[0035] exist Figure 1AIn the embodiment of the present invention, the gas delivery system 101 may include a mixing vessel 104 for mixing and / or conditioning process gases for delivery to the showerhead 106. One or more mixing vessel inlet valves 120 may control the introduction of process gases into the mixing vessel 104. Certain reactants may be stored in liquid form prior to vaporization and subsequent delivery to the process station 102 of the process chamber. Figure 1A Embodiments may include a vaporization point 103 for vaporizing liquid reactants to be supplied to the mixing vessel 104. In some implementations, the vaporization point 103 may include a heated liquid injection module. In some other implementations, the vaporization point 103 may include a heated vaporizer. In still other implementations, the vaporization point 103 may be eliminated from the processing station. In some implementations, a liquid flow controller upstream of the vaporization point 103 may provide for controlling the mass flow of liquid for vaporization and delivery to the processing station 102.

[0036] The showerhead 106 is operable to distribute process gases and / or reactants (e.g., film precursors) toward the substrate 112 at the processing station, with the flow rate being controlled by one or more valves (e.g., valves 120, 120A, 105) upstream of the showerhead. Figure 1A In the illustrated embodiment, the substrate 112 is depicted as being located below the showerhead 106 and is shown as being placed on the pedestal 108. The showerhead 106 can include any suitable shape and can include any suitable number and arrangement of ports to distribute process gases to the substrate 112. In some embodiments involving two or more stations, the gas delivery system 101 can include a valve or other flow control structure upstream of the showerhead that can independently control the flow of process gases and / or reactants to each station so that gas flow can be switched to one station while simultaneously inhibiting gas flow to a second station. In addition, the gas delivery system 101 can be configured to independently control the process gases and / or reactants delivered to each station in a multi-station apparatus so that the gas composition provided to different stations is different; for example, at the same time, the partial pressure of the gas components will vary between multiple stations.

[0037] exist Figure 1AIn some implementations, the gas volume 107 is depicted as being located below the showerhead 106. In some implementations, the pedestal 108 can be raised or lowered to expose the substrate 112 to the gas volume 107 and / or to change the size of the gas volume 107. The spacing between the pedestal 108 and the showerhead 106 is sometimes referred to as a "gap". Optionally, the pedestal 108 can be lowered and / or raised during portions of the deposition process to adjust the process pressure, reactant concentration, etc. within the gas volume 107. The showerhead 106 and the pedestal 108 are depicted as being electrically coupled to an RF signal generator 114 and a matching network 116 to couple power to the plasma generator. Thus, the showerhead 106 can be used as an electrode for coupling RF power into the processing station 102. The RF signal generator 114 and the matching network 116 can operate at any suitable RF power level that can operate to form a plasma having a desired composition of radical species, ions, and electrons. Additionally, the RF signal generator 114 can provide RF power having more than one frequency component, such as a low frequency component (e.g., less than about 2 MHz) and a high frequency component (e.g., greater than about 2 MHz). In some implementations, plasma ignition and maintenance conditions can be controlled using appropriate hardware and / or appropriate machine-readable instructions in a system controller, which can provide control instructions via an input / output control instruction sequence.

[0038] In general, the disclosed embodiments may be implemented with any plasma-assisted manufacturing tool, including with integrated sensors (including, for example, sensors or sensor packages 117) configured to acquire data (including, for example, images) related to plasma and / or plasma phenomena. Exemplary deposition apparatus include, but are not limited to, those from Product Series, Product range and / or Product Series, Product Series, Product Series and The LASER® product line of devices, each available from Lam Research Corp. of Fremont, Calif., or any of a variety of other manufacturing tools that employ plasma.

[0039] In addition, in some embodiments, the sensors or sensor packages described herein can be used for multiple purposes. As an example, the sensor or sensor package can be or include a hyperspectral sensor (which is configured to capture intensity values ​​in multiple narrow wavelength intervals) or a multispectral sensor (which is configured to capture intensity values ​​on a wider wavelength interval, usually less than the multiple narrow bands sensed by the hyperspectral sensor). A hyperspectral or multispectral sensor can thereby provide spatial and spectral sensor information, wherein the spectral information can correspond to wavelengths outside the visible spectrum (e.g., at least a portion of the ultraviolet (UV) spectrum, at least a portion of the infrared (IR) spectrum) (infrared) spectrum). As another example, a sensor can have sufficiently enhanced capabilities to provide the functionality of any two or more types of sensors. For example, a camera with a very fast frame rate (e.g., 120 frames per second or higher) can capture spatial and temporal information, and is therefore used as a spatial sensor and a temporal sensor.

[0040] Reference now Figure 1B , depicts an implementation of a multi-station manufacturing tool 150 according to some embodiments. For simplicity, Figure 1A The processing apparatus 100 is shown as a separate station 102 of a processing chamber that is used to maintain a low pressure environment. However, some manufacturing tools use multiple processing stations, such as Figure 1B In some embodiments, the multi-station manufacturing tool 150 may employ a processing chamber 165 that includes a plurality of manufacturing processing stations, each of which may be used to process a wafer held on a wafer holder (e.g., Figure 1A The processing operation is performed on the substrate in the susceptor 108. Figure 1B In the embodiment of the present invention, the processing chamber 165 is shown as having four processing stations 151, 152, 153 and 154. However, in some other embodiments, the multi-station processing apparatus may have more or fewer processing stations, depending on the implementation, for example, the desired level of parallel wafer processing, size or space limitations, or cost limitations. Figure 1B Also shown is a substrate handler robot 175, which can operate under the control of the system controller 190. The substrate handler robot 175 can be configured to load a wafer cassette ( Figure 1B The wafer is moved from the processing station 151 to the processing station 152 (not shown) and enters the multi-station processing chamber 165 and is placed on one of the processing stations 151, 152, 153 or 154.

[0041] like Figure 1BAs shown, processing station 153 has an associated sensor or sensor package 121, which is located within processing station 153 and is configured to obtain in-situ information (e.g., image, spectral and / or temporal data) from within processing station 153 and, in some embodiments, from within processing chamber 154. Processing station 151 can have two associated sensors or sensor packages 123 and 125. Sensor or sensor package 123 is positioned and configured to obtain in-situ information from within processing station 151 and, in some embodiments, from within processing chamber 152. Sensor or sensor package 125 is positioned and configured to obtain in-situ information from within processing station 151 and, in some embodiments, from within processing chamber 153. Processing station 152 can have an associated sensor or sensor package 127, which is positioned and configured to obtain in-situ information from within processing station 152 and, in some embodiments, from within processing chamber 154. When implemented as a sensor package, elements 121, 123, 125, 127 can include two or more different sensors, which can be different types of sensors. Any one or more of the sensors or sensor packages 121, 123, 125, or 127 may be coupled to the interior of the processing chamber 165 via a viewport or other window disposed in the chamber wall. Figure 1B Not shown, but some embodiments may include one or more sensors or sensor packages adjacent to the processing station 154. Exemplary positioning of the sensors or sensor packages will be referred to below. Figure 2 and Figure 3 Further described. In addition, as will be described in detail below, each sensor or sensor package can be a combination of at least one spatial sensor, at least one spectral sensor, or at least one temporal sensor, such as two or more of these types of sensors. Some sensors can be stand-alone sensors of one type, but the use applications described in the present disclosure include multiple types in terms of detectors to achieve greater accuracy, precision, stability, and integrity of measurements.

[0042] In the context of this disclosure, the terms precision, stability, and match may refer to the sigma (σ, standard deviation) of process metrics (measurements) reported by a sensor or sensor package in high volume manufacturing (HVM). Apparently, multiple replicas of a sensor or sensor package (measurement device) are used to repeatedly measure the same process over an extended period of time. The same sensor package may be installed at multiple stations within a process module and at multiple process modules, where the process modules are installed on multiple tools of multiple HVM production lines that may be located in different locations. These online HVM process metrics from the sensor or sensor package can be used to develop correlations with other HVM process metrics, including process metrics obtained using precision offline professional metrology techniques. Such correlations can then achieve near real-time process optimization and root cause analysis of process deviations.

[0043] The measurement result or metric can be recorded by a host device (e.g., a computer device) that receives the data. Precision can be defined as the sigma of a metric repeated in the shortest possible time under almost identical conditions. Precision can quantify or represent the basic limit of a metric, and typically requires more than 10 metric input values ​​to calculate sigma, where the values ​​are acquired in a fast sequence without unnecessary interruptions. Stability can be defined as the sigma of a metric reported within a defined time period. Stability can quantify or represent a process study or a gage study (which indicates the repeatability and reproducibility of a measurement), which typically occurs over multiple days, such as over the duration of a process suite. For many metrics, process stability and gage stability may not be easily separated in the data. Careful use of stability results of multiple metrics can help identify gage shifts from process shifts. High stability is beneficial because predictive maintenance and process yield improvement must be based on metrics with known stability. Matching can be defined as the sigma (stability data set of a sensor subsystem) of a metric of multiple sensors or sensor packages acquired over multiple days. An example of matching is within a single process module (PM), which contains multiple sensor subsystems. In particular, in intra-PM matching, multiple subsystems on a single PM can be used to quantify the PM matching sigma, a metric. With inter-PM matching, matching can be measured, for example, within a manufacturing tool, across manufacturing tools, or across a mounting base at multiple locations. Subsystem (e.g., a subsystem of a sensor or sensor package) specific control limits can be used to manage non-idealities in matching, but this approach is cumbersome. Metrics that favor good matching of subsystems that function well are favored. In general, process accuracy is easiest (single chamber, short time), process stability is more difficult to achieve (because it drifts over time, for example), and process variation from chamber to chamber or tool to tool is the most difficult.

[0044] In some implementations, the sensor or sensor package can report measurements as a function of time to the host device. This data can be further refined to define customized measurements by the end user of the sensor or sensor package. The accuracy, stability, and matching of the measurements help define the limits of the sensor package to manage the underlying process. Finally, the sensor or sensor package requires process accuracy, process stability, and process matching, for example, which is used to develop process metrology technology, perform process quality control (e.g., determine whether the behavior is normal, whether the abnormal behavior is an impulse or offset problem, whether the behavior requires corrective action), and establish process tolerance limits for measurements that the end user considers important.

[0045] In some embodiments, the manufacturing tool 150 may include a system controller 190 configured to control process conditions and hardware states of the manufacturing tool 150. In some embodiments, the system controller 190 may interact with one or more sensors, gas flow subsystems, temperature subsystems, and / or plasma subsystems (collectively referred to as block representative subsystems 191) to control process gas flows, thermal conditions, and / or plasma conditions as appropriate for controlling a manufacturing process. In various implementations, the system controller 190 and subsystems 191 may be used to implement a recipe or other process condition in one or more processing stations (e.g., 151-154) of the processing chamber 165. The system controller may be located entirely on or adjacent to the manufacturing tool (e.g., as an edge computer in a manufacturing facility), or the system controller may be located entirely remote from the manufacturing tool (e.g., on a host cloud computing resource), or the system controller may be located partially on the manufacturing facility and partially remote.

[0046] In a multi-station manufacturing tool, the RF signal generator can be coupled to an RF signal distribution unit, which is configured to divide the power of the input signal into, for example, four output signals. The output signals from the RF signal distribution unit can have similar levels of RF voltage and RF current, which can be delivered to each processing station (e.g., 151-154) of the multi-station manufacturing tool.

[0047] Figure 1C A top view of an electronic device manufacturing system 182 having four four-station manufacturing tools 188, 189, 193, and 195 is provided. The four-station manufacturing tools 188, 189, 193, and 195 may be Figure 1B 151-154 of the four-station tool. Each four-station tool includes four processing stations, each of which is configured to hold and process substrates. At the front end of the electronic device manufacturing system 182 are three front-opening unified wafer boxes (FOUPs) 183a, 183b, and 183c accessible to a front-end wafer handler robot 185, which is configured to transfer wafers between the FOUPs and a first load lock 187. The first wafer handler 170 can be positioned and configured to transfer wafers between the first load lock 187 and the four-station manufacturing tools 188 and 189. The first wafer handler 170 can also be configured to transfer the wafer to the second load lock 171, which makes the wafer available to the four-station manufacturing tools 193 and 195 via the second wafer handler 172.

[0048] In some embodiments, a four-station tool 195 (as an example) may include three sensors or sensor packages 196, 197, and 198 disposed about its outer wall. Figure 1C196-198 are shown as being vertically secured to three of the four-sided chamber of tool 195. In the illustrated embodiment, the only side without a sensor or sensor package is the side proximate wafer handler 172. Although Figure 1C 1, but similar sensors or sensor packages can be provided in the system in any one or more of the three other four-station chambers 188, 189 or 193. In addition, in some implementations, the sensor or sensor package can still be provided near the side of the wafer carrier 172. Figure 2 and Figure 3 Different possible sensor or sensor package locations are shown. A variety of other positioning of one or more sensors (including sensor packages together or separately in a sensor package) are possible, depending on, for example, the measurement, angle, desired measurement location. It should be understood that in some cases, the system controller (e.g., Figure 1B 190) can be configured to adjust the position or orientation (e.g., up, down, left, right, diagonal, azimuth, elevation) of a given sensor or sensor package.

[0049] Figure 2 and Figure 3 A manufacturing tool is schematically shown with sensors or sensor packages oriented to capture information (e.g., spatial, spectral, and / or temporal) from horizontally oriented and vertically oriented lines of sight, respectively. In some embodiments, each sensor or sensor package described can include two or more of the types described below: spatial, spectral, or temporal sensors, combined in such a way that each type is capable of obtaining information and measurements sequentially (e.g., based on a sensor type meeting or not meeting a trigger condition) or in parallel (e.g., two or more sensor types collecting information simultaneously).

[0050] Figure 2 A processing chamber 270 is shown having chamber walls 271, showerhead 272, and pedestal 273, which may be designed and constructed in any manner known in the art. Figure 2 Exemplary locations and viewing angles of exemplary sensors or sensor packages 274 are shown only. Figure 2Also shown is an exemplary sensor or sensor package 274 that is configured to capture in-situ information from the interior of the process chamber 270 via a viewport or window 276 that is designed for access by the sensor or sensor package (e.g., a sapphire rod having a diameter of 1 to 10 mm). The sensor or sensor package 274 may have a field of view defined by edges 277 and 278. As shown, the sensor or sensor package 274 and window 276 are configured to allow the sensor 274 to, for example, capture plasma images, spectral information associated therewith, and / or temporal information associated therewith, including the vertical edges of the pedestal 273. Other configurations may allow the sensor or sensor package to capture thermal images of other vertical edges in the process chamber 271, as well as other information, such as pressure, temperature, voltage, current, or other measurements of the process chamber or a particular station, depending on whether sensors capable of capturing such information are incorporated into the sensor or sensor package 274.

[0051] Figure 3 A similar process chamber 381 is shown, but with exemplary sensors or sensor packages 394 and 394' having vertical lines of sight to allow for capturing information of other features within the process chamber 381. In certain embodiments, the process chamber 381 can have chamber walls 379, showerhead 384, and base 386, all of which are designed and constructed in any manner known in the art. Figure 3 Exemplary locations and viewing angles of exemplary sensors or sensor packages 394 and 394' are shown only. Figure 3 Vertically oriented sensors or sensor packages 394 and 394' are shown, which are arranged to capture in-situ information from the interior of the processing chamber 381 via viewports or windows 399 and 399', which are designed for access by the sensor or sensor package. Sensor or sensor package 394 can have a field of view defined by edges 361 and 362, while sensor or sensor package 394' can have a field of view defined by edges 363 and 364. As shown, sensor or sensor package 394 and window 399 are arranged to allow sensor or sensor package 394 to capture plasma images, including the stem or underside of base 386. As shown, sensor or sensor package 394' and window 399' can be arranged to allow sensor or sensor package to capture, for example, plasma images, spectral information related thereto, and / or temporal information related thereto, including the edges of the stem and / or backside of showerhead 384. Other setups may allow sensors or sensor packages to capture thermal images of other edges within processing chamber 381, as well as other information such as pressure, temperature, voltage, current, or other measurements of the processing chamber or a particular station, depending on whether sensors capable of capturing this information are incorporated into sensors or sensor packages 394 and 394'.

[0052] The sensor or sensor package may be disposed externally to the manufacturing tool, but in some embodiments, it may be integrated with the chamber wall or other components or elements within the chamber, such as Fig. 9 and Fig.10 In some embodiments, a viewport or window specifically configured for the sensor can be integrated into the chamber wall. In some embodiments, the sensor or sensor package can be coupled to the interior of the manufacturing tool using an access hole disposed in or on the chamber wall to allow visual inspection of the interior of the tool. Sensors and sensor configuration

[0053] In various embodiments of the present disclosure, multiple sensors (e.g., two or more of at least one spatial sensor (e.g., a camera), at least one spectral sensor (e.g., an optical emission spectroscopy (OES) sensor), and / or at least one temporal sensor (e.g., a photodiode)) are implemented in an integrated sensor "package" or sensor system that can be used for various purposes and tasks in a semiconductor system, such as a manufacturing tool, which may include one or more processing chambers and / or stations (e.g., multi-station manufacturing tool 150, manufacturing system 182). Such sensors can be positioned or otherwise configured to sense plasma or monitor conditions in situ (meaning that the wafer does not need to be moved from the processing chamber to a separate metrology chamber for sensing or monitoring after processing) and in real time (i.e., while the process is being performed on the wafer and on a time scale comparable to the time scale of events occurring in the processing chamber). To this end, the sensors can be integrated with a viewport or other window to provide visual access to the interior of the processing chamber, as described below with respect to Figure 8 described.

[0054] In the disclosed embodiments, such a sensor package can allow monitoring equipment or personnel to solve several problems currently faced in a given semiconductor device by providing two or more of the spatial, spectral or temporal information from the corresponding type of sensor. Such spatial, spectral and / or temporal information, when combined (e.g., using a combination of possibilities, machine learning models or other algorithms) analysis, can provide a better understanding of the state of the system associated with the semiconductor device. This information can also provide control opportunities, resulting in benefits, including cost savings, improved machine availability (MA) (unit time of a machine that can be used to process, for example, a wafer), improved green to green (GtG) time period (reduced downtime) cycle) and output, and improved wafer to wafer (WtW) uniformity and wafer quality, as well as higher accuracy, stability and compatibility by combining different types of sensors that are conventionally used independently. In various use cases, different sensors can be used for primary indicators, secondary indicators and / or third indicators. In some implementations, optionally, in addition to spatial, spectral and / or temporal sensors, other sensors in the monitoring system can also obtain readings including voltage and current (VI), pressure, temperature, etc. Spatial Sensors

[0055] In various embodiments, the spatial sensor can be an image capture device (e.g., a camera or camera sensor) configured to acquire visual or optical information (e.g., image data, time-of-flight (TOF) data, or both) such that an image of radiation intensity in at least two dimensions can be provided (e.g., a 2D area array of image data with x×y pixels). The spatial sensor can provide images of features within the chamber, such as plasma features. Plasma features can include plasma boundaries, plasma locations, plasma anomalies, or parasitics. The spatial sensor can also provide images of chamber component boundaries and gas flow patterns (e.g., over time or at a given moment).

[0056] A camera sensor, as an example of a spatial sensor, can be characterized by a variety of parameters, including the number of pixels, the range of wavelengths captured, etc. In some embodiments, a camera sensor for capturing information about a plasma is capable of multispectral or hyperspectral imaging and sensing intensity values ​​for a range of the electromagnetic spectrum, including visible radiation at a wavelength that includes at least a portion of the visible spectrum. As an example, a camera sensor can be configured to sense intensity values ​​for one or more ranges (e.g., including 100 nm to 1000 nm). In some embodiments, a camera sensor can distinguish signals from wavelengths of at least a portion of the visible spectrum, at least a portion of the UV spectrum, at least a portion of the IR spectrum, or any combination thereof. For example, a camera sensor that captures IR and visible light information can provide information related to thermal processes as well as plasma processes.

[0057] As an example of any embodiment herein, the camera sensor can be constructed as a charge coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) array. In some implementations, quantum image sensors (QIS) and single photon avalanche diode (SPAD) array detectors for imaging and time of flight can be included in the CMOS array as many forms. In some embodiments, the camera sensor used herein may have at least about 5 megapixels or at least about 12 megapixels. In some embodiments, the camera sensor used herein may have as few as about 2 megapixels. Multispectral sensors can use a color filter array (CFA, color filter array) tiled across the entire detector array. Another type of array tiled across the entire detector array of the camera can be a polarizer filter array (PFA, polarizer filter array). Polarized images can achieve spatially resolved thin film stress detection and other applications.

[0058] In some embodiments, the image capture device is a linear or one-dimensional sensor or pixel array. Such a device can be configured to scan a two-dimensional field of view. The scanning direction can be substantially perpendicular to the axis of the sensor line. In some embodiments, the one-dimensional image capture device is perpendicular to the wafer or chamber assembly orientation, and is optionally configured to scan from one side of the chamber to the other side (or scan in other parts or fields of view in the chamber). In some implementations, the image capture device includes a multidimensional detector array (e.g., a two-dimensional detector array). The image capture device can be configured to access the multidimensional field of view in the processing chamber.

[0059] In some embodiments, the camera used in any embodiment herein may be configured with a shutter. In some embodiments, the camera may be configured to capture video data of the plasma in the manufacturing tool. In some embodiments, the camera may be configured to capture video information of the plasma in the manufacturing tool at a frame rate of about 30 to 120 frames per second, or about every 1 ms to 100 ms. In some embodiments, the camera does not need to provide a fast capture rate (e.g., more frequently than about every 1 ms or longer), and may have a capture rate lower than that of standard video, for example, less than about 30 frames per second, less than about 10 frames per second, or less than about 1 frame per second. In some implementations, the shutter can be used to manually capture frames.

[0060] Some manufacturing tools include static image or video displays. When a camera sensor or light conduit blocks access to the viewport from outside the tool, such a display can be used to allow a process engineer or other staff member to observe the interior. In some embodiments, a view of the interior of the chamber is provided via an image or video that is streamed electronically (e.g., using a real-time streaming protocol (RTSP: real time streaming protocol), a real-time messaging protocol (RTMP, real-time messaging protocol), a low-latency HTTP real-time streaming (HLS: HTTP live streaming), a secure reliable transport (SRT, secure reliable transport), WebRTC, etc.), optionally transmitted to a remote location via a Web application. Examples of remote locations include a factory monitoring room or facility, a smart phone, a tablet computer, and / or a desktop computer system. In some embodiments, the communication of the image or video is carried out via a network that includes a camera on a processing chamber as a node. Such a network can be wired or wireless, such as a mesh network. In some embodiments, the network can use a communication protocol that uses Wi-Fi, Bluetooth, cellular, etc.

[0061] In some embodiments, the image analysis logic may be configured to receive sensed values ​​from one or more camera sensors on a manufacturing tool. In some embodiments, the input to the image analysis logic includes pixel-by-pixel intensity values ​​as a function of observable parameters, such as wavelength, time, polarization, or any combination thereof. In some embodiments, the input data from the camera sensor is provided in the form of image data, video data, spectral values, time series data, wafer metrology data, etc. In some embodiments, the input data is filtered by wavelength, polarization, etc. In some embodiments, the analysis logic is configured to receive additional input information other than the camera sensor intensity data and work on it. Such additional input information may include metadata about the camera sensor and / or related camera components, substrate metrology information, historical information about the manufacturing tool, etc.

[0062] The analysis logic may be configured to output one or more characteristics of a plasma in a manufacturing tool, and / or a classification of a state of the manufacturing tool or a component thereof. In some cases, the analysis logic may be configured to perform an image processing routine, such as a segmentation or other edge finding routine. In some embodiments, the analysis logic may be configured to determine plasma characteristic information relative to system components using segmentation or other edge detection methods.

[0063] In addition, in some embodiments, the camera sensor analysis logic can include any of a variety of types of classifiers or models, such as deep neural networks (e.g., convolutional neural networks, autoencoders, U-Net), conventional or classical computer vision methods (e.g., edge detection), image adjustments (e.g., blurring, changing contrast), intensity thresholds, color channel thresholds, etc.

[0064] The logic may employ any of a variety of techniques for edge detection or segmentation. For example, the logic may employ a threshold-based approach, a deep learning model, etc. In some embodiments, the edge of a plasma or the boundary of a sub-region within a plasma having a defined plasma feature may be determined. Spectral Sensor

[0065] Spectral sensors provide wavelength-specific radiation intensity detection. They output a value for the radiation intensity as a function of wavelength or spectral region. Spectral sensors have multiple functions. The information they provide can distinguish or identify specific chemicals, each of which has its own specific spectrum. In some embodiments, spectral sensors can distinguish signals of wavelengths in one or more regions of the electromagnetic spectrum, including UV, visible light, and IR.

[0066] Spectral sensors can be implemented in a variety of ways. Examples of wavelength separation components include (a) a dispersive device (which separates wavelengths in a dispersive medium such as a prism), (b) a diffractive device (which separates wavelengths by diffraction using, for example, a diffraction grating), (c) a filter that is placed in front of an intensity detector so that the detector receives only the wavelength of interest.

[0067] An example of a spectral sensor detector is a linear array of optical detectors configured to detect wavelength specific intensities at different detectors in the linear array. The optical detector can be configured to provide an output of intensity as a function of wavelength. A wavelength separation component disposed in front of the linear array provides wavelength separation in one dimension (along the linear array) such that each component of the linear detector is associated with a specific wavelength or wavelength range.

[0068] Unlike spatial sensors, spectral sensors do not need to provide multi-dimensional images. In addition, for some applications, the speed at which spectral sensors acquire data does not need to be fast (e.g., a capture rate of ≥1 ms or about 1 ms to 1 s).

[0069] In some embodiments, the optical detector for the OES sensor is a linear, one-dimensional array or two-dimensional array of detector devices. For example, the OES sensor is a two-dimensional charge coupled device (2-D CCD) array, wherein the spectrally separated light components are detected by different regions of the 2-D CCD. In some implementations, a linear array of detectors can be used, which detects wavelength-specific intensities separately at different detectors in the linear array. Refraction (e.g., prism) or diffraction (e.g., grating) can be used to perform wavelength separation along the linear array in one dimension. Each element of the linear detector can be associated with a specific wavelength or wavelength range. A variety of implementations can be used for different use cases. In some embodiments, the OES sensor for capturing spectral information about plasma is capable of sensing the intensity value of electromagnetic radiation of wavelengths including the following items and / or distinguishing signals from wavelengths including the following items: at least a portion of the visible spectrum, at least a portion of the ultraviolet (UV) spectrum, at least a portion of the infrared (IR) spectrum, or any combination thereof. In some cases, the optical detector does not need to provide a fast capture rate (e.g., more frequently than about every 1 ms or more). The optical detector may be operated at a frame rate of, for example, approximately every 1 ms to 1 second.

[0070] Figure 4 4 is a diagram of an exemplary OES sensor 400 according to some embodiments. The OES sensor 400 can be coupled to one or more transmission fibers 402. The transmission fibers 402 can collect spectral information from a corresponding process, such as at a process chamber or a station within a process chamber. For example, the transmission fibers 402 can collect spectral information from a multi-station process chamber (e.g., reference 400) via light collection optics 403 during a process. Figure 1B and 1C In some embodiments, each of the four optical fibers 402 is associated with its own processing station and collects optical signals emitted from its processing station. The transmission optical fibers 402 can be grouped into bundles and received by the OES sensor 400. Figure 4 As shown, as an exemplary implementation, the OES sensor 400 can receive spatial information from four stations of a four-station processing chamber via four transmission optical fibers 402 .

[0071] The OES sensor 400 may also include dispersive optics, such as one or more mirrors 404, some or all of which may have optically dispersive properties by being coated at different depths so that different wavelengths have different penetration lengths, thereby allowing the mirrors to reflect light of different wavelengths. The OES sensor 400 may include an image sensor 406 that detects spatial information, which may include spectrally separated light components.

[0072] Various other implementations of the OES sensor are possible, including, for example, mounting the OES sensor directly at the viewport, removing at least some of the transmission optical fibers 402, removing some or all of the light collection optical devices 403, switching between different optical inputs (e.g., fiber switches, microelectromechanical systems (MEMS) mirrors), using different types of dispersive optical device configurations (e.g., gratings, prisms, computed tomography (CT), echelle gratings), using spectral filters (e.g., filter wheels, or in some cases etalons or adjustable liquid crystals), using time-domain spectroscopy (e.g., in some cases, using a spectrum analyzer).

[0073] In some embodiments, a spectral reflectometer device can be used as a spectral sensor. The reflectometer device may include a light source for irradiating the surface to be treated, and an optical detector. The optical detector may include one or more photodetectors. An optical fiber cable may be connected to the spectral reflectometer device. The optical cable may include a transmission fiber and a receiving fiber, wherein each receiving fiber may be connected to a corresponding respective photodetector. In some cases, multiple receiving fibers may be connected to the same photodetector.

[0074] In some embodiments, position-sensitive spectroscopic ellipsometry can be used. In some embodiments, a reflectometer with polarization control can be used particularly with highly polarizing structures.

[0075] Note that while spatial information may be based on the intensity collected by each pixel of a sensor (eg, a camera), spectral sensor intensity counts may be collected for a larger area. Time sensor

[0076] The time sensor can detect and measure the time characteristics of pulsed light or other light bursts (e.g., arcs formed in plasma systems). In the environment of, for example, a processing chamber, the time sensor can be used to very quickly identify the occurrence of process changes due to its fast reaction time (e.g., in the range of about 100 ms or picoseconds (ps) to microseconds (μs), such as 10ps to 100ms). By measuring the time characteristics (e.g., the time change of light emission as an electrical signal), the time sensor can measure and / or provide signal data of such process changes. Examples of time sensors include photodiode optical sensors, micro-photodiodes, phototransistors, photoelectric cells, pulse characterization sensors, photomultiplier tubes, solid-state photomultipliers, or other fast one-dimensional radiation (usually but not necessarily limited to visible light) sensors suitable for use with the type of processing chamber or multi-station manufacturing tool (e.g., 150, 182). The time sensor can be configured to acquire single point data in the spatial domain. In some embodiments, a filter or lens can be used in conjunction with the time sensor.

[0077] In embodiments using photodiodes or arrays thereof, for example, the photodiode may include at least one semiconductor junction (pn, np, pnp, npn, ppn, etc.), wherein light falling on the junction may result in the formation of electron-hole pairs. As known to those of ordinary skill in the relevant art, the electron-hole pairs may migrate to different parts of the junction, thereby generating a voltage and / or current (if connected to a circuit). In this manner, even substantially instantaneous light or bursts of light or photons (e.g., on the scale of the aforementioned ns to μs range) may be detected. Furthermore, there is no need to provide wavelength discrimination or to obtain multi-dimensional images.

[0078] Changes in the architecture of the photodiode or time sensor, such as an increase in the width of the depletion region of the photodiode, the shape of the mesa of the photodiode, the number and depth of n or p layers used, the dopants used, the junctions used, or the amount of bias applied across the junction, may result in a change in the detectability of a process variation, or a change in the indication of a process variation (e.g., a light emitting diode turns on based on a process variation of not emitting light, or if visible light falls below a certain level), as required or desired for a particular application (which is described in the present disclosure as a combined sensor group, which may be one or more of many sensor packages, as will be discussed further below), for example, by a change in quantum yield (ratio of emitted photons to absorbed photons), or sensitivity (ratio of current to light irradiance), or other metrics. The range of radiation detected may vary depending on the materials used, as known to those of ordinary skill in the relevant art. Examples of detected wavelengths may include 300 nm to 1200 nm. In some variations, the photodiode may be a PN photodiode (having a P-type layer and an N-type layer), a PIN photodiode (having an intrinsic "I-region"), a Schottky photodiode, or an avalanche photodiode. Arrays of the foregoing may also be used.

[0079] In some embodiments, a given processing chamber, multi-station manufacturing tool, or processing chamber thereof may include one or more spatial sensors, one or more spectral sensors, one or more temporal sensors, or a combination thereof. The sensors may be disposed inside or outside the chamber. The sensors may be attached to an interior or exterior surface of the chamber, or a platform, or other component, and may be configured to communicate signals or data, for example, with controller 190 or subsystem 191.

[0080] In some embodiments, one or more additional or alternative sensors may be used, such as sensors configured to detect or measure other parameters within or associated with the processing chamber, such as VI, pressure, or temperature. Irradiation source

[0081] In some cases, the manufacturing tool may include an illumination system or one or more illumination sources configured to illuminate all or one or more portions of the interior of the manufacturing tool. In some embodiments, the illumination system may be configured to allow the camera to obtain an illumination image when the plasma is turned off (e.g., outside of operation or between pulses). It should be noted that in some cases, the illumination system is not used, and illumination from the plasma itself is used. In some embodiments, the illumination system uses one or more light emitting diodes (LEDs) or other light sources. The light source may be monochromatic, polychromatic with discrete emission wavelengths, or wide spectrum. The light source may be continuously activated, pulsed synchronously with the operation of a spatial sensor (e.g., one or more camera shutters), pulsed asynchronously with the operation of a spatial sensor (e.g., one or more camera shutters), or pulsed synchronously with other process parameters (e.g., RF generators or gas delivery valves). In some cases, the pulses of the light source may facilitate detail extraction and provide an ongoing process within the manufacturing tool. For example, the light may be pulsed synchronously with optical sensing (e.g., camera shutter closed and / or opened to capture image data, such as optical images or videos of visible effects and events, such as plasma behavior and characteristics), while spectral sensing does not occur during pulsed illumination so as not to corrupt the spectral signal captured between optical imaging. However, pulsing of the light source asynchronously with optical sensing (e.g., operation of the camera shutter) may allow for the capture of other visible information, such as the location of the plasma arc, which may be better detected and located in the dark than when the chamber is illuminated. In some implementations, a temporal sensor may additionally or alternatively be used to detect the occurrence and presence of an arc when the chamber is not pulsed. It would be beneficial to detect arcs or other artifacts or anomalies that may damage components, which may be more difficult to discern or locate when the chamber is pulsed. These exemplary illumination pulse schemes increase the efficiency and amount of information extracted from the ongoing process, which is particularly useful for short-term (e.g., above-mentioned on the ms or ps level) events and process variations occurring within a manufacturing tool. In other implementations, multiple light sources are employed at different locations indoors or outdoors. These multiple light sources can be powered continuously or sequentially, and the timing can be controlled so that structured illumination can be used to construct super-resolution images of indoor features. In some embodiments, one or more notches or bandpass filters are provided in front of the light sources to produce effects that can support analysis (e.g., identifying specific chemicals by their emission spectra). In some implementations, strobe lighting and other consistent structured lighting (e.g., using periodic bursts of illumination at, for example, 10, 30, 60 Hz) can be used to provide additional discrete visual information at a consistent frequency.

[0082] Additional uses of the illumination source include illuminating the internal components of the processing chamber, which would otherwise be dark. This allows the spatial sensor to image such components. The images of these components can provide a frame of reference for locating other features in the chamber, such as the gap between the pedestal and the showerhead, or with respect to the chamber walls. It can also allow images of adjacent stations in a multi-station chamber. The location of the adjacent stations can also provide a reference frame for key features to be imaged in the station of interest.

[0083] The illumination source can also be used as a "health check", for example, to verify or confirm that systems and subsystems such as spatial, spectral and temporal sensors are functioning or operating properly and correctly, to track system drift over time, and to determine whether the sensor or illumination source is the source of system decay. The illumination source can also be used as a calibration mechanism for the sensor. For example, strobed illumination at a specific wavelength can support calibration of a spectral OES sensor. In some cases, OES calibration can be performed by software for specific wavelengths of light to correct for any drift. In some methods, the illumination source can also be used to determine the sensitivity of an optical detector (e.g., a photodetector), and to adjust any threshold drift to minimize false positives in the event that the photodiode itself drifts over time. method

[0084] Figure 5 is a flow chart showing a method 500 for determining a state of semiconductor manufacturing equipment according to some embodiments. One or more functions of the method 500 may be performed or caused by a computerized device or system. Figure 5 The structure of the functions shown in one or more of the steps shown may include hardware and / or software components of such a computerized device or system, such as a controller device, a computerized system, or a computer-readable device including a storage medium that stores computer-readable and / or computer-executable instructions that are configured to cause at least one processor device or computerized device to perform operations when executed by a processor device. A controller may be an example of a computerized device or system. A subsystem (e.g., 191) may be an example of a computerized device or system. A processing chamber may be another example of a computerized device or system. Exemplary components of a processing chamber and a controller are respectively described in Figure 1A and Figure 1B and Fig.13 , and described in more detail elsewhere herein.

[0085] It should also be noted that the operations of method 500 may be performed in any suitable order, not necessarily in the order in which they are performed. Figure 5 In addition, method 500 may include Figure 5More or less than the operations depicted may be performed to determine the status of semiconductor manufacturing equipment.

[0086] As described above, when implementing the present disclosure, for example, using method 500, an opportunity to control a manufacturing tool arises. Management of process parameters and control of a manufacturing tool can be implemented in three stages: (1) detection, (2) understanding, and (3) control (e.g., changing a set point, desired value, or target value). An integrated sensor group, for example, regarding Figures 8 to 10 The described multi-sensor system can implement method 500, which can enhance, enhance or replace conventional plasma monitoring, film growth, and mechanical or wafer movement sensor equipment. Current needs expose the inadequacy of monitoring chamber conditions and expected mechanical motion. To this end, the multi-sensor system can enable active monitoring and control of chamber conditions based on plasma, parasitic effects, HCD, particles, process excursions, and mechanical changes around the gap and / or pedestal or showerhead (as well as other use cases as described above), as well as support real-time analysis capabilities in an integrated group.

[0087] At frame 510, method 500 may include determining a signal group from a plurality of sensors of semiconductor manufacturing equipment to monitor during a process performed by the semiconductor manufacturing equipment. In some embodiments, frame 510 may involve setting data capture parameters for a particular process operation. These parameters may involve selecting two or more sensors from a plurality of sensors available on a hardware tool. Sensors may be selected because they are collectively capable of detecting and / or characterizing a particular event or condition in a processing chamber. Some of these events or conditions are described herein as use cases and applications. In some implementations, frame 510 may be performed during the setting period, and then no longer performed, while other operations in the process are repeated.

[0088] In some embodiments, the plurality of sensors include sensors of multiple types. In some implementations, the plurality of sensors include one or more spatial sensors, one or more spectral sensors, and one or more temporal sensors. An example of a spatial sensor may be a camera.

[0089] According to some implementations, the one or more spatial sensors may include one or more cameras or one or more camera arrays that are aligned toward the interior of the semiconductor manufacturing equipment and are configured to obtain image data related to the interior before, during, and / or after the process. In some implementations, the one or more spatial sensors may be configured to obtain data related to characteristics of the plasma present in the interior during the process. Such characteristics of the plasma may include intensity, location, whether it is parasitic, anomalies related thereto, etc.

[0090] An example of a spectral sensor may be an OES sensor. An OES sensor may be configured to detect one or more wavelengths of interest emitted by one or more substances produced during a process.

[0091] An example of a time sensor may be a photodiode. The photodiode may be configured to detect changes in light intensity occurring within a time period of less than 1 ms, including the nanosecond and microsecond range.

[0092] In some embodiments, these spatial, spectral, and temporal sensors may be a sensor group having multiple sensor types, which are combined into an integrated package (or device) that can be incorporated into a viewport (e.g., 900) or other window of a semiconductor manufacturing equipment (e.g., multi-sensor manufacturing tool 800) and described below. Figure 8 The device may be operated according to the description provided. Although sensor types can be combined into one device, not all of them may be used. In some cases, a signal group may be provided by all three sensor types. However, in some cases, a signal group may be provided by two types of sensors. For example, depending on the application, only a camera and an OES sensor may be used, or a camera and a photodiode may be used. In some implementations, the multiple sensors may also include other sensor types, such as VI, pressure, or temperature sensors, to provide additional measurements.

[0093] In some implementations, the plurality of sensors may include two types of one or more spatial sensors, one or more spectral sensors, or one or more temporal sensors. For example, in these implementations, the plurality of sensors may be only cameras and OES sensors, without photodiodes.

[0094] At block 520, method 500 may include obtaining measurements associated with the signal set from the selected plurality of sensors. The measurements may be obtained during a process to be performed and may be obtained by at least one spatial sensor, at least one spectral sensor, at least one temporal sensor, or a combination thereof, wherein the combination includes at least two types—spatial and spectral; spectral and temporal; spatial and temporal; or spatial, spectral, and temporal. The sensors may be used in series or in parallel simultaneously to collect corresponding signals from a processing environment (e.g., a processing chamber) and as primary, secondary, and / or tertiary indicators.

[0095] In some cases, the process performed by the semiconductor manufacturing equipment may include detecting an undesirable amount of one or more substances inside the semiconductor manufacturing equipment. Herein, the sensor types used to detect the undesirable amount of substances may include at least one spatial sensor and at least one spectral sensor. In some cases, the process performed by the semiconductor manufacturing equipment may include endpoint detection of chamber cleaning inside the semiconductor manufacturing equipment. Herein, the sensor types used to detect the endpoint of chamber cleaning may include at least one spatial sensor and at least one spectral sensor. In other cases, the process performed by the semiconductor manufacturing equipment may include any one or more use cases described elsewhere herein.

[0096] At block 530, method 500 may include determining an indication of a state of the semiconductor manufacturing equipment based on a combination of data generated from measurements associated with the signal group (e.g., the measurements from block 520). Advantageously, the combination of data more accurately characterizes the state of the semiconductor manufacturing equipment than a signal from only one of the multiple sensors in a conventional monitoring system.

[0097] In some embodiments, method 500 may include adjusting control parameters of semiconductor manufacturing equipment based on the state of the semiconductor manufacturing equipment. Various changes may be made to the control parameters to implement changes to the processing environment, such as the amount of gas species or their rate of flow into a gas-specific pipeline, clamping and unclamping the substrate from the pedestal (e.g., by varying an electrical signal of a voltage applied to an electrostatic pedestal), the position of the substrate, the endpoint of chamber cleaning, the timing of plasma ignition, the opening and closing of valves, the shutdown of the manufacturing process (e.g., based on excessive HCD) to prevent hardware damage, changing lighting, providing alarms, providing process-related data for manual review (e.g., adjusting process recipes, equipment design, etc.), and other adjustments to implement the various use cases described above.

[0098] Otherwise, the status of the semiconductor manufacturing equipment may be confirmed, stored, or provided (eg, displayed, recorded in a computer file, transmitted to another computer device) without changing the control parameters.

[0099] In some embodiments, a machine learning model may be implemented to improve the accuracy of the output. Determining an indication of the state of the semiconductor manufacturing equipment (block 530) may include performing a classification task or a regression task on at least a portion of the data combination using a machine learning model to obtain measurements associated with the signal group configured to be performed. For example, in some cases, a regression model may be implemented using labeled data (e.g., previous measurements and correct labels) to enhance the predictability and accuracy of the measurements (compared to measurements from individual sensors). In other cases, when the signal-to-noise ratio of only a single sensor is reduced, the classification probability (e.g., using a logic model) may be improved for the primary sensor (e.g., whether the detected plasma is parasitic, whether there is HF contamination, whether there is film growth). The machine learning model used may be triggered simultaneously throughout the process, or may be triggered at specific intervals depending on the use case.

[0100] Fig. 6A and Figure 6B is a flow chart showing methods 600 and 650 for determining a state of semiconductor manufacturing equipment according to some embodiments. One or more of the functions of methods 600 and 650 may be performed or caused by a computerized device or system. Fig. 6A and Figure 6B The structure of the functions shown in one or more of the steps shown may include hardware and / or software components of such a computerized device or system, such as a controller device, a computerized system, or a computer-readable device, which includes a storage medium storing computer-readable and / or computer-executable instructions, which are configured to cause at least one processor device or computerized device to perform operations when executed by a processor device. A controller can be an example of a computerized device or system. A subsystem (e.g., 191) can be an example of a computerized device or system. A processing chamber can be another example of a computerized device or system. Exemplary components of a processing chamber and a controller are respectively described in Figure 1A and Figure 1B ,and Fig.13 , and described in more detail elsewhere herein.

[0101] It should also be noted that the operations of methods 600 and 650 may be performed in any suitable order, not necessarily in the order in which they are performed. Fig. 6A and 6B In addition, methods 600 and 650 may include Fig. 6A and Figure 6B The operations depicted may be performed in greater or lesser amounts to determine the status of semiconductor manufacturing equipment.

[0102] refer to Fig. 6AAt block 610, method 600 may include obtaining a measurement signal via a first sensor. In some embodiments, the first sensor may be a spatial sensor (e.g., a camera), a spectral sensor (e.g., an OES sensor), or a temporal sensor (e.g., a photodiode).

[0103] At block 620, method 600 may include determining whether a trigger condition corresponding to the first sensor is met. The trigger condition may be a condition that triggers different operations, such as collecting information from different sensors, processing sensor signals from one or more sensors in a specific manner, determining whether to make real-time adjustments to the process, etc. In some examples, the trigger condition is a specific threshold of sensor measurements obtained during a process performed by semiconductor manufacturing equipment. An example may be a measurement below a specified threshold (determined for each type of sensor or application or use case), which may indicate uncertainty as to whether the measurement is a valid signal or noise that should be ignored. The specific example discussed herein may be, for example, a plasma condition (e.g., intensity) within a processing chamber (determined by in-situ monitoring of a sensor type (e.g., camera) described throughout this disclosure) during a process performed by semiconductor manufacturing equipment.

[0104] Another example of a trigger condition in a particular situation may be an intensity shift from an OES sensor, as discussed elsewhere above. As a further example, an OES sensor may detect gas composition and its shift (below or above a certain threshold), or the presence of an undesirable substance (e.g., oxygen). These measurements may be primary indications, for example, for determining a gas burst or chamber clean endpoint. Secondary or tertiary confirmation by at least one other type of sensor may be useful for confirmation of the primary indication by the above sensor examples (e.g., visual confirmation or additional information from another angle).

[0105] To this end, if the trigger condition is met, method 600 may proceed to block 630, which may include obtaining a measurement signal via a second sensor. The second sensor may be a different sensor type than the first sensor.

[0106] For example, if the first sensor is a spectral sensor, the second sensor can be a spatial sensor or a temporal sensor. A spectral sensor (e.g., an OES sensor) can be used primarily to determine the chemical state inside a processing chamber, for example, to determine how the chamber cleaning is progressing. However, there may be slow cleaning sections, or the response of the spectral sensor may be weak and sufficient to determine whether trace amounts remain on the film to be cleaned. Such uncertainty in the response can be an example of a trigger condition (box 620). A camera aimed at the location of the film trace amount can provide a secondary confirmation of the cleaning endpoint. Therefore, in such a multi-sensor synthesis method, the OES sensor is the primary indicator in such a situation, and the camera is a secondary indicator for visual confirmation of the chemically uncertain parts of the system.

[0107] However, in some implementations, the first sensor and the second sensor can be the same type of sensor. For example, the first sensor can be a camera at a wall of the processing chamber pointing toward the top of the pedestal, and the second sensor can be a camera at an opposite wall pointing toward the bottom of the pedestal. In this way, a more complete view of the state inside the processing chamber can be obtained.

[0108] Otherwise, if the triggering condition at block 620 is not met, the method 600 may return to block 610 .

[0109] At block 640, method 600 may include determining whether a trigger condition corresponding to the second sensor is met. An example of a trigger condition may be a measurement result obtained by the second sensor that meets or exceeds a specified threshold (determined for each type of sensor or application or use case), which may indicate the validity of the measurement result.

[0110] If the trigger condition at block 640 is met, the measurement signal obtained via the second sensor can be used as a secondary confirmation of the measurement signal obtained via the first sensor (e.g., the weak measurement of the first sensor is verified). In the case where the second sensor can obtain a measurement result of interest (e.g., the plasma intensity can be derived from visual information or spectral measurements), the measurement signals obtained via the first and second sensors can both be used as separate measurements, or the measurement signal obtained via the second sensor can be used as the measurement result in some cases. If the trigger condition at block 640 is not met, the method 600 can return to block 610. However, in some implementations, the method 600 can return to block 630 to obtain additional measurement signals from the second sensor.

[0111] refer to Figure 6BIn some embodiments, if the trigger condition is not met at block 620, the method may proceed to block 622, which may include determining whether a change is desired in a process performed by the semiconductor manufacturing equipment. In these embodiments, the process may be changed rather than obtaining an additional confirmation measurement signal via the second sensor (block 630) because, for example, obtaining a measurement signal via the first sensor (block 610) when the trigger condition (block 620) is not met still provides sufficient information to conclude that an abnormal or undesirable condition exists in the state of the processing chamber or process that requires some adjustment to the process to eliminate, such as parasitic plasma, gas leaks, gas composition shifts, gas bursts, arcs, etc. In some cases, it may be determined that a change in the process is required based on other factors that are unrelated to the internal conditions of the processing chamber, such as the time taken by the process.

[0112] If a change is required, at block 624, the method may include changing one or more parameters of the process, and the method may return to block 610 to continue acquiring the measurement signal via the first sensor. By changing the parameters of the process, the process may be changed, paused, or aborted. According to various applications, an example of a parameter may be a power state of a processing chamber or a manufacturing tool, wherein changing the power state may, for example, at least partially shut down the manufacturing tool so that the process may be paused or stopped. Another example may be a change in gas flow. The amount of undesirable gas (e.g., gas burst or gas leak) or the amount of undesirable species in the processing chamber may be reduced, or failure to introduce a sufficient amount of desired gas may result in an increase in the desired gas. Another example may be the amount of etching performed during chamber cleaning. If the trigger condition indicates that a large amount of accumulation has occurred, for example, on the chamber wall, a larger amount of etching may be performed without damaging the chamber wall. Other examples of parameter (e.g., temperature, voltage, plasma timing, recipe conditions, other operating parameters) changes will become apparent in view of the present disclosure.

[0113] If the change is not desired (eg, because the state of the chamber or process may benefit from additional confirmation from the second indicator), the method may return from block 622 to block 610 to continue acquiring measurement signals via the first sensor.

[0114] In some embodiments, if the trigger condition is not met at block 640, the method may proceed to block 642, which may include determining whether a change is desired to a process performed by the semiconductor manufacturing equipment. Herein, the determination may be based on a measurement signal obtained via the second sensor and / or the first sensor.

[0115] If a change is desired, then at block 644, one or more parameters of the process may be changed, similar to block 624, and the method may return to block 610 to continue acquiring measurement signals via the first sensor. In some embodiments, the method may return to block 630 to acquire measurement signals via a second sensor, for example, for further confirmation.

[0116] If no change is expected, the method may return from block 642 to block 610 to continue acquiring measurement signals via the first sensor. In some embodiments, the method may return to block 630 to acquire measurement signals via a second sensor, such as for further confirmation.

[0117] Methods 600 and 650 represent the continuous collection and / or processing of information from two or more types of sensors (e.g., from a process chamber, station, or other part of a manufacturing tool). The collection and / or processing of information can also be performed simultaneously in parallel, as will be described below. Fig. 7A and 7B Described in the content.

[0118] Fig. 7A and Figure 7B is a flow chart showing methods 700 and 750 for determining a state of semiconductor manufacturing equipment according to some embodiments. One or more of the functions of methods 700 and 750 may be performed or caused by a computerized device or system. Fig. 7A and 7B The structure of the functions shown in one or more of the steps shown may include hardware and / or software components of such a computerized device or system, such as a controller device, a computerized system, or a computer-readable device including a storage medium that stores computer-readable and / or computer-executable instructions that are configured to cause at least one processor device or computerized device to perform operations when executed by a processor device. A controller may be an example of a computerized device or system. A subsystem (e.g., 191) may be an example of a computerized device or system. A processing chamber may be another example of a computerized device or system. Exemplary components of a processing chamber and a controller are respectively described in Figure 1A and Figure 1B and Fig.13 and described in more detail elsewhere herein.

[0119] It should also be noted that the operations of methods 700 and 750 may be performed in any suitable order, not necessarily Fig. 7A and 7B In addition, methods 700 and 750 may include Fig. 7A and 7B More or less than the operations depicted may be performed to determine the status of semiconductor manufacturing equipment.

[0120] refer to Fig. 7A , at box 710, method 700 may include obtaining a measurement signal via a plurality of sensors, which may include at least a first sensor and a second sensor. In some embodiments, the first sensor may be a spatial sensor (e.g., a camera), a spectral sensor (e.g., an OES sensor), or a temporal sensor (e.g., a photodiode). In some embodiments, the second sensor may be a spatial sensor (e.g., a camera), a spectral sensor (e.g., an OES sensor), or a temporal sensor (e.g., a photodiode). In some implementations, the second sensor may be a different type from the first sensor. For example, the first sensor may be a spectral sensor, and the second sensor may be a spatial sensor.

[0121] In some embodiments, the measurement signal can be obtained via at least one third sensor. In some embodiments, the first sensor can be a spatial sensor (e.g., a camera), a spectral sensor (e.g., an OES sensor), or a time sensor (e.g., a photodiode). In some implementations, the first sensor, the second sensor, and the third sensor can each be a different type. For example, the first sensor can be a spectral sensor, the second sensor can be a spatial sensor, and the third sensor can be a time sensor. However, in some other embodiments, different combinations of sensor types can be utilized, such as two spatial sensors and one spectral sensor.

[0122] In some embodiments, one or more additional sensors may be utilized. Such additional sensors may include sensors that can obtain readings of voltage and current (VI), pressure, temperature, etc.

[0123] At block 720, method 700 may include determining whether trigger conditions corresponding to the plurality of sensors are met. As described above, the trigger conditions may be conditions that trigger different operations (e.g., collecting information from different sensors, processing sensor signals from one or more sensors in a particular manner, determining whether to make real-time adjustments to the process, etc.).

[0124] If the trigger condition is met at block 720, the method 700 may stop. In some cases, the processing chamber or manufacturing tool may pause or stop operation in view of meeting the trigger condition. For example, this may be appropriate if continued operation may damage the tool or the wafer being processed in the tool. In some cases, the processing chamber or manufacturing tool may perform a different process or continue the process.

[0125] If the trigger condition is not met at block 720 , the method 700 may return to block 710 to obtain measurement signals via a plurality of sensors.

[0126] However, in some embodiments, reference is now made to Figure 7B If the trigger condition is not met at block 720, method 700 may include determining whether a change to a process performed by the semiconductor manufacturing equipment is desired at block 722. Examples of determining the desirability of a change to a process are described above with reference to block 622.

[0127] If a change is desired, at block 724, the method may include changing one or more parameters of the process, and the method may return to block 710 to continue acquiring measurement signals via the plurality of sensors. Examples of parameters are described above with reference to block 624. A process may be changed, paused, or aborted by changing a parameter of the process.

[0128] If no changes are desired, the method may return from block 722 to block 710 to continue acquiring measurement signals via the plurality of sensors.

[0129] In these ways, two or more sensors of different types may acquire information from a process chamber, station, or other portion of a manufacturing tool collectively rather than sequentially (eg, depending on a trigger condition). Device Multi-sensor capability

[0130] Figure 8 800 (e.g., multi-station manufacturing tool 150 or 182) according to some embodiments, the multi-sensor manufacturing tool 800 implements a combined sensor group (or "suite") that utilizes multiple types of sensors for manufacturing equipment 850. In some embodiments, the multi-sensor manufacturing tool 800 may include a combined sensor group 810 associated therewith, the combined sensor group 810 being, for example, disposed internally or externally to or connected to a controller (e.g., system controller 190) or a subsystem (e.g., subsystem 191) of the multi-sensor manufacturing tool 800. Each sensor may send or receive a signal from a controller or subsystem based on information obtained from, for example, the manufacturing equipment 850. In some embodiments, the combined sensor group 810 may be encapsulated in a sensor package 810 that engages a viewport or window of a processing chamber (e.g., a processing chamber of the multi-station manufacturing tool 150 or 182). According to different implementations, one or more sensor packages may be used for each processing chamber or station.

[0131] In some implementations, the multi-sensor manufacturing tool 800 may include multiple sensor types, particularly two or more of one or more spatial sensors 812, one or more spectral sensors 814, or one or more temporal sensors 816. These sensor types may be packaged into a sensor group 810 and may not be customized for a particular application (which may be used with a given sensor type), but the sensor group 810 may be combined in different ways, depending on the implementation. Such an "integrated" approach may improve availability (and reduce customization) for multiple applications and thereby reduce production costs, while improving machine availability (MA, the unit time a machine can be used to process, for example, a wafer), improved green-to-green (GtG) time (reduced downtime or "red" time) and yield, as well as improved wafer-to-wafer (WtW) uniformity and wafer quality, as well as higher accuracy, stability, and matching by combining different types of sensors (which are conventionally used independently).

[0132] In various embodiments, the sensor package 810 may include one or more physical hardware features that enable the ability to use multiple sensors in combination as described herein. In some embodiments, the sensor package 810 can be contained by a housing having at least one major dimension (e.g., a major axis, a diagonal axis, a height, a diameter, a radius, or other geometry-related feature) that is no greater than about 10 inches. Other dimensions are possible depending on the needs and capabilities of a particular sensor package that may have the various components described herein. Examples of housings may be as described below with respect to Fig.10 The housing 1001 discussed. The housing can be shaped to accommodate at least one spatial sensor 812, at least one spectral sensor 814, at least one temporal sensor 816, or, depending on the variation, a combination of at least two of the aforementioned types of sensors. In some implementations, the housing can include at least one interface that allows hardware to be attached to a wall of the processing chamber. The interface can be a physical interface that allows adhesion, or interlocking, or other mechanisms or structures involving attachment to the wall. In some cases, the wall can be part of a thermal barrier (e.g., Fig.10 1002, 1003), and vice versa. In some implementations, the housing may also include a wired or wireless communication interface and / or a power interface (e.g., a cable), or a structure that may point to or accommodate such a communication or power interface. The sensor package 810 may include a structure for fixing some or all of the above components, such as a printed circuit board (PCB) or a printed wiring board (PWB). In some implementations, the sensor package 810 may also include an illumination source (e.g., at least one LED), on-board processing capabilities, filters (e.g., optical filters), and / or related to the following Fig. 9 and 10 Other components mentioned.

[0133] In a particular implementation, two of the aforementioned sensor types may be implemented using a multi-sensor manufacturing tool 800. In a particular implementation, all three of the aforementioned sensor types may be implemented using a multi-sensor manufacturing tool 800. Examples of spatial sensors 812 may include, but are not limited to, a camera or another image capture device configured to acquire visual or optical information. Examples of spectral sensors 814 may include, but are not limited to, an OES sensor or a spectral reflectometer device. Examples of time sensors 816 may include, but are not limited to, a photodiode or another type of light sensor. In some implementations, in addition to the multiple sensor types described above, one or more other sensors (e.g., VI, pressure, temperature) may also be included.

[0134] In various embodiments, the sensor group 810 can be used to obtain spatial, spectral and / or temporal information, which when analyzed can provide a better understanding of the system status (e.g., observation of the process of the multi-sensor manufacturing tool 800), as well as provide opportunities for control and adjustment or management of process parameters. Two or more types of sensors (including use cases with three types of sensors and / or additional types of sensors (e.g., temperature sensors)) can be considered as primary sensors or indicators, secondary sensors or indicators, tertiary sensors or indicators, fourth sensors or indicators, etc., respectively.

[0135] In this context, the primary sensor can collect data that is used to detect or measure or otherwise obtain a signal (e.g., from radiation) from the process, where the signal can independently determine the presence or change of the process without any other sensor (e.g., secondary, third sensor). The secondary sensor can collect the following data: the data can help obtain signals that the primary sensor may not be able to fully or accurately obtain. For example, in some cases, the OES sensor can capture the first part of the process environment, but the camera can capture the second part. Using illustrative exemplary numbers, one sensor is able to capture 70% of the information, and another sensor is able to capture the remaining 30% of the information by capturing additional viewing angles or additional types of signals. For example, although the OES sensor as the primary sensor can detect the film left on the back of the base, the camera can visually detect the film that may exist on the side or top of the base. In the absence of a camera, the OES sensor may not be able to detect the film on the top of the base. As another example, the OES sensor may not be able to detect the presence of any film because the amount of tracking material left on the wall is not enough, which indicates to the conventional manufacturing tool that no cleaning needs to be performed. However, the camera is able to sense the film and provide an indication that the film is detected. As another example, the OES sensor can detect a sufficient amount of material on the wall, but is not sure whether the film exists. The camera can provide a secondary indication to provide confirmation, for example, (i) only when it is determined that the amount of film detected is not enough, which can be a trigger condition for starting a spatial sensor such as a camera to confirm the existence of the film, or (ii) in parallel and simultaneously with the OES sensor to provide a powerful combined indication. In these ways, the primary sensor and the secondary sensor can be used together to obtain a more complete state of the process or its by-products. In this context, the third sensor can be used as a redundant sensor to serve as a backup sensor, for example, if the primary sensor or the secondary sensor fails. In the case where the first or second sensor does not work, the third sensor can provide at least partial information. The third sensor can only be used as a temporary solution, for example, until the system is offline for maintenance. In some cases, all three types of sensors can be used to obtain primary, secondary and tertiary indications.

[0136] In some implementations, combining measurements from the primary and secondary sensors can improve the accuracy and / or precision of the measurements, or obtain additional relevant information (e.g., using temporal sensing to detect chamber arcing, and using spatial or spectral sensing to determine location or source). The third or other sensor can be a redundant sensor, used as a backup sensor in the event that one of the primary or secondary sensors fails, and is not used for measurements. However, in some implementations, measurement data from the third or other sensor can be used in conjunction with the primary and / or secondary sensors to further improve the accuracy and / or precision of the measurements of interest.

[0137] A particular embodiment of the disclosed system (e.g., multi-sensor manufacturing tool 800) combines one or more cameras for providing spatial information in the visible and IR spectra, an optical emission spectrometer (OES) for spectral sensing in the UV, visible, and IR spectra, a photodiode for rapid temporal sensing of events, and one or more illumination sources for improved detection capabilities. The data obtained from these sensors is combined to produce a stronger signal that allows for automatic information response in the system, thereby addressing one of the many use cases described below.

[0138] In some cases, a machine learning model or algorithm can be used to improve the accuracy of the output. For example, in some cases, a regression model can be implemented using labeled data (e.g., previous measurements and correct labels) to enhance the predictability and accuracy of the measurements (compared to measurements from individual sensors). In other cases, when the signal-to-noise ratio of only a single sensor is reduced, the classification probability (e.g., using a logistic model) can be improved (e.g., using a logistic model) for the primary sensor (e.g., whether the detected plasma is parasitic, whether there is HF contamination, whether there is film growth). The machine learning model used can be triggered simultaneously throughout the process, or can be triggered at specific intervals depending on the use case. Viewport Widget

[0139] Fig. 9 A cross-sectional view of an exemplary viewport 900 of a manufacturing tool (e.g., multi-sensor manufacturing tool 800) is shown with a sensor group (e.g., 810) associated therewith, according to some embodiments. The sensor group includes multiple types of sensors, such as two or more of the aforementioned spatial, spectral, or temporal sensors (or additional sensors, such as VI, temperature, pressure, etc.). Viewport 900 may include spatial, spectral, or temporal sensors (three types of sensors) and be integrated into a process chamber of a station of a manufacturing tool.

[0140] In some embodiments, viewport 900 may include a spatial sensor, such as camera 901. Camera 901 may be positioned for optimal view of the process chamber to monitor the process being performed therein. In some implementations, a manufacturing tool, a given station, or a process chamber may include multiple cameras, and Fig. 9 Camera 901 shown may be one of a plurality of cameras. Similarly, a manufacturing tool, a given station, or a processing chamber may include a plurality of viewports, and viewport 900 may be one of the plurality of viewports. Depending on the implementation of the manufacturing tool, camera 901 or a viewport having camera 901 may be configured in a plurality of ways. Figure 23, or in other locations on the chamber wall. Camera 901 may have a cover or shutter that can be kept open or closed to protect the lens. In some embodiments, camera 901 can be moved in a direction so that camera 901 can obtain optical information from different perspectives (e.g., facing the surface of the nozzle or base). The movement can be a translational movement along the vertical 910, horizontal 911, or diagonal 912 (at a certain angle relative to the vertical or horizontal), which are relative to each other and do not have an absolute direction orientation. In some embodiments, camera 901 can be tilted in azimuth and elevation while remaining in the same position relative to viewport 900.

[0141] In some embodiments, the viewport 900 can include a spectral sensor (e.g., an OES sensor configured to obtain spectral measurements as described herein) and an OES collimator 902 that can narrow the radiation beam emitted from the processing chamber. The OES collimator 902 can have a protective cover or shutter that can be kept open or closed. In some embodiments, the viewport 900 can include a time sensor, such as a photodiode 903, which is configured to obtain measurement results as described herein. In some embodiments, the viewport 900 can include an illumination system or one or more illumination sources, such as one or more LEDs 904.

[0142] In some embodiments, the above components can be packaged into a sensor group (e.g., see Figure 8 900 ) at a viewport 900 of a manufacturing tool. The viewport 900 may be arranged relative to other components of the manufacturing tool, such as a showerhead 905, a focus ring 906, a pedestal 907, and / or an RF pad plate 908.

[0143] Fig.10 An external perspective view of a chamber 1000 showing an exemplary viewport 900 of a manufacturing tool according to some embodiments. In some embodiments, the chamber 1000 may include a housing 1001 that may contain Fig. 9 The housing 1001 may be configured to include one or more components, such as a camera 901, a spectral sensor, an OES collimator 902, a photodiode 903, one or more LEDs 904, or other components described above with respect to the sensor package 810. The housing 1001 may be in a variety of shapes (e.g., Fig.10The housing 1001 may be circular, polygonal (e.g., square), or at least partially conical with corner edges and / or a flat top surface as shown and protrude from the exterior (e.g., the outside of the wall of the processing chamber of the manufacturing tool). In some implementations, the housing 1001 may take into account the material selection and component geometry of the illumination source to avoid reflection into the spatial sensor. For example, the components within the housing 1001 may have a matte or other non-reflective surface to prevent inaccurate illumination of the interior of the manufacturing tool. As described above, the housing may have a major dimension of no more than about 10 inches. As described above, the manufacturing tool or its processing chamber may include multiple viewports. These multiple viewports may accommodate one or more components, but not necessarily the same components. For example, a first viewport may have a camera but no photodiode, and a second viewport may have a photodiode but no OES sensor. The configuration may depend on whether other views visible from the viewport or window have sufficient field of view. Multiple such housings 1001 may be used with a manufacturing tool or processing chamber (eg, implemented as multiple sensor packages 123 and 125, multiple sensor packages 196, 197, and 198, or multiple sensor packages 394 and 394').

[0144] In some embodiments, housing 1001 can include various physical components and structures of supporting and stabilizing sensor. According to implementation, such structure can include viewport frame, LED holder, mounting plate, radiator, data connector and power line, printed circuit board (PCB) or printed circuit board assembly (PCBA), lens and / or outer cover. In some embodiments, thermal barrier or radiation filter (for example, UV filter) may be included or not included in some parts. Other standard structures can also be used, such as ring and fastener, communication interface (for example, universal serial bus (USB) adapter), wire channel, fin, reflective surface, vent, fan and coating. Sensor or sensor assembly (for example, OES collimator assembly, photodiode, camera) can be positioned, assembled or otherwise positioned in the above-mentioned structure, and the material (for example, aluminum) selected can be used to construct the structure with various configurations, which will be obvious to those of ordinary skill in the relevant field.

[0145] In some embodiments, the exterior of the processing chamber can include a first thermal barrier 1002 (e.g., a top thermal barrier) and / or a second thermal barrier 1003 (e.g., a bottom thermal barrier). Data connectors and / or power connectors can be placed below the first thermal barrier 1002 or the second thermal barrier 1003. For example, a fiber optic connector physical contact (FC-PC) adapter for the photodiode 903 can run under the thermal barriers. The thermal barriers 1002, 1003 can accommodate at least a portion of other components. For example, a viewport can pass through the second thermal barrier 1003, such as Fig.10In other embodiments, the viewport and one or more components thereof (e.g., a camera) may be positioned between the first thermal barrier 1002 and the second thermal barrier 1003. In fact, the viewport may be placed in any location suitable for a manufacturing tool or application.

[0146] In some embodiments, thermal management is also employed for the viewport 900 or housing 1001 by using passive and / or active cooling systems to ensure reliable operation.Other UV, thermal, and RF safety management components and features may be included in the housing 1001 . Example use cases and applications

[0147] There are many examples of applications and uses for multiple types of sensors in a processing chamber. Some applications use two or more sensors, at least two of which are some combination of spatial sensors, spectral sensors, and time sensors. In some applications, one of a plurality of such sensors is used in combination with different types of sensors (e.g., current and / or voltage sensors at locations in the processing chamber of manufacturing tool 800). Various applications involve using a combination of sensors to detect and / or characterize conditions or events occurring in a processing chamber. If such an event or condition is abnormal, a specific response may be required, such as suspending operation, replacing parts, or changing process conditions. The following are various examples of conditions and events.

[0148] An example of such an event is a change in plasma intensity in a processing chamber. Such a change may have temporal, spectral and / or spatial characteristics. For example, the change may have a spike or pulse-like temporal structure characterized by duration, leading edge shape, etc. Such a change may be associated with a condition or event in a plasma generator, physical structure, process gas flow, etc. in a processing chamber. A time sensor such as a photodiode may provide information that allows the temporal shape of a change in plasma intensity to be characterized in full or in part. Such a change may also additionally have a spectral composition associated with a specific gas or other material ionized in a plasma. Information collected by a spectral sensor (e.g., an OES sensor) may help clarify materials associated with changes in plasma intensity. In addition, such a change may be located in a specific region or regions in a processing chamber. The location may indicate a specific component or components that cause a change in plasma intensity. To this end, information collected by a spatial sensor such as a camera sensor may allow a system or process engineer to focus on one or more components associated with a change in plasma intensity.

[0149] A related example of an event or condition is the presence of a parasitic plasma in a process chamber. Parasitic plasmas typically have a consistent location that can be determined using spatial sensors. They can also be of spectral composition that can be determined using spectral sensors. Using information collected from spatial and spectral sensors, the system can determine the location, intensity, and / or ionized gas of the parasitic plasma. With this information, affected components can be replaced, cleaned, or adjusted and / or process conditions adjusted.

[0150] Another example of an event in a processing chamber is an RF pulse. As an example, such a pulse can be used to generate a pulsed plasma. The RF pulse and the associated plasma pulse can have various characteristics, such as light emitted at a specific location, having a specific spectral composition, having a specific time profile, having a specific voltage, etc. In general, such information allows the system to detect and / or characterize the RF pulse in the processing chamber. Moreover, such information can be captured using multiple sensors, such as spatial sensors (e.g., camera sensors), spectral sensors (e.g., OES sensors), time sensors (e.g., photodiodes), voltage / current sensors, or any combination thereof. In some embodiments, in order to detect and / or characterize RF pulses that last more than a specific threshold (e.g., more than 35ms) in the processing chamber, the system uses a spatial sensor as a primary indicator, a spectral sensor as a secondary indicator, and a time sensor as a third indicator. In some implementations, a VI sensor can also be used to obtain voltage and / or current data to supplement the characteristics of the RF pulse.

[0151] Another example of an event in a processing chamber is a gas composition shift. This may be due to a variety of reasons, some of which are expected and some of which are unexpected. Examples include the introduction of new gases, failure to introduce expected gases, and leaks. A gas composition shift may change some characteristics of the plasma in the processing chamber. Such changes may have spatial and / or spectral characteristics that can be detected by appropriate sensors. For example, a spatial sensor such as a camera may be used to detect a shift in plasma intensity at a process gas input location or at a leak location. Additionally or alternatively, a spectral sensor such as an OES sensor may be used to detect a shift in the spectral composition of the plasma (caused by a shift in the gas composition). As an example, in order to detect and / or characterize a shift in the gas composition in the processing chamber, a spectral sensor may be used as a primary indicator and a spatial sensor may be used as a secondary indicator. In other implementations, the spatial sensor may be the primary indicator and the spectral sensor may be the secondary indicator.

[0152] Another example of an event in a processing chamber is an arc in the chamber. The arc may damage chamber components, such as a showerhead or pedestal, introduce defects to the wafer, consume system power used to generate the plasma, etc. To detect chamber arcs within a processing chamber, a time sensor (e.g., a diode) can be used as a primary indicator (which detects the start of the arc by capturing high-intensity light emissions), and a spatial sensor (e.g., a camera) can be used as a secondary indicator to locate the arc. In some implementations, the VI sensor can also be used to obtain voltage and / or current data related to the arc. In some applications, the detection of the arc and its location can help troubleshoot defect clusters on the wafer. For example, if a defect cluster appears disproportionately on an area of ​​the wafer, and the sensor information indicates that there is an arc in the vicinity of the cluster, the defects can be reduced by addressing the underlying problem that caused the arc discharge.

[0153] Room cleaning endpoint: As another example, when operating a process chamber, accurate detection of the end point of chamber cleaning may be of interest. Processes performed within a process chamber, such as depositing a conformal material film onto a substrate using, for example, a chemical vapor deposition (CVD) process, may result not only in film deposition on the substrate, but also deposition on various chamber surfaces as a byproduct of the process. Over time, unwanted deposits accumulate on the chamber surfaces resulting in particles and potential contamination, which can have a negative impact on wafer yield. The process chamber is cleaned regularly to remove the accumulation of byproduct particles. Removal of chamber deposits can be performed, for example, by reacting a film tracking the byproduct with a reactive gas (e.g., radical fluorine) to produce silicon tetrafluoride (SiF), which can then be removed from the chamber. The optimal cleaning time for a given chamber varies depending on many factors, including the type of deposited material, temperature, pressure, reactive gas delivery, and the like. It would be advantageous to determine the end point of chamber cleaning to prevent over-cleaning caused by reacting with the surface itself.

[0154] Chamber cleaning can be performed (e.g., periodically after precursor chemicals introduced into the process chamber are deposited onto the substrate and / or internal surfaces) to maintain the life of the susceptor and / or improve the performance of a deposition or other process. However, as described above, trace amounts of chemical byproducts on components of the process chamber (e.g., walls, susceptor, or showerhead) gradually and incrementally accumulate. Detection of a cleaning endpoint is required to stop the etching process so that the buildup is completely removed but not beyond (without etching the walls or susceptor itself). Current chamber cleaning methods involve either timed cleaning (which does not account for system or process variability and / or accumulation of variations based on different processes) or the use of IR-EPD. IR-EPD looks for specific voltages and signal slopes, and adds an over-etch step. IR-EPD may cause significant etching to occur in some areas of the chamber, thereby reducing the life of the susceptor (e.g., due to AlF 3 caused by the formation of ).

[0155] In some embodiments of the present disclosure, a spectral sensor (e.g., an OES sensor) may be used as the primary sensor to indicate the cleaning status. As cleaning proceeds, the OES signal in the light collection area may weaken to the point where a small film remains, such as on the back of the susceptor, which may not have any direct impact on the plasma response captured by the OES. If this sensor is used alone, the system may leave a small amount of film, and repeated cycles may lead to particle problems. Systematically increasing over-etching similar to current methods may cause excessive formation of reaction byproducts (e.g., AlF 3 However, OES used in conjunction with a spatial sensor (e.g., a camera) directed toward the slowest etching portion (e.g., used simultaneously or at a different time than spectral sensing) can provide a secondary trigger to indicate that cleaning is complete without significant over-etching.

[0156] Process engineering (e.g., undesired gas detection) : As another exemplary use case, consider a gas leak scenario, where an undesirable gas is introduced into the system (e.g., a chamber of a multi-sensor manufacturing tool 800) or a gas burst occurs during a process step in an amount greater than expected. Process performance checks are typically done by visually verifying how the plasma behaves when viewed through a viewport, based on the response of a sensor to one of hundreds of relevant channels, or (most commonly) by performance on a wafer after the process is completed. Conventional methods do not provide real-time response. In addition, in rapidly changing systems, subtle system excursions may not be easily understood by visual inspection; plasma dynamics are on the order of μs-ms. Continuous monitoring of relevant channels is also not feasible without an (automatic) monitoring system, and currently used sensors (visual) may not accurately capture subtle changes, or may be seen as noise in the system. A reliable indication of process variation is the effect of a particular recipe after the process. Post-process metrology can indicate whether the system state has changed based on the response of characteristics (thickness, RI, etc.) on the wafer. However, not all wafers are measured. This does not make real-time control feasible and may result in undesirable wafer scrapping.

[0157] In some embodiments of the present disclosure, a spatial sensor 812 (e.g., a camera) can be used to monitor plasma during a process step. A visual model can indicate changes in peak or average intensity within a region of interest. Although the system can generate or display real-time warnings or alarms, the cause of the impact and any corrective measures can only occur after the product (e.g., a wafer) is completed to fully utilize gas flow, local pressure, power, etc. to determine the state of the system. This may lead to one or more wafer quality issues, in which the wafer may be assessed as lower quality or yield and / or completely scrapped. When the camera signal is accompanied by an OES signal indicating that a specific substance and its deposition rate may increase, real-time control of the system can be enhanced. Therefore, the deposition time or number of cycles can be reduced in real time to compensate for the excessive deposition rate at a given step (or multiple steps) of the process.

[0158] Further use cases for multi-sensor manufacturing tool 800 using combined sensor set 810 are possible.

[0159] For example, the combined sensor set 810 can enable detection of parasitic plasmas, where plasma intensity excursions in two (or more) different regions of the process chamber (e.g., the main chamber, the rim or stem of the pedestal, or the showerhead) can be monitored using a camera as a primary indicator and an OES sensor as a secondary indicator. In some implementations, this can be done by defining a plasma profile (which may involve additional computer cycles), performing a Fourier transform from the time domain to the frequency domain (>500 Hz is associated with excitation failures or particles), determining how often plasma flickering occurs, and looking for unusual excursions in intensity that may be associated with flickering.

[0160] In some embodiments, the hollow cathode discharge (HCD) of the process chamber and plasma can be monitored, for example, at the area of ​​the lift pins that define the storage position of the substrate or wafer. More specifically, in an exemplary process, the camera can be used as a primary indicator to mark the area where the lift pins are located, monitor the parasitic plasma near the lift pins (e.g., within a specified distance) to indicate incorrect position, and capture changes in plasma intensity over time.

[0161] As another example, a camera can be used as a primary indicator to detect hardware deformation. In some cases, nozzle deformation or relative gap profile changes can be determined based on changes in plasma shape. In addition, hardware damage can be prevented by monitoring the HCD in the main plasma. In some cases, the number of HCDs may indicate whether to continue the process. For example, if there are more than 10 HCDs, the process can be stopped and / or the system can be shut down, and if there are less than 10 (or other number) HCDs, the system can initiate a "soft" shutdown (e.g., deactivating non-essential modules or processes).

[0162] As another example, film flaking can be detected and prevented by monitoring film growth or flaking in different areas of a process chamber. More specifically, in one exemplary process, a film can be grown in a process chamber using, for example, one of a number of chemical vapor deposition (CVD) processes, and a camera can be used as a primary indicator to monitor film growth in the chamber or on a wafer.

[0163] As another example, process diagnostics can be performed. Using the spectral sensor as a primary indicator, process gas composition can be measured, gas connections can be confirmed, purge timing can be evaluated, and / or other related applications can be performed. To this end, an exemplary method can be to ignite the plasma and then measure the rate of change of the spectral measurement associated with a specific line of the gas.

[0164] Another example is detecting an air leak or a gas leak. An exemplary method may include igniting a plasma in a flowing gas, closing all valves, and measuring N using a spectral sensor as the primary indicator. 2 In some implementations, the spatial sensors may be spatially isolated, and the combination of the spatial sensors and the spectral sensors may enable detection of where a leak has occurred and what material has leaked.

[0165] Other examples of use cases and applications of sensor groups (e.g., of multi-sensor manufacturing tool 800) may include plasma spatial variations (primarily spatial sensors), gas impurity sensing (primarily spectral sensors), wafer placement (primarily spatial sensors), detecting droplets from rinse processes (primarily spatial sensors) to allow substrate drying, determining plasma health or quality (spectral and spatial sensors), chamber-to-chamber or station-to-station transition matching (spectral, spatial, and temporal sensors), alerting a spatial sensor (e.g., camera) or controller if information changes (e.g., within the process environment) occur faster than the frame rate of the camera (primarily temporal sensor), determining plasma ignition time when byproducts may not be effectively swept and may cause ignition delays or altered ignition profiles (primarily temporal sensors), and reconstructing the location and quality of plasma uniformity (spectral and spatial sensors), as well as other applications and functions where the accuracy, strength, or integrity of the detection signal can benefit from multiple sensor types. Using a combination of all sensors, pulses of more than 100 samples per second can be detected.

[0166] In each of the above-described use scenarios and methods, any type of spatial, spectral, or temporal sensor not explicitly mentioned may be used as a secondary and / or tertiary indicator in conjunction with other sensors (e.g., VI, temperature, pressure) and / or illumination sources associated with the processing environment.

[0167] In some cases, the primary and secondary indicators may be reversed. That is, a secondary indicator may be used as a primary indicator (independently or partially capable of obtaining measurements) even if the primary indicator is capable of independently obtaining measurements without the secondary indicator. Some configurations or packages of sensors for manufacturing tools may not include the required primary indicators. For example, the sensor group 810 mounted to the manufacturing tool may not include a photodiode, but the process may require chamber arc detection suitable for rapid light detection, for which the photodiode would be the primary indicator. In such a case, the camera is capable of detecting arcs, at least arcs lasting at least in the ms range, and therefore, the camera may instead serve as the primary indicator.

[0168] Fig.11 1 is a flow chart showing a method 1100 for multi-sensor determination of a process chamber cleaning endpoint according to some embodiments. One or more functions of the method 1100 may be performed or caused by a computerized device or system. Fig.11The structure of the functions shown in one or more steps shown may include hardware and / or software components of such a computerized device or system, such as a controller device, a computerized system, or a computer-readable device including a storage medium that stores computer-readable and / or computer-executable instructions that are configured to cause at least one processor device or computerized device to perform operations when executed by a processor device. A controller may be an example of a computerized device or system. A subsystem (e.g., 191) may be an example of a computerized device or system. A processing chamber may be another example of a computerized device or system. Exemplary components of a processing chamber and a controller are respectively described in Figure 1A and Figure 1B and Fig.13 , and described in more detail elsewhere herein.

[0169] It should also be noted that the operations of method 1100 may be performed in any suitable order, not necessarily Fig.11 In addition, method 1100 may include Fig.11 More or fewer operations may be performed to determine the cleaning endpoint.

[0170] At block 1110, method 1100 may include initiating a process chamber cleaning process. In some embodiments, the cleaning process may include initiating remote plasma cleaning and cleaning at least a portion of a process chamber of a manufacturing tool for a fixed duration. Various methods known in the related art may be used to initiate the cleaning process.

[0171] At block 1120, method 1100 may include pausing the cleaning process and igniting an inert plasma inside the manufacturing tool. The pause may be used to prevent over-etching (or cleaning beyond byproducts present, for example, as a film) inside the processing chamber (e.g., walls, components, or other surfaces within the processing chamber). Over-etching may cause damage, and determining an endpoint where damage is minimized may advantageously extend the usability and life of the equipment.

[0172] At box 1130, method 1100 may include collecting spectral information via a spectral sensor (e.g., at least one OES sensor). The spectral sensor can be part of and incorporated into a sensor package having multiple types of sensors, including a spectral sensor and at least one other type (e.g., a spatial sensor). In some implementations, the sensor package used may also include at least another type of sensor (e.g., a time sensor). The spectral information may indicate the presence of some materials and byproducts from inside the processing chamber. The presence of different tracking substances may result in different wavelengths being emitted and detected. Because the surface is made of a material different from the process byproducts (e.g., aluminum), the surface exposed after cleaning may result in the emission and detection of corresponding wavelengths.

[0173] At block 1140, method 1100 may include determining whether the spectral information indicates that the cleaning process is complete. In some cases, the spectral information may not change or may not change enough, which may indicate that the chamber cleaning is incomplete because the walls or other internal components have not been exposed from the cleaning period. In these cases, more cleaning (e.g., etching) may be required to be performed, and method 1100 may return to block 1110. In some cases, the spectral information (e.g., detected wavelengths) may shift to indicate that the surface being cleaned is exposed rather than still coated with a previous process byproduct film. In these cases, method 1100 may proceed to block 1150.

[0174] At block 1150, method 1100 may include collecting spatial information at a particular location. This may be used for secondary confirmation purposes in conjunction with the spectral information as a primary indication. In some embodiments, the spatial information may be collected via a spatial sensor (e.g., at least one camera). The spatial sensor may be part of and incorporated into a sensor package having multiple types of sensors (e.g., the same sensor package that includes a spectral sensor). Both the spectral sensor and the spatial sensor may use a common viewport or window to access the interior of the processing chamber, as described above with respect to Fig. 9 and Fig.10 In some embodiments, an illumination source (e.g., LED) may be activated while capturing spatial information. For example, the illumination source may illuminate one or more walls within the processing chamber to visually obtain an image of at least the illuminated area.

[0175] At box 1160, method 1100 may include determining whether a particular location is clean. In some embodiments, spatial information may indicate the presence of tracking byproducts at a particular location. For example, an image may show that an unclean portion of a wall may have a color different from the known color of a wall without any byproducts or films. In some implementations, the contour or boundary of a cleaned area or film may be identified to determine or estimate the percentage or quantity of a cleaned area and the percentage or quantity of a not yet cleaned area. In some cases, a ratio or percentage of a cleaned area that meets or exceeds a specified threshold amount (e.g., 80%) may indicate that a particular location is clean. In some implementations, a transparency level that meets a specified threshold (e.g., based on color composition, overall shape, edge, brightness, contrast) may indicate that a particular location is clean.

[0176] If the particular location is clean or sufficiently (above a threshold) clean, the method 1100 can end. If not, the method 1100 can return to block 1110.

[0177] In some variations, data from the spatial sensor may be captured (block 1150) at the end of the spectral data collection (block 1130), regardless of whether the spectral information indicates completion (block 1130). In such a case, the collection of spectral information and spatial information may be performed sequentially or in parallel. This method may be used to track the cleaning rate of some areas, in addition to collecting spectral data, and may also be used to optimize the cleaning process. At block 1160, the cleaning rate may further inform whether a particular location is clean.

[0178] In some variations, infrared endpoint detection (IR-EPD) may be used in conjunction with the spectral and spatial information.

[0179] In some variations, spatial information may be collected first as a primary indication, with spectral information used as secondary confirmatory information.

[0180] Fig. 12A and Fig. 12B 1 is a flow chart showing methods 1200 and 1250 for determining the presence of undesirable substances in a processing chamber according to some embodiments. One or more functions of methods 1200 and 1250 may be performed or caused by a computerized device or system. Fig. 12A and Fig. 12B The structure of the functions shown in one or more steps shown may include hardware and / or software components of such a computerized device or system, such as a controller device, a computerized system, or a computer-readable device including a storage medium that stores computer-readable and / or computer-executable instructions that are configured to cause at least one processor device or computerized device to perform operations when executed by a processor device. A controller may be an example of a computerized device or system. A subsystem (e.g., 191) may be an example of a computerized device or system. A processing chamber may be another example of a computerized device or system. Exemplary components of a processing chamber and a controller are respectively described in Figure 1A and Figure 1B and Fig.13 , and described in more detail elsewhere herein.

[0181] It should also be noted that the operations of methods 1200 and 1250 may be performed in any suitable order, not necessarily Fig. 12A and Fig. 12B In addition, methods 1200 and 1250 may include Fig. 12A and Fig. 12B More or less than the operations depicted may be performed to determine the presence of undesirable substances.

[0182] At block 1210, method 1200 may include collecting spectral information and spatial information via a spectral sensor and a spatial sensor, respectively, during a process (e.g., occurring within a processing chamber). In some embodiments, the spectral sensor may include at least one OES sensor, and the spatial sensor may include at least one camera, both of which are incorporated into a sensor package and mounted on a manufacturing tool (e.g., a processing chamber). Such a sensor package may access the interior or exterior of the processing chamber and may be able to obtain spectral and spatial information through a viewport or window.

[0183] In some variations, only spectral information may be collected, and spatial information may be related and collected at a later time. For example, a spatial sensor may be triggered to detect spatial information (block 1220) only when spectral information indicates a problem exists.

[0184] At block 1220, method 1200 may include determining whether there is a problem (e.g., the presence of an undesired gas species). In some embodiments, a spectral sensor may be used to make the determination. Consider the following scenario, where the plasma is expected to have a particular spectral signature and / or emission wavelength, such as violet. An external leak involving air entering the processing chamber or manufacturing system may occur, which may change the spectral signature and / or emission wavelength to, for example, orange. Such spectral information and / or changes in spectral information may be detected via a spectral sensor.

[0185] If detected, the method 1200 may proceed to block 1230. If not detected, the method 1200 may return to block 1210 and continue collecting spectral and / or spatial information.

[0186] At block 1230, method 1200 may include using a spatial sensor to determine whether a gas leak associated with an undesirable gas species is located. The spatial sensor (e.g., a camera) may visually isolate the problem (e.g., a gas leak) identified by the spectral sensor. The gas leak and the presence of the undesirable gas may occur outside or inside the process chamber. Visually isolating the gas leak may allow an end user or personnel to understand the leak and / or perform root cause and corrective action (RCCA) to address its cause to prevent the defect from recurring.

[0187] If positioned, the method 1200 terminates at block 1290. In some implementations, termination may occur after or concurrently with RCCA. For example, a manufacturing tool or a portion thereof (e.g., a process chamber) may be placed in a shutdown mode or at least partially disabled (e.g., some gas lines).

[0188] If not, the method 1200 may return to block 1210 to continue acquiring measurement signals via the spectral and spatial sensors. In some embodiments, the method 1200 may return to block 1220 or repeat block 1230 to attempt to locate a gas leak.

[0189] However, in some implementations, reference is now made to Fig. 12B If not located at block 1230 , the method 1200 may proceed to block 1240 , which may include determining whether to terminate the process, rather than terminating the process (block 1290 ).

[0190] If it is determined that the process should not be terminated, the method 1200 may proceed to block 1242, which may include determining whether to change a process parameter. Changing a process parameter may have a beneficial effect on the process, such as improving product yield or saving process.

[0191] If it is determined that changing the process parameters will not have a beneficial effect, or is unlikely to have a beneficial effect, the method 1200 may return to block 1210. In some variations, the method may terminate at block 1290.

[0192] If it is determined that changing the process parameters will result in a beneficial effect, or is likely to result in a beneficial effect, the method 1200 may proceed to block 1244, which may include changing one or more parameters of the process, and the method may return to block 1210 to continue capturing measurement signals via the spectral and spatial sensors. Examples of process parameters are described elsewhere herein (e.g., power states, reduction or increase of gases, plasma timing, recipe conditions). Computer and controller implementation plan

[0193] Fig.13 1 is a block diagram of an example of a computing device 1300 suitable for implementing some embodiments of the present disclosure. For example, the device 1300 may be suitable for implementing some or all of the functionality of the image analysis logic disclosed herein.

[0194] The computing device 1300 may include a bus 1302 that directly or indirectly couples the following devices: a memory 1304, one or more central processing units (CPUs) 1306, one or more graphics processing units (GPUs) 1308, a communication interface 1310, an input / output (I / O) port 1312, an input / output component 1314, a power supply 1316, and one or more presentation components 1318 (e.g., a display). In addition to the CPU 1306 and the GPU 1308, the computing device 1300 may also include Fig.13 Additional logic devices not shown include, but are not limited to, an image signal processor (ISP), a digital signal processor (DSP), an ASIC, an FPGA, etc.

[0195] although Fig.13 The various blocks of are shown as being connected by wires via bus 1302, but this is not intended to be limiting and is for clarity only. For example, in some embodiments, presentation components 1318 such as a display device may be considered I / O components 1314 (e.g., if the display is a touch screen). As another example, CPU 1306 and / or GPU 1308 may include memory (e.g., memory 1304 may also represent a storage device in addition to the memory of GPU 1308, CPU 1306, and / or other components). In other words, Fig.13 The computing devices in the description are illustrative only. No distinction is made between categories such as "workstations," "servers," "laptops," "desktops," "tablets," "client devices," "mobile devices," "handheld devices," "electronic control units (ECUs)," "virtual reality systems," and / or other device or system types, as all of these are contemplated under Fig.13 within the range of computing devices.

[0196] The bus 1302 may represent one or more buses, such as an address bus, a data bus, a control bus, or a combination thereof. The bus 1302 may include one or more bus types, such as an Industry Standard Architecture (ISA) bus, an Extended Industry Standard Architecture (EISA) bus, a Video Electronics Standards Association (VESA) bus, a Peripheral Component Interconnect (PCI) bus, a Peripheral Component Interconnect Express (PCIe) bus, and / or other types of buses.

[0197] Memory 1304 may include any of a variety of computer-readable media. Computer-readable media may be any available media that can be accessed by computing device 1300. Computer-readable media may include volatile and non-volatile media, and removable and non-removable media. By way of example and not limitation, computer-readable media may include computer storage media and / or communication media.

[0198] Computer storage media may include volatile and nonvolatile media and / or removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules, and / or other data types. For example, memory 1304 may store computer readable instructions (e.g., representing programs and / or program elements, such as an operating system). Computer storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by computing device 1300. As used herein, computer storage media itself does not include signals.

[0199] Communication media may embody computer-readable instructions, data structures, program modules, and / or other data types in a modulated data signal such as a carrier wave or other transport mechanism, and include any information delivery media. The term "modulated data signal" may refer to a signal having one or more of its characteristics that are set or changed in such a way as to encode information in the signal. By way of example and not limitation, communication media may include wired media such as a wired network or a direct wire connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above items should also be included within the scope of computer-readable media.

[0200] The CPU 1306 may be configured to execute computer-readable instructions to control one or more components of the computing device 1300 to perform one or more of the methods and / or processes described herein. The CPUs 1306 may each include one or more cores (e.g., one, two, four, eight, twenty-eight, seventy-two, etc.) capable of processing multiple software threads simultaneously. The CPU 1306 may include any type of processor and may include different types of processors, depending on the type of computing device 1300 implemented (e.g., a processor with fewer cores for a mobile device and a processor with more cores for a server). For example, depending on the type of computing device 1300, the processor may be an ARM processor implemented using reduced instruction set computing (RISC) or an x86 processor implemented using complex instruction set computing (CISC). In addition to one or more microprocessors or auxiliary coprocessors (e.g., math coprocessors), the computing device 1300 may also include one or more CPUs 1306.

[0201] GPU 1308 can be used by computing device 1300 to render graphics (e.g., 3D graphics). GPU 1308 can include many (e.g., tens, hundreds, or thousands) cores capable of processing many software threads simultaneously. GPU 1308 can generate pixel data of an output image in response to a rendering command (e.g., a rendering command from CPU 1306 received via a host interface). GPU 1308 may include a graphics memory, such as a display memory, for storing pixel data. Display memory may be included as part of memory 1304. GPU 1308 may include two or more GPUs operating in parallel (e.g., via a link). When combined, each GPU 1308 can generate pixel data of different parts of an output image or different output images (e.g., a first GPU for a first image and a second GPU for a second image). Each GPU may contain its own memory or may share memory with other GPUs.

[0202] In examples where computing device 1300 does not include GPU 1308 , CPU 1306 may be used to render graphics.

[0203] The communication interface 1310 may include one or more receivers, transmitters, and / or transceivers that enable the computing device 1300 to communicate with other computing devices via an electronic communication network (including wired and / or wireless communications). The communication interface 1310 may include components and functionality that enable communication over any of a variety of different networks, such as wireless networks (e.g., Wi-Fi, Z-Wave, Bluetooth, Bluetooth LE, ZigBee, etc.), wired networks (e.g., communicating over Ethernet), low-power wide area networks (e.g., LoRaWAN, SigFox, etc.), and / or the Internet.

[0204] I / O ports 1312 can enable computing device 1300 to be logically coupled to other devices, including coupling to I / O components 1314, presentation components 1318, and / or other components, some of which can be built into (e.g., integrated into) computing device 1300. Illustrative I / O components 1314 include microphones, mice, keyboards, joysticks, tracking pads, satellite antennas, scanners, printers, wireless devices, etc. I / O components 1314 can provide a natural user interface (NUI) that processes air gestures, voice, or other physiological inputs generated by users. In some cases, the input can be transmitted to appropriate network elements for further processing. NUI can implement any combination of speech recognition, stylus recognition, facial recognition, biometric recognition, gesture recognition on and near the screen, air gestures, head and eye tracking, and touch recognition (as described in more detail below) related to the display of computing device 1300. Computing device 1300 may include a depth camera, such as a stereo camera system, an infrared camera system, an RGB camera system, a touch screen technology, and a combination of these, for gesture detection and recognition. Additionally, computing device 1300 may include an accelerometer or gyroscope that enables detection of motion (e.g., as part of an inertial measurement unit (IMU)). In some examples, computing device 1300 may use the output of the accelerometer or gyroscope to render an immersive augmented reality or virtual reality.

[0205] The power supply 1316 may include a hardwired power supply, a battery power supply, or a combination thereof. The power supply 1316 may provide power to the computing device 1300 to enable the components of the computing device 1300 to operate.

[0206] The presentation component 1318 may include a display (e.g., a monitor, a touch screen, a television screen, a head-up display (HUD), other display types, or combinations thereof), speakers, and / or other presentation components. The presentation component 1318 may receive data from other components (e.g., GPU 1308, CPU 1306, etc.) and output the data (e.g., as images, video, sound, etc.).

[0207] The present disclosure may be described in the general context of computer code or machine-usable instructions, including computer-executable instructions (e.g., program modules) executed by a computer or other machine (e.g., a personal data assistant or other handheld device). In general, program modules, including routines, programs, objects, components, data structures, etc., refer to code that performs specific tasks or implements specific abstract data types. The present disclosure may be practiced in a variety of system configurations, including in handheld devices, consumer electronics, general-purpose computers, more specialized computing devices, etc. The present disclosure may also be practiced in distributed computing environments, where tasks are performed by remote processing devices that are linked through a communications network.

[0208] In some implementations, a "controller" (e.g., 190) is part of a system that includes various types of sensors as described herein. Such a system includes a manufacturing tool with a camera sensor. Such a system may include a semiconductor processing device that 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 electronics to control their operations before, during, and after processing semiconductor wafer substrates. The controller can be implemented or coupled to analysis logic using analysis logic as described above. The controller can be implemented as logic, such as an electronic device with one or more integrated circuits, a memory device, and / or software that receives instructions, issues instructions, controls operations, and / or enables sensing operations.

[0209] The electronic device may be referred to as a "controller" which may control various components or subcomponents of one or more systems. Depending on the processing requirements and / or system type, the controller may 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 in some systems, RF matching circuit settings, frequency settings, flow rate settings, fluid delivery settings, position and operation settings, wafer transfer to and from tools and other transfer tools and / or load locks connected or docked with a particular system.

[0210] In general, 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 circuit 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 microcontrollers that execute program instructions (e.g., software). The program instructions can be instructions sent to the controller in the form of various individual settings (or program files) that define operating parameters for performing a specific process 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 during the processing of one or more (kinds of) layers, materials, metals, oxides, silicon, silicon dioxide, surfaces, circuits, and / or dies of a wafer.

[0211] The controller may be configured to control or cause control of various components or subcomponents of one or more systems. Depending on the processing requirements and / or system type, the controller may be programmed to control any process that may be used by the manufacturing tool during manufacturing operations, including adjusting or maintaining the delivery of process gases, temperature settings (e.g., heating and / or cooling) (including substrate temperature and chamber wall temperature), pressure settings (including vacuum settings), plasma settings, RF matching circuit settings, and substrate position and operation settings (including substrate transfer into and out of the manufacturing tool and / or load lock). Process gas parameters include process gas composition, flow rate, temperature and / or pressure. Of particular relevance to the disclosed embodiments, controller parameters may relate to plasma generator power, pulse rate and / or RF frequency.

[0212] The process parameters under the control of the controller may be provided in the form of a recipe and may be input using a user interface. Signals for monitoring the process may be provided via analog and / or digital input connections of the system controller. Signals for controlling the process may be output on analog and digital output connections of the deposition apparatus.

[0213] In one example, instructions for causing or maintaining plasma are provided in the form of a process recipe. The related process recipes may be arranged sequentially so that at least some of the processing instructions may be executed simultaneously. In some implementations, instructions for setting one or more plasma parameters may be included in a recipe prior to the plasma ignition process. For example, a first recipe may include instructions for a first delay, instructions for setting the flow rate of an inert gas (e.g., helium) and / or a reactive gas, and instructions for setting a plasma generator to a first power set point. A second subsequent recipe may include instructions for a second time delay and instructions for enabling a plasma generator to supply power under a defined parameter set. A third recipe may include instructions for a third time delay and instructions for disabling a plasma generator. It should be understood that within the scope of the present disclosure, these recipes may be further subdivided and / or repeated in any suitable manner. In some deposition processes, the duration of plasma triggering may correspond to a duration of several seconds, such as from about 3 seconds to about 15 seconds, or may involve a longer duration, such as a duration of up to about 30 seconds. In certain implementations described herein, a much shorter plasma excitation may be applied during a processing cycle. Such plasma excitation duration may be on the order of less than about 50 milliseconds, with about 25 milliseconds being utilized in a particular example.As explained, the plasma may be pulsed.

[0214] In some embodiments, the controller is configured to control and / or manage the operation of the RF signal generator. In certain implementations, the controller is configured to determine an upper threshold and / or a lower threshold of the RF signal power to be transmitted to the manufacturing tool, thereby determining the actual (e.g., real-time) level of the RF signal power transmitted to the integrated circuit manufacturing chamber, the RF signal power activation / deactivation time, the RF signal on / off duration, the duty cycle, the operating frequency, etc.

[0215] As a further example, the controller can be configured to control: the timing of various operations, the mixing of gases, the pressure in the manufacturing tool, the temperature in the manufacturing tool, the temperature of the substrate or pedestal, the position of the pedestal, chuck and / or base, and a number of cycles performed on one or more substrates.

[0216] The controller may include one or more programs or routines for controlling the design subsystems associated with the manufacturing tool. Examples of such programs or routines include substrate positioning programs, process gas control programs, pressure control programs, heater control programs, and plasma control programs. The substrate positioning program may include program code for processing tool components that are used to load the substrate onto the pedestal and control the spacing between the substrate and other parts of the manufacturing tool. The positioning program may include instructions for moving the substrate into and out of the reaction chamber to deposit films on the substrate and clean the chamber.

[0217] The process gas control program may include program code for controlling gas composition and flow rate and for causing gas to flow into one or more processing stations before deposition to achieve stabilization of pressure in the processing station. In some implementations, the process gas control program includes instructions for introducing gas during film formation on a substrate in a reaction chamber. This may include introducing gas for one or more substrates in a batch of substrates for different numbers of cycles. The pressure control program may include program code for controlling the pressure in the processing station by adjusting, for example, a throttle valve in an exhaust system of the processing station, a gas flow rate flowing into the processing station, and the like. The pressure control program may include instructions for maintaining the same pressure during deposition of different numbers of cycles on one or more substrates during batch processing.

[0218] The heater control program may include program code for controlling the current flowing to a heating unit for heating the substrate. Alternatively, the heater control program may control the delivery of a heat transfer gas (eg, helium) to the substrate.

[0219] In some implementations, there may be a user interface associated with the controller. The user interface may include a display screen, a graphical software display of the device and / or process conditions, and user input devices such as a pointing device, a keyboard, a touch screen, a microphone, and the like.

[0220] 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 thereof. For example, the controller may be in the "cloud" or may be all or part of a wafer fab host system that may allow remote access to wafer processing. The computer may enable remote access to the system to monitor the current progress of a processing operation, check the history of past processing operations, check trends or performance criteria for multiple processing operations, change parameters of a current processing, set processing steps to follow the current processing, or start a new processing. 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 sent 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 process to be performed and the type of tool that the controller is configured to interface with or control. Thus, as described above, the controller may be distributed, for example, by including one or more discrete controllers networked together and working toward a common purpose (e.g., the process and control described herein). An example of a distributed controller for such a purpose is one or more integrated circuits on a chamber communicating with one or more integrated circuits remotely (e.g., at a platform level or as part of a remote computer), which combine to control the process on the chamber.

[0221] Exemplary systems may include, but are not limited to, plasma etching chambers or modules, deposition chambers or modules, spin rinse chambers or modules, metal plating chambers or modules, cleaning chambers or modules, chamfer edge etching 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 etching (ALE) chambers or modules, ion implantation chambers or modules, track chambers or modules, and any other semiconductor processing system that may be associated with or used in the manufacture and / or preparation of semiconductor wafers.

[0222] The system software may be organized in many different ways that may have different architectures. For example, various chamber component subroutines or control objects may be written to control the operation of the chamber components necessary to perform deposition processes (and other processes in some cases) according to the disclosed embodiments.

[0223] 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.

[0224] Various adjustments to the implementations described in this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other implementations without departing from the spirit or scope of the disclosure. Therefore, the claims are not intended to be limited to the implementations shown herein, but should be consistent with the broadest scope consistent with the disclosure, the principles and novel features disclosed herein.

[0225] Some features described in the content of separate implementations in this specification may also be implemented in combination in a single implementation. Conversely, various features described in the content of a single embodiment may also be implemented individually in multiple embodiments or in any suitable sub-combination. In addition, although features may be described above as working in some combinations and even initially requiring such protection, one or more features from the claimed combination may be deleted from the claimed combination in some cases, and the claimed combination may be directed to a sub-combination or a variant of a sub-combination.

[0226] Similarly, although the operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring such operations to be performed in the particular order shown or in a continuous order, or performing all the operations shown, to implement the desired result. In addition, the accompanying drawings can schematically depict another exemplary process in the form of a flow chart. However, other operations not depicted can be incorporated into the schematically displayed exemplary process. For example, one or more additional operations can be performed before, after, simultaneously or between any of the operations shown. In some cases, multitasking and parallel processing may be advantageous. In addition, the separation of various system components in the above-mentioned implementation scheme should not be understood as requiring such separation in all implementation schemes, and it should be understood that the described program components and systems can generally be integrated together, in a single software product or packaged into multiple software products. In addition, other implementation schemes are also within the scope of the appended claims. In some cases, the actions recorded in the claims can be performed in different orders and still achieve the desired result.

Claims

1. A manufacturing system comprising: Semiconductor manufacturing equipment; a plurality of sensors including one or more spatial sensors, one or more spectral sensors, and one or more temporal sensors disposed around the semiconductor manufacturing equipment; as well as a controller communicatively coupled to the plurality of sensors, the controller being configured such that: determining a set of signals from the plurality of sensors to monitor during a process performed by the semiconductor manufacturing equipment; during the process, obtaining measurements associated with the set of signals from the plurality of sensors; as well as An indication of a status of the semiconductor manufacturing equipment is determined based on a combination of data generated from the measurements associated with the signal groups.

2. The manufacturing system according to claim 1, wherein: The signal group is provided by at least two types of sensors, the at least two types of sensors comprising at least two of the following: (i) at least one of the one or more spatial sensors, (ii) at least one of the one or more spectral sensors, or (iii) at least one of the one or more temporal sensors; as well as At least one of the one or more spatial sensors includes a camera aligned toward the interior of the semiconductor manufacturing equipment, at least one of the one or more spectral sensors includes an optical emission spectroscopy (OES) sensor configured to detect one or more wavelengths of interest emitted by one or more substances produced during the process, and at least one of the one or more time sensors includes a photodiode configured to detect changes in light intensity occurring within a time period of about 1 ms or less.

3. The manufacturing system according to claim 2, wherein: The process to be performed includes detecting an undesirable amount of one or more substances within an interior of the semiconductor manufacturing equipment; and The at least two types of sensors include the one or more spatial sensors and the one or more spectral sensors.

4. The manufacturing system according to claim 2, wherein: The process to be performed includes endpoint detection of chamber cleaning inside the semiconductor manufacturing equipment; and The at least two types of sensors used simultaneously include the one or more spatial sensors and the one or more spectral sensors.

5. The manufacturing system according to claim 1, wherein: The controller is further configured to adjust a control parameter of the semiconductor manufacturing equipment based on the state of the semiconductor manufacturing equipment.

6. The manufacturing system according to claim 1, wherein: The combination of data more accurately characterizes the state of the semiconductor manufacturing equipment than a signal from only one of the plurality of sensors.

7. The manufacturing system according to claim 1, wherein: The controller includes a machine learning model; and Wherein determining the indication of the state of the semiconductor manufacturing equipment includes utilizing the machine learning model to perform a classification task or a regression task on at least a portion of the data combination to obtain measurement results associated with the signal group configured to be performed.

8. The manufacturing system according to claim 1, wherein: The one or more spatial sensors include one or more cameras or one or more camera arrays that are aligned toward the interior of the semiconductor manufacturing equipment and are configured to acquire image data related to the interior before, during, and / or after the process.

9. The manufacturing system of claim 8, further comprising one or more illumination sources configured to provide illumination to the interior of the semiconductor manufacturing equipment, Wherein the illumination enables the one or more cameras or the one or more camera arrays to obtain the image data related to the interior before, during, and / or after the process.

10. The manufacturing system according to claim 1, wherein: The one or more spatial sensors include: one or more cameras or one or more camera arrays, which are aligned toward the interior of the semiconductor manufacturing equipment and are configured to obtain data related to one or more characteristics of a plasma present in the interior during the process, the one or more characteristics of the plasma including one or more of an intensity, a boundary, or a position of the plasma.

11. The manufacturing system according to claim 1, further comprising: One or more illumination sources are configured to provide one or more light pulses to at least a portion of an interior of the semiconductor manufacturing facility, the one or more light pulses being asynchronous with one or more operations of the one or more spatial sensors.

12. A method for determining a state of a semiconductor device using multiple sensors, the method comprising: determining a signal set from a plurality of sensors of the semiconductor device to be monitored during a process to be performed by the semiconductor device, the plurality of sensors comprising at least one spatial sensor, at least one spectral sensor, and at least one temporal sensor; Based on the process to be performed, performing sensor measurements via two or more of: (i) the at least one spatial sensor, (ii) the at least one spectral sensor, or (iii) the at least one temporal sensor; as well as The state of the semiconductor device is determined based on the sensor measurements.

13. The method according to claim 12, wherein: The at least one spatial sensor comprises a camera, the at least one spectral sensor comprises an optical emission spectroscopy (OES) sensor, and the at least one temporal sensor comprises a photodiode.

14. The method of claim 12, wherein: The process to be performed includes detecting an undesirable amount of one or more substances within an interior of the semiconductor device; and The two or more of (i)-(iii) include the at least one spatial sensor and the at least one spectral sensor.

15. The method according to claim 12, wherein: The process to be performed includes detecting an endpoint of a chamber clean inside the semiconductor device; as well as The two or more of (i)-(iii) include the at least one spatial sensor and the at least one spectral sensor. 16 . The method of claim 12 , further comprising adjusting a control parameter of the semiconductor device based on the state of the semiconductor device.

17. The method of claim 12, further comprising using one or more illumination sources of the semiconductor device to verify functionality of one or more of (i) the at least one spatial sensor, (ii) the at least one spectral sensor, or (iii) the at least one temporal sensor.

18. The method of claim 12, further comprising providing pulsed light to at least a portion of an interior of the semiconductor device in a pulsed manner using one or more illumination sources, the pulsed light being asynchronous with operation of the at least one spatial sensor.

19. A multi-sensor measurement device for accessing the interior of a manufacturing tool, the multi-sensor measurement device comprising: a housing having a major dimension of about 10 inches or less; a spectral sensor in the housing; a spatial sensor in the housing; a time sensor in the housing; as well as A physical interface having a shape that allows the housing to be attached to a surface of the manufacturing tool and receive electromagnetic signals related to a process performed by the manufacturing tool through a window of the manufacturing tool.

20. The multi-sensor measuring device according to claim 19, wherein: The fabrication tool includes a processing chamber having one or more stations.

21. The multi-sensor measuring device according to claim 19, wherein: The spatial sensor includes an optical emission spectroscopy (OES) sensor configured to obtain spectral information; The spectral sensor includes a camera configured to acquire visual information; and The time sensor includes a photodiode configured to obtain time information.

22. The multi-sensor measuring device according to claim 21, wherein: The multi-sensor measurement device also includes a communication interface configured to transmit measurement results related to at least two of the spectral information, the visual information, the time information, or a combination thereof.

23. The multi-sensor measuring device according to claim 22, wherein: The process performed by the manufacturing tool includes detecting an undesirable amount of one or more substances within the interior of the manufacturing tool; and The measurement results relate to the spectral information and the visual information.

24. The multi-sensor measuring device according to claim 22, wherein: The process performed by the fabrication tool includes endpoint detection of a chamber clean of the interior of the fabrication tool; and The measurement results relate to the spectral information and the visual information.

25. The multi-sensor measurement device of claim 19, further comprising an illumination source within the housing, the illumination source configured to provide illumination to at least a portion of an interior of the manufacturing tool; wherein the illumination: enables the spatial sensor to obtain visual information related to at least a portion of the interior of the manufacturing tool before, during, and / or after the process performed by the manufacturing tool; includes light pulses that are asynchronous with obtaining the visual information related to at least a portion of the interior of the manufacturing tool; facilitates calibration or verification of one or more of the spectral sensor, the spatial sensor, or the temporal sensor, or a combination thereof.

Images