Condensation detection for vapor precursor delivery in semiconductor manufacturing

Abstract

Methods and systems for detecting condensation for vapor precursor delivery in a manufacturing system. A flow of a vapor in a vapor delivery line is terminated and one or more pressure measurements of the vapor delivery line are obtained. A vapor quantity is determined based on the one or more pressure measurements. The vapor quantity and a supply pressure of the vapor delivery line are processed using a model that outputs a condensation risk level in the vapor delivery line. Upon determining that the condensation risk level satisfies a condensation risk criterion, an alert indicating detection of a condensate in the vapor delivery line is generated.

Description

Attorney Docket No.: 08090.1223 (L1056PCT)CONDENSATION DETECTION FOR VAPOR PRECURSOR DELIVERY IN SEMICONDUCTOR MANUFACTURINGTECHNICAL FIELD

[0001] Embodiments of the present disclosure relate, in general, to manufacturing sy stems and more particularly to systems and methods for detecting condensation for vapor precursor delivery.BACKGROUND

[0002] In semiconductor manufacturing, vapor delivery lines transfer vaporized chemical precursors to processing chambers for use in substrate deposition processes. However, vapor precursors can condense (i.e., revert to a liquid state) durin the transfer process due to factors such as insufficient heating, high moisture levels, or impurities in the vapor precursors, causing defects on substrates or clogging vapor delivery lines. Due to the small size and gradual accumulation of condensate, condensate formation can be difficult to detect when caused by factors other than manufacturing equipment failure. For example, impurities in vapor precursors, such as precursor molecules that have bonded and formed chains with lower vapor pressures, can cause condensation even when equipment temperatures are maintained at appropriate levels. The difficulty in detection can cause significant delays in identifying condensate buildup, increasing response time for implementing corrective maintenance actions, clogging vapor delivery lines, and causing substrate defects from unstable deposition processes.SUMMARY

[0003] Embodiments described herein provide systems, methods, and non -transitory computer-readable media for detecting condensate for vapor precursor delivery in a manufacturing system.

[0004] In one aspect, a method includes terminating a flow of a vapor in a vapor delivery' line and obtaining one or more pressure measurements of the vapor delivery' line. The method further includes determining, based on the one or more pressure measurements, a vapor quantity in the vapor delivery line. The method further includes processing the vapor quantity' and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery line in view of the vapor quantity and the supply pressure. The method further includes, responsive to determining that the condensation riskAttorney Docket No.: 08090.1223 (L1056PCT)level indicated by the one or more outputs satisfies a condensation criterion, generating an alert indicating detection of a condensate in the vapor delivery line.

[0005] In one aspect, a system includes a processing chamber, a vapor delivery line, and a computing device. The computing device is to terminate a flow of a vapor in the vapor delivery line and obtain one or more pressure measurements of the vapor delivery line. The computing device is further to determine, based on the one or more pressure measurements, a vapor quantity in the vapor delivery line. The computing device is further to process the vapor quantity and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery line in view of the vapor quantity and the supply pressure. Responsive to determining that the condensation risk level indicated by the one or more outputs satisfies a condensation criterion, the computing device is further to generate an alert indicating detection of a condensate in the vapor delivery line.

[0006] In one aspect, a non-transitory computer-readable medium includes instructions that, when executed by a set of one or more processing devices, cause the set of one or more processing devices to terminate a flow of a vapor in a vapor delivery line and obtain one or more pressure measurements of the vapor delivery line. The set of one or more processing devices are further to determine, based on the one or more pressure measurements, a vapor quantity in the vapor delivery line. The set of one or more processing devices are further to process the vapor quantity and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery line in view of the vapor quantity and the supply pressure. Responsive to determining that the condensation risk level indicated by the one or more outputs satisfies a condensation criterion, the set of one or more processing devices are further to generate an alert indicating detection of a condensate in the vapor delivery' line.BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The present disclosure is illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like references indicate similar elements. It should be noted that different references to “an” or “one” embodiment in this disclosure are not necessarily to the same embodiment, and such references mean at least one.

[0008] FIG. 1 depicts an illustrative system architecture, according to aspects of the present disclosure.Attorney Docket No.: 08090.1223 (L1056PCT)

[0009] FIG. 2 is a block diagram of an example condensate engine, according to aspects of the present disclosure.

[0010] FIG. 3 is a flow chart of an example method for training an artificial intelligence (Al) model, according to aspects of the present disclosure.

[0011] FIG. 4 is a flow chart of an example method for detecting condensation in a vapor delivery line, according to aspects of the present disclosure.

[0012] FIG. 5 depicts a block diagram of an example chemical precursor delivery system, in accordance with aspects of the present disclosure.

[0013] FIG. 6 depicts a pressure versus time graph illustrating pressure decay in a vapor delivery line after flow termination, in accordance with aspects of the present disclosure.

[0014] FIG. 7 depicts a scatter plot illustrating condensate detection in a vapor delivery line in accordance with aspects of the present disclosure.

[0015] FIG. 8 depicts a block diagram of an illustrative computer system operating in accordance with one or more aspects of the present disclosure.DETAILED DESCRIPTION

[0016] Implementations described herein provide systems and methods for condensate detection in vapor delivery lines for vapor precursor delivery. In semiconductor manufacturing vapor precursor delivery refers to the process of transporting vaporized chemical precursors to a reaction chamber for use in a chemical process, such as chemical vapor deposition. For example, a liquid chemical compound (e.g., trimethyl aluminum) can be heated and changed to a gaseous state, then transported through vapor delivery lines to a processing chamber, where it is deposited on a sub strate. Condensation refers to when vaporized chemical precursors revert from a gaseous state to a liquid state.

[0017] While being transported in a vapor delivery line, vapor precursors can condense due to insufficient heating, high moisture levels, and / or impurities in the vapor precursors. For example, impurities can include precursor molecules that have bonded and formed chains with lower vapor pressures than unbonded precursor molecules, causing the bonded molecules to condense more readily than the unbonded molecules. However, condensates are often small in size (e.g., microscopic), making them hard to visually detect. Additionally, condensate accumulates slowly over time, so changes to flow or pressure may appear gradually rather than suddenly. As such, condensate formation can be difficult to detect in a vapor delivery line before there is a tangible indication of the formation. Tangible indications can include defects on a substrate caused by condensate particles or reduced flow rates due to progressiveAttorney Docket No.: 08090.1223 (L1056PCT)condensate accumulation that clogs the vapor delivery line. However, such indications appear after a problem has developed due to condensate accumulation.

[0018] The difficulty in detection can cause significant delays in identifying condensate buildup, which increases response time for implementing corrective maintenance actions. Failure to detect condensate in a timely manner can also cause a manufacturing system to continue operating under compromised conditions. Continuing under compromised conditions can cause condensate particles to be deposited onto substrate surfaces, creating localized areas of non-uniform film growth that compromise wafer quality' and reduce manufacturing yield. Additionally, prolonged condensate buildup can clog vapor delivery lines, causing unstable deposition processes due to inconsistent precursor flow rates. Excessive accumulation can also cause tool shutdowns and unscheduled downtime to address the condensate accumulation, which disrupts production schedules and increases operational costs for emergency repairs.

[0019] One conventional approach to preventing condensation includes monitoring equipment related to vapor phase changes, such as the heat output of heater jackets around vapor delivery' lines, to maintain vapor precursors at temperatures that prevent liquid formation. However, condensate that forms due to other factors, such as impurities in vapor precursors, are not detectable through the conventional approaches that monitor equipment because such condensation can occur regardless of equipment functionality. For example, impurities in vapor precursors can cause condensation even when temperatures are maintained at appropriate levels, as the condensation is driven by the chemical properties (e.g., lower vapor pressure) of the impurities rather than thermal conditions.

[0020] Additionally, precursor chemistry and supply pressures can vary across manufacturing equipment. Differences in vapor properties and operating conditions can cause the same amount of condensation to produce significantly different pressure and flow signatures, complicating detection of condensation. For example, condensates are easier to detect in equipment operating at a high supply pressure due to the larger density’ of vapor molecules that make pressure and flow deviations more pronounced. In contrast, at a lower supply pressure, the density is smaller, so the same amount of condensate produces proportionally less pressure and flow deviations, making detection more difficult. Such variation in pressure and flow signatures prevents the use of standard detection thresholds across different operating conditions (e.g., different supply pressures).

[0021] Aspects of the present disclosure address the above noted and other deficiencies by providing systems and methods for condensate detection in vapor delivery lines. A system can train a model (e.g., a digital twin model and / or an Al model) to predict whether a condensateAttorney Docket No.: 08090.1223 (L1056PCT)is present in a vapor delivery line based on a vapor quantity and a supply pressure of the line. The vapor quantity refers to the amount of residual vapor present in the vapor delivery line after the flow of vapor is terminated. In some embodiments, the vapor quantity can be calculated by determining an area under a pressure versus time curve derived from pressure measurements obtained after terminating the flow of vapor in the delivery line. The supply pressure refers to the operating supply pressure of the vapor delivery line before the flow is terminated. In some embodiments, the model can also be trained to normalize the vapor quantity by using the supply pressure to determine an expected vapor quantity and comparing the measured vapor quantity with the expected vapor quantity.

[0022] In some instances, the model can be trained using training data that includes a set of training inputs and a set of target outputs. A training input can include a historical vapor quantity in the vapor delivery line determined from historical pressure measurements and a historical supply pressure. A target output of the training input can include historical condensation data indicating the condensation risk level in the vapor delivery line based on the historical vapor quantity and historical supply pressure. The condensation risk level can be a numerical value, percentage, or classification that quantifies the probability that condensate is present in the vapor delivery line. In some embodiments, the historical condensation data can be provided by an operator of the system. Further details regarding training the model are provided herein.

[0023] Upon trainingthe model, the model can be applied to vapor quantities derived from real-time pressure measurements from a vapor delivery line. For example, the system can obtain pressure measurements of a vapor delivery line after the flow of vapor in the vapor delivery line is terminated. The system can determine a vapor quantity based on the pressure measurements (e.g., by calculating an area under a pressure versus time curve derived from the pressure measurements). The system can process the vapor quantity and a supply pressure of the vapor delivery' line using the model, which outputs an indication of a condensation risk level in the vapor delivery line in view of the vapor quantity and the supply pressure. The system can determine whether the condensation risk level satisfies a condensation criterion (e.g., by determining whether the condensation risk level exceeds a condensation threshold). Upon determining that the condensation risk level satisfies the condensation criterion, the system can generate an alert indicating detection of a condensate in the vapor delivery line. In some embodiments, the system can transmit the alert to a client device for presentation to one or more users. In some embodiments, the alert may include an indication of a region (e.g., the region by the pressure sensor) of the vapor delivery line that includes the anomaly. In someAttorney Docket No.: 08090.1223 (L1056PCT)embodiments, other remedial actions are performedresponsive to detecting condensation, such as scheduling of maintenance (e.g., to clean or replace a portion of the vapor delivery line), taking the process chamber off line, and so on.

[0024] Aspects of the present disclosure address the above noted and other deficiencies by providing techniques for detecting condensate in vapor delivery' lines. By using the vapor quantity in a vapor delivery line to determine whether condensate is present, condensate due to impurities in precursor vapor can be detected even when manufacturing equipment is functioning as intended. The use of supply pressure to normalize the calculated vapor quantity¬ accounts for supply pressure variations to provide consistent detection performance across different operating conditions. Enhanced accuracy in detecting condensation caused by impurities in vapor precursors enables faster corrective action, minimizes downtime, and supports predictive maintenance scheduling to maximize system uptime. Additionally, early and accurate identification of condensate accumulation can prevent sub strate defects caused by condensate particles, reducing manufacturing scrap and improving production yield.

[0025] FIG. 1 depicts an illustrative system architecture 100, according to aspects of the present disclosure. System architecture 100 can include a client device 120, manufacturing equipment 124, metrology equipment 128, a predictive server 112 (e.g., to generate predictive data, to provide model adaptation, to use a knowledge base, etc.), and / or a data store 140. The predictive server 112 can be part of a predictive system 110. The predictive system 110 can further include serv er machines 170 and 180. In some embodiments, system architecture 100 can be included as part of or otherwise connected to a manufacturing system for processing substrates.

[0026] Manufacturing equipment 124 can produce products, such as electronic devices, following a recipe or performing runs over a period of time. Manufacturing equipment 124 can include a process chamber. Manufacturing equipment 124 can perform a process for a substrate (e.g., a wafer, etc.) at the process chamber. Examples of substrate processes include a deposition process to deposit a film on a surface of the substrate (e.g., using chemical vapor deposition (CVD), atomic layer deposition (ALD), physical vapor deposition (PVD), etc.), an etch process to form a pattern on the surface of the substrate, a polishing process to polish a material on the surface of the substrate, etc. Manufacturing equipment 124 can perform each process according to a process recipe. A process recipe defines a particular set of operations to be performed for the substrate during the process and can include one or more settings associated with each operation. For example, a deposition process recipe can include a temperature setting for the process chamber, a pressure setting for the process chamber, a flowAttorney Docket No.: 08090.1223 (L1056PCT)rate setting for a precursor for a material included in the film deposited on the substrate surface, etc. Substrates that are processed according to a process recipe (e.g., for manufacturing a portion of an electronic device, etc.) are referred to herein as production substrates.

[0027] Manufacturing equipment 124 can include one or more sensors that capture data for a substrate being processed at the manufacturing system. In some embodiments, the manufacturing equipment 124 and the sensors can be part of a sensor system that includes a sensor server (e.g., field service server (FSS) at a manufacturing facility) and sensor identifier reader (e.g., front opening unified pod (FOUP) radio frequency identification (RFID) reader for sensor system). Sensor data may include a value of one or more of temperature (e.g., heater temperature), spacing(SP), pressure, high frequency radio frequency (HFRF), RF bias, voltage of electrostatic chuck (ESC), electrical current, flow, power, voltage, pressure, etc. Sensor data may be associated with or indicative of manufacturing parameters such as hardware parameters, such as settings or components (e.g., size, type, etc.) of the manufacturing equipment 124, or process parameters of the manufacturing equipment 124. The sensor data can be provided while the manufacturing equipment 124 is performing manufacturing processes (e.g., equipment readings when processing products). In some embodiments, sensor data can include trace data collected during performance of one or more processes (e.g., substrate processes, maintenance processes, etc.) at manufacturing equipment 124. Trace data refers to data that indicates how components in a process chamber are operating and / or a state of an environment within a process chamber before, during, or after performance of an operation.

[0028] Manufacturing equipment 124 can include one or more vapor delivery lines 126 to transport vapor precursors to processing chambers in the manufacturing system. In some implementations, the vapor deliver}' lines 126 can include one or more injection valves that control the introduction of vaporized chemical precursors into carrier gas streams. A carrier gas stream is a flow of an inert gas (e.g., nitrogen) that transports vaporized precursors through the vapor delivery lines. In some embodiments, the vapor delivery lines 126 can include one or more mass flow controllers (MFCs) that regulate the flow rate of vapor precursors and / or carrier gases. In some embodiments, the vapor delivery lines 126 can include a pressure sensor positioned at a specified location within the vapor delivery line 126 to monitor pressure conditions. For example, in some embodiments, a pressure sensor can be located in a vapor delivery line between a MFC for a carrier gas and an injection valve for a vapor precursor. The sensor can be used to monitor the carrier gas line for pressure changes due to precursor condensate that forms in the vapor delivery line and causes vapor precursors to migrateAttorney Docket No.: 08090.1223 (L1056PCT)backwards from the injection valve into the carrier gas line. In some embodiments, more than one pressure sensor canbe used. In some embodiments, eachvapor delivery' line 126 can supply a different precursor to a processing chamber.

[0029] FIG. 5, described in greater detail below, shows components of a vapor delivery line, in accordance with embodiments of the present disclosure.

[0030] Metrology equipment 128 can be used to perform metrology on substrates at various stages of manufacturing. The metrology equipment 128 may include an inspection station where substrates are inspected after being processed in a process chamber in some embodiments. In some embodiments, the inspection station can be located in the vacuum environment of, for example, a process chamber or transfer chamber. In other embodiments, the inspection station can be located outside the vacuum environment (e.g., as external equipment from the manufacturing equipment).

[0031] The client device 120 may include a computing device such as personal computers (PCs), laptops, mobile phones, smart phones, tablet computers, netbook computers, network connected televisions (‘"smart TV s”), network-connected media players (e.g., Blu-ray player), a set-top box, over-the-top (OTT) streaming devices, operator boxes, etc. In some embodiments, the metrology' data may be received from the client device 120. In some embodiments, client device 120 displays a graphical user interface (GUI), where the GUI enables the user to provide, as input, metrology' measurement values for substrates processed at the manufacturing system. In other or similar embodiments, client device 120 can display another GUI that enables the user to provid e, as input, an indication of a type of substrate to be processed at the manufacturing system, a type of process to be performed for the substrate, and / or a type of equipment at the manufacturing system. In yet other or similar embodiments, client device 120 can display another GUI thatthatpresents sensordata collected by the sensors before, during, or after performance of a process (e.g., a substrate process, a maintenance process, etc.). It should be noted that one or more GUIs of client device 120 can provide and / or receive any data described herein.

[0032] Data store 140 canbe a memory / (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or another type of component or device capable of storing data. Data store 140 can include multiple storage components (e.g., multiple drives or multiple databases)thatcan span multiple computing devices (e.g., multiple server computers). The data store 140 can store data associated with processing a substrate at manufacturing equipment 124. For example, data store 140 can store data collected by the sensors at manufacturing equipment 124 before, during, or after a substrate process (referred to as processAttorney Docket No.: 08090.1223 (L1056PCT)data). Process data can refer to historical process data (e.g., process data generated for a previous substrate processed at the manufacturing system) and / or current process data (e.g., process data generated for a current substrate processed at the manufacturing system). Current process data can be data for which predictive data is generated (e.g., such as predictive data identifying condensates in a vapor delivery line). In some embodiments, data store can store metrology data including historical metrology data (e.g., metrology measurement values fora prior substrate processed at the manufacturing system). The data store 140 can also store contextual data associated with one or more substrates processed at the manufacturing system. Contextual data can include a recipe name, recipe operation number, preventive maintenance indicator, operator, etc. In some embodiments, contextual data can also include an indication of a difference between two or more process recipes or process operations. In some embodiments, data store 140 can additionally or alternatively store image data collected (e.g, by imaging component 126) for substrate before, during, or after processing of the substrate, as described herein.

[0033] In some embodiments, data store 140 can be configured to store data that is not accessible to a user of the manufacturing system. For example, process data, spectral data, nonspectral data, and / or positional data obtained for a substrate being processed at the manufacturing system may not be accessible to a user of the manufacturing system. In some embodiments, all data stored at data store 140 is inaccessible by a user (e.g., an operator) of the manufacturing system. In other or similar embodiments, a portion of data stored at data store 140 is inaccessible by the user while another portion of data stored at data store 140 is accessible by the user. In som e embodim ents, one or m ore portions of data stored at data store 140 are encrypted using an encryption mechanism that is unknown to the user (e.g, data is encrypted using a private encryption key). In other or similar embodiments, data store 140 includes multiple data stores where data that is inaccessible to the user is stored in one or more first data stores and datathatis accessible to the user is storedin one or more second data stores.

[0034] Computing system 150 can include a condensate engine 152 that detects whether condensate accumulation is present in a vapor delivery line based on pressure measurements and supply pressure data obtained from vapor delivery lines 126 in manufacturing equipment 124. The pressure measurements are obtained afterthe flowof vapor (e.g., the carrier gas and / or the vapor precursor) is terminated. The condensate engine 152 can calculate the amount of vapor (i.e., vapor quantity) in the vapor delivery line 126 usingthe pressure measurements. For example, the condensate engine 152 can calculate the vapor quantity by determining an area under a pressure versus time curve derived from the pressure measurements, in someAttorney Docket No.: 08090.1223 (L1056PCT)embodiments. In some embodiments, the condensate engine 152 can apply a model (e.g., model 190) to the vapor quantity' and the supply pressure to determine whether condensate is present in the vapor delivery line. The model 190 may include an Al model and / or a physics-based model (e.g., a digital twin). In some embodiments, condensate engine 152 can include or have access to a predictive component 114 (e.g., of predictive system 110) which provides the vapor amount and supply pressure data as inputs to the model 190. In other or similar embodiments, predictive system 110 can train the model 190 to predict condensate anomalies in vapor delivery lines, as described below with respect to FIG. 3.

[0035] As stated above, the model 190 can include a physics-based digital twin model, in some embodiments. The physics-based digital twin model can be capable of solving systems of equations describing physical phenomena that occur in vapor delivery' lines, such as equations governing gas flow, pressure decay, and / or the ideal gas law. The physics-based digital twin model can be trained on data from vapor delivery lines operating under normal conditions without condensate accumulation, in some embodiments. A vapor delivery line operating under normal conditions can be expected to produce measured pressure readings that approximately match simulated pressure readings of the trained digital twin model.

[0036] While the above description provides examples of a physics-based digital twin model, the model 190 can include any model that is capable of predicting condensation risk levels in vapor delivery lines. These can include, but are not limited to, machine learning models (e.g., neural networks, support vector machines, random forests), statistical models, or any other Al-based or physics-based modeling approach. In some embodiments, the model 190 can include a hybrid model that combines physics-based calculations with machine learning techniques. In some embodiments, the various models discussed in connection with model 190 may be combined in one model (e.g., an ensemble model), or may be separate models. Data may be passed back and forth between several distinct models included in model(s) 190.

[0037] In some embodiments, predictive system 110 includes server machine 170 and server machine 180. Server machine 170 includes atraining set generator 172 thatis capable of generating training data sets (e.g., a set of data inputs and a set of target outputs) to train, validate, and / or test a model 190. Some operations of training set generator 172 are described in detail below with respect to FIG. 3. In some embodiments, the training set generator 172 can partition the training data into a training set, a validating set, and a testing set. In some embodiments, the predictive system 110 generates multiple sets of training data.

[0038] Server machine 180 includes a training engine 182, a validation engine 184, a selection engine 186, and / or a testing engine 188. An engine can refer to hardware (e.g.,Attorney Docket No.: 08090.1223 (L1056PCT)circuitry, dedicated logic, programmable logic, microcode, processing device, etc.), software (such as instructions run on a processing device, a general purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. Training engine 182 can be capable of training a model 190. The model 190 can referto the model artifact thatis created by the training engine 182 using the training data that includes training inputs and corresponding target outputs (correct answers for respective training inputs). The training engine 182 can find patterns in the training data that map the training input to the target output (the answer to be predicted), and provide the patterns to the model 190 that captures these patterns. In som e embodim ents, the model 190 uses one or more of support vector machine (SVM), Radial Basis Function (RBF), clustering, supervised machine learning, semisupervised machine learning, unsupervised machine learning, k-nearest neighbor algorithm (k-NN), linear regression, random forest, neural network (e.g., artificial neural network, a recurrent neural network, a convolutional neural network, etc.), clustering techniques (e.g, hierarchical clustering techniques), association techniques (e.g., apriori techniques), classification techniques (e.g., decision trees, random forest techniques, etc.), a variational recurrent auto-encoder, etc. It should be noted that although some embodiments of the present disclosure describe model 190 as an Al model and / or physics-based model, such embodiments can be applied to any type of Almodel, non-AIbased model (e.g., a statistical model, a physical model, etc.), and / or a hybrid model (e.g., implementing Al techniques and non-AI techniques).

[0039] The validation engine 184 can be capable of validating a trained model 190 using a corresponding set of features of a validation set from training set generator 172. The validation engine 184 can determine an accuracy of each of the train ed models 190 based on the corresponding sets of features of the validation set. The validation engine 184 can discard a trained model 190 that has an accuracy that does not meet a threshold accuracy. In some embodiments, the selection engine 186 can be capable of selecting a train ed model 190 thathas an accuracy that meets a threshold accuracy. In some embodiments, the selection engine 186 can be capable of selecting the trained model 190 that has the highest accuracy of the trained models 190.

[0040] The testing engine 188 can be capable of testing a trained model 190 using a corresponding set of features of a testing set from training set generator 172. For example, a first train ed model 190 that was trained using a first set of features of the training set can be tested using the first set of features of the testing set. The testing engine 188 can determine a trained model 190 thathas the highest accuracy of all of the trained modelsbased on the testing sets.Attomey Docket No.: 08090.1223 (L1056PCT)

[0041] Predictive server 112 includes a predictive component 114 that is capable of providing data as an input to a trained model 190 and obtaining one or more outputs of the trained model 190, As described herein, predictive component 114 can be a component of or otherwise associated with condensate engine 152 and can provide the vapor quantity and supply pressure as an inputto trained model 190. Predictive component 114 can obtain one or more outputs of trained model 190, which can include an indication of the condensation risk level, as described herein.

[0042] As illustrated by FIG. 1, the client device 120, manufacturing equipment 124, metrology equipment 128, predictive server 112, data store 140, server machine 170, and server machine 180 can be coupled to each othervia a network 130. In some embodiments, network 130 is a public network that provides client device 120 with access to predictive server 112, data store 140, and otherpublicly available computing devices. In some embodiments, network 130 is a private network that provides client device 120 access to manufacturing equipment 124, metrology equipment 128, data store 140, and other privately available computing devices. Network 130 can include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet network), wireless networks (e.g., an 802.11 network or a Wi-Fi network), cellular networks (e.g., a Long-Term Evolution (LTE) network), routers, hubs, switches, server computers, cloud computing networks, and / or a combination thereof.

[0043] It should be noted that in some other implementations, the functions of server machines 170 and 180, as well as predictive server 112, can be provided by a fewer number of machines. For example, in some embodiments, server machines 170 and 180 can be integrated into a single machine, while in some other or similar embodiments, server machines 170 and 180, as well as predictive server 112, can be integrated into a single machine.

[0044] In general, functions described in one implementation as being performed by server machine 170, serv er machine 180, and / or predictive server 112 can alsobe performed on client device 120. In addition, the functionality attributed to a particular component can be performed by different or multiple components operating together.

[0045] In embodiments, a “user” can be representedas a single individual. However, other embodiments of the disclosure encompass a “user” being an entity controlled by one or more users and / or an automated source. For example, a set of individual users federated as a group of administrators can be considered a “user.”

[0046] FIG. 2 is a block diagram of an example condensate engine 152, according to aspects of the present disclosure. As described above, condensate engine 152 can obtain aAttorney Docket No.: 08090.1223 (L1056PCT)prediction of the condensation risk level associated with a vapor delivery' line 126 in manufacturing equipment 124. In some embodiments, the condensate engine 152 can provide alerts for detected anomalies for display via the UI of the client device 120. The condensate engine 152 may additionally or alternatively perform other corrective actions, such as automatically scheduling maintenance, taking a processing chamber offline, and so on. As illustrated by FIG. 2, condensate engine 152 can include a measurement component 210, an alert component 212, a predictive component 114, and / or a process control component 214. In some embodiments, condensate engine 152 can be connected to the predictive system 110 and / or memory 250 (e.g., via network 130). The predictive system 110 can be connected to a client device 120 in some embodiments (e.g., via network 130). In other or similar embodiments, memory' 250 can include any memory' of or accessible to a component of system 100. For example, memory 250 can include, or be included in, data store 140, a memoiy of client device 120, and so forth.

[0047] As described herein, condensate engine 152 can obtain a predicted indication of the condensation risk level in a vapor delivery line 126 based on one or more outputs of a model 190. The model 190 may be trained based on historical data and / or physics-based models associated with system 100 (or another system) to predict the condensation risk level based on a given vapor quantity' and supply pressure of a vapor delivery line. Details regarding training the model 190 are provided herein with respect to FIG. 3.

[0048] FIG. 3 is a block diagram depicting an example method 300 for training an Al model, according to aspects of the present disclosure. Method 300 is performed by processing logic that can include hardware (circuitry, dedicated logic, etc.), software (such as is run on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one implementation, method 300 can be performedby a computer system, such as computer system architecture 100 of FIG. 1. In other or similar implementations, one or more operations of method 300 can be performed by one or more other machines not depicted in the figures. In some aspects, one or more operations of method 300 can be performed by training set generator 172.

[0049] For simplicity of explanation, method 300 is depicted and described as a series of acts. However, acts in accordance with this disclosure can occur in various orders and / or concurrently, and with other acts not presented and described herein. Furthermore, not all illustrated acts can be performed to implement the methods in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that the methods could alternatively be represented as a series of interrelated states via a state diagramAttorney Docket No.: 08090.1223 (L1056PCT)or events. Additionally, it should be appreciated that the methods disclosed in this specification are capable of being stored on an article of manufacture to facilitate transporting and transferring such methods to computing devices. The term article of manufacture, as used herein, is intended to encompass a computer program accessible from any computer-readable device or storage media.

[0050] At block 310, processing logic initializes a training set T to an empty set (e.g., {}). At block 312, processing logic identifies a historical vapor quantity in a vapor delivery line 126 and a historical supply pressure of the vapor deliver}' line 126. In some embodiments, the measurement component 210 can calculate the historical vapor quantity from historical pressure measurements obtained after termination of vapor flow in the vapor delivery line 126. The historical pressure measurements can be obtained from a pressure sensor positioned in the vapor delivery line 126 between a carrier gas MFC and an injection valve, in some embodiments. In some embodiments, the historical vapor quantity can be calculated by determining an area under a historical pressure versus time curve based on the historical pressure measurements. In some embodiments, the processing logic can retrieve the historical vapor quantity from vapor quantity data 256 in memory 250. The historical supply pressure refers to the operating pressure of the vapor delivery line before the flow was terminated in the historical operation. The model 190 can be trained using historical vapor amounts and corresponding historical supply pressures obtained under different operating conditions, in some embodiments. In some embodiments, the processing logic (e.g., training set generator 172) can identify or otherwise obtain the historical pressure measurements from pressure data 254 and supply pressure data 252 in memory 250. In other or similar embodiments, the pressure data 254, vapor quantity’ data 256, and / or supply pressure data 252 can be provided by an operator of manufacturing equipment 124 and / or a developer or operator of system 100, in some embodiments.

[0051] At block 314, processing logic obtains a historical indication of the condensation risk level associated with the historical vapor quantity and the historical supply pressure. For example, the historical indication can identify whether condensate accumulation was present in the vapor delivery line for the corresponding historical measurements. In some embodiments, the historical indication of the condensation risk level can be identified from condensation data 258 stored atmemory 250. In some embodiments, the historical indication of the condensation risk level can be obtained from maintenance records (e.g., operator log;) documenting equipment downtime or repair activities related to condensate buildup in vapor delivery lines.Attorney Docket No.: 08090.1223 (L1056PCT)

[0052] At block 316, processing logic generates a training input based on the historical vapor quantity and the historical supply pressure. In some embodiments, the training input can include the historical pressure measurements. In some embodiments, the training input can include additional contextual data such as precursor chemistry information, equipment configuration parameters, or process recipe settings associated with the historical measurements.

[0053] At block 318, processing logic generates a target output based on the historical indication of the condensation risk level. At block 320, processing logic generates a mapping between the training input and the target output. At block 322, processing logic adds the mapping to the training set T. At block 324, processing logic determines whether the training set T includes a sufficient amount of training data to train an Al model. It should be noted that in some implementations, the sufficiency of training set T can be determined based simply on the number of mappings in the training set, while in some other implementations, the sufficiency of training set T can be determined based on one or more other criteria (e.g., a measure of diversity of the training examples, etc.) in addition to, or instead of, the number of input / output mappings. Responsive to determining the training set does not include a sufficient amount of training data to train the Al model, method 300 returns to block 312. Responsive to determining the training set, T, includes a sufficient amount of training data to train the Al model, method 300 continues to block 326.

[0054] At block 326, processing logic provides training set T to train the Al model. In one implementation, the training set T is provided to training engine 182 of server machine 180 to perform the training. In the case of a neural network, for example, input values (e.g., the training input) of a given input / output mapping are input to the neural network, and output values (e.g., the target output) of the input / output mapping are stored in the output nodes of the neural network. The connection weights in the neural network are then adjusted in accordance with a learning algorithm (e.g., backpropagation, etc.), and the procedure is repeated for the other input / output mappings in the training set T. After block 326, the model 190 can be used to predict a condensation risk level in a vapor delivery line 126, as described herein.

[0055] Referring back to FIG. 2, the condensate engine 152 can determine whether condensate is present based on one or more outputs of the model 190. Embodiments relating to detecting condensation are described with respect to FIG. 2 and FIG. 4.

[0056] FIG. 4 is a block diagram of an example method 400 for detecting condensation in a vapor delivery' line, in accordance with embodiments of the present disclosure. Method 400 is performedby processing logic that can include hardware (circuitry, dedicated logic, etc.),Attorney Docket No.: 08090.1223 (L1056PCT)software (such as is run on a general-purpose computer system or a dedicated machine), firmware, or some combination thereof. In one implementation, method 400 can be perfomied by a computer system, such as computer system architecture 100 of FIG. 1. In other or similar implementations, one or more operations of method 400 can be performed by one or more other machines not depicted in the figures. In some aspects, one or more operations of method 400 can be performed by condensate engine 152.

[0057] For simplicity of explanation, method 400 is depicted and described as a series of acts. However, acts in accordance with this disclosure can occur in various orders and / or concurrently, and with other acts not presented and described herein. Furthermore, not all illustrated acts can be performed to implement the methods in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that the methods could alternatively be represented as a series of interrelated states via a state diagram or events. Additionally, it should be appreciated that the methods disclosed in this specification are capable of being stored on an article of manufacture to facilitate transporting and transferring such methods to computing devices. The term article of manufacture, as used herein, is intended to encompass a computer program accessible from any computer-readable device or storage media.

[0058] At block 402, processing logic terminates the flow of a vapor in a vapor delivery line. The flow of the vapor may be terminated, for example, by closing a valve. In some embodiments, process control component 214 can perform one or more operations to terminate the flow of the vapor precursor and / or carrier gas in the vapor delivery line. For example, in som e embodiments, process control component 214 can transmit one or more signals to a liquid flow meter (LFM) that controls the flow of chemical precursors and / or a carrier gas MFC to pause or stop flow of vapors in the vapor delivery line. The termination of flow may be performed as part of a routine maintenance process or during process transitions where manufacturing equipment 124 changes operational states (e.g., from a processing state to a cleaning state). In some embodiments, the termination can be performed as manufacturing equipment 124 changes the process recipe being performed (e.g., changes the precursors used, changes process parameters). In some embodiments, condensation detection operations can be performed periodically, after a predetermined time interval, or after a predetermined number of substrate processing operations.

[0059] At block 404, processing logic obtains one or more pressure measurements of a vapor delivery line. In some embodiments, the measurement component 210 obtains the pressure measurements from a pressure sensor positioned within the vapor delivery line 126.Attorney Docket No.: 08090.1223 (L1056PCT)For example, in some embodiments, the pressure sensor can be located between a carrier gas mass flow controller and an injection valve in the vapor delivery line 126. When condensate forms near the injection valve and subsequently evaporates, the resulting vapor increases the total vapor quantity within the segment of the vapor delivery' line 126 between the injection valve and the carrier gas mass flow controller (MFC). The pressure sensor located in this region can detect changes in pressure that reflect the presence of addition al vapor from evaporated condensate. The pressure measurements capture the decrease in pressure as residual vapor (i.e., vaporremainingin the vapor delivery line 126 after flowtermination)leaves the vapor delivery line 126. Residual vapor can include carrier gas and / or vapor precursors remainingin the vapor delivery line 126. The pressure measurements can be collected continuously over a period of time to generate pressure versus time data that reflects the amount of residual vapor evacuated from the vapor delivery line 126, in some embodiments.

[0060] At block 406, processing logic determines a vapor quantity in the vapor delivery line. In some embodiments, the measurement component 210 can calculate the vapor quantity from the pressure measurements obtained after termination of vapor flow in the delivery line. When vapor flow is terminated in the delivery line, residual vapor molecules remain trapped in the line. As the line is evacuated (pumped out), the pressure decreases over time as these residual vapor molecules are removed. The relationship between pressure and vapor quantity is governed by the ideal gas law, where pressure times volume is proportional to the number of moles of gas present in the contained volume. After terminating the flow of vapor precursor and carrier gas into the line, pressure measurements may be continuou sly obtained from a pressure sensor positionedin the vapor delivery line. In some embodiments, the pressure sensor may be located between the earner gas mass flow controller and the injection valve. These pressure measurements may be collected over a period of time to generate a pressure versus time curve that captures the pressure decay as the line is pumped out. An example of such a pressure versus time curve is illustrated in FIG. 6 below.

[0061] The vapor quantity represents the total amount of residual vapor molecules present in the delivery line when the flow is terminated. In some embodiments, the measurement component 210 calculates the vapor quantity by determining an area under a pressure versus time curve derived from the pressure measurements. This area corresponds to an integral of pressure over time (JP dt), which is proportional to the total amount of vapor present in the delivery line. The pressure-time curve typically exhibits an exponential decline as residual vapor evacuates from the line until reaching a baseline pressure level. This tail of the pressure curve, sometimes referred to as the "tail of the precursor," represents the residual vapor beingAttorney Docket No.: 08090.1223 (L1056PCT)removed from the system. The area corresponds to an integral of pressure overtime, which is proportional to the total amount (e.g., the number of moles) of vapor present.

[0062] The calculated vapor quantity alone may not provide sufficient information to detect condensation, as the amount of residual vapornaturally varies with operating conditions. In some embodiments, the supply pressure (the base pressure at which the line was pressurized before flow termination) may be usedto normalize the vapor quantity measurement. The supply pressure affects the density of vapor molecules in the line, which in turn affects the expected vapor quantity. By accounting for the supply pressure, the system can distinguish between normal variations in vapor quantity due to different operatingpressures and abnormal in creases in vapor quantity caused by condensate evaporation.

[0063] At block 408 processing logic processes the vapor quantity and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery line in view of the vapor quantity and the supply pressure. In some embodiments, the predictive component 114 can provide the calculated vapor quantity and supply pressure as inputs to the model 190 (e.g., which may be an Al model, a digital twin, or a combination thereof), which is trained to detect condensation based on the relationship between vapor quantity and supply pressure. The supply pressure refers to the operating pressure of the vapor delivery line 126 before the flow was terminated. For example, in some embodiments, the predictive component 114 can determine the supply pressure using the pressure sensors to measure the pressure before the vapor flow is terminated.

[0064] In some embodiments, the condensation risk level can be a numerical value, percentage, or classification (e.g., high risk, low risk) that quantifies the probability that condensate is present in the vapor delivery line. In some embodiments, the classification can be a binary classification indicating whether condensate is present. In some embodiments, the classification can be a percentage indicating the likelihood of condensation. In some embodiments, the output can include a confidence level associated with the prediction.

[0065] In some embodiments, the model 190 is trained to normalize the vapor quantity using the supply pressure. For example, the m odel 190 can use the supply pressure to determine an expected vapor quantity for the supply pressure. The actual vapor quantity is then divided by the expected vapor quantity to determine a condensation index to create a ratio indicating how much the measured vapor quantity deviates from the expected vapor quantity. By calculating an expected vapor quantity based on the supply pressure, the model can account for different vapor quantities due to different supply pressures. The model 190 can then use theAttorney Docket No.: 08090.1223 (L1056PCT)condensation index to determine the condensation risk level in the vapor delivery line 126, in some embodiments.

[0066] At block 410, processing logic determines whether the condensation risk level satisfies a condensation criterion. In some embodiments, the condensation criterion can be satisfied if the condensation risk level is a classification value indicating the presence of condensate. For example, the predictive component 114 can determine that the condensation criterion is satisfied if the condensation risk level is a binary value indicating that condensate is present. In some embodiments, the condensation criterion can be satisfied if a value (e.g, a percentage, a numerical value) of the condensation risk level exceeds a predefined condensation threshold. In some embodiments, the condensation threshold can be defined by analyzing historical condensation risk level values to identify which values correspond to known condensation events in a vapor delivery line 126. In some embodiments, the predictive component 114 can compare the confidence level to a predetermined confidence threshold to determine whether a condensation criterion is satisfied. In some embodiments, the condensation criterion (e.g., condensation threshold, confidence threshold) can be stored as condensation criteria 260 in memory 250. In some embodiments, the prediction component 114 can determine whether the condensationrisk level satisfies multiple condensation criterion. Responsive to a determination that the condensation criterion is not satisfied, method 400 proceeds to block 412. Responsive to a determination that the condensation criterion is satisfied, method 400 proceeds to block 414.

[0067] At block 412, processing logic proceeds with a manufacturing process. In some embodiments, the process control component 214 can transmit one or signals to the manufacturing equipment 124 to continue normal operations. For example, the flow of vapor in the vapor delivery line is resumed for substrate processing operations.

[0068] At block 414, processing logic generates an alert indicating detection of a condensate in the vapor delivery line. In some embodiments, the alert component 212 can transmit the alert to a client device 120 indicating that there is condensate accumulation in vapor delivery line 126. In some embodiments, the alert further includes an indication of a region of the vapor delivery line that includes the anomaly. For example, the alert can indicate the region of the vapor delivery' line 126 where the pressure sensor is located, in some embodiments. In some embodiments, the alert can include a classification indicating that the condensate is due to impurities in the vapor precursor. In some embodiments, instead of, or in ad dition to, outputting an alert, processing logic performs one or more other corrective actions, such as scheduling maintenance or taking a tool offline.Attorney Docket No.: 08090.1223 (L1056PCT)

[0069] FIG. 5 depicts a block diagram of an example chemical precursor delivery system 500, in accordance with aspects of the present disclosure. The chemical precursor delivery system 500 includes a liquid chemical precursor container 510 that stores a liquid chemical precursor (e.g., trimethyl aluminum) to be delivered to a processing chamber. A liquid flow meter (LFM) 512 connected to the liquid chemical precursor container 510 measures the flow rate of the liquid chemical precursor. The liquid precursor is vaporized and introduced to a carrier gas stream through an injection valve 514. In some embodiments, heated vaporization chambers or heated delivery lines are used to heat the precursor and convert it to a gaseous state.

[0070] The carrier gas transports the vaporized precursor through the vapor delivery line (e.g., vapor delivery line 126) to a final valve 540, which controls the flow of gas to a processing chamber. The flow of carrier gas can be regulated by a carrier gas mass flow controller (MFC) 530. In some embodiments, a pressure sensor 520 is positioned between the injection valve 514 and the carrier gas MFC 530. The pressure sensor 520 monitors pressure conditions in the vapor delivery line 126 and obtains pressure measurements used to determine vapor quantity in the delivery line. For example, the pressure sensor 520 can measure the pressure after the LFM 512 and carrier gas MFC 530 stop the flow of precursors and carrier gas in the vapor delivery line 126. When condensate formsnear the injection valve 514 and subsequently evaporates, the evaporated vapor increases the total vapor quantity within the segment of the vapor delivery line 126 betweenthe injection valve 514 and the carrier gas MFC 530. The pressure sensor 520 located in this region can measure changes in pressure that reflect the presence of additional vapor from evaporated condensate. The pressure values are then used by the condensate engine 152 to determine wh ether condensate is present in the vapor delivery line 126 and generate an alert. In some embodiments, the alert can include an indication that the condensate is located near the injection valve 514.

[0071] In some embodiments, the chemical precursor delivery' system 500 is represented using a digital twin.

[0072] FIG. 6 depicts a pressure versus time graph 600 illustrating pressure decay in a vapor delivery line after flow termination in accordance with aspects of the present disclosure. Time is depicted as the horizontal axis of the graph 600. The graph depicts the flow rate 610 of vapors (e.g., vapor precursors, carrier gas) in a vapor delivery line 126. The flow rate 610 is maintained at constant value until the flow is terminated (e.g., by sending one or more signals to the LFM 512 and / or carrier gas MFC 530). The flow termination is depicted in the graph 600 by the drop in the flow rate 610.Attorney Docket No.: 08090.1223 (L1056PCT)

[0073] The graph also depicts the pressure 620 in the vapor delivery line 126. In some embodiments, the pressure can be obtained using a pressure sensor 520. The pressure 620 begins at an elevated level corresponding to the flow rate 610, then decreases after the flow rate 610 is terminated. As depicted by the graph 600, the pressure 620 experiences an exponential decline as residual vapor evacuates from the delivery line until reaching a baseline pressure level. A shaded area 622 under the pressure-time curve corresponds to the integral of pressure overtime, which is proportional to the total vapor quantity (e.g., number of moles) present in the vapor delivery line 126. When condensate is present and evaporates after flow termination, the additional vapor from the evaporated condensate increases the total vapor quantity, creating a larger area under the pressure-time curve than would be expected from residual gas alone at a given supply pressure. For example, in some embodiments, the measurement component 210 of the condensate engine 152 can calculate the area 622 under the pressure 620 curve to determine the vapor quantity. The condensate engine 152 can then provide the vapor quantity and a supply pressure to a model 190 to determine whether there is condensate in the vapor delivery line 126.

[0074] FIG. 7 depicts a scatter plot 700 illustrating condensate detection in a vapor delivery line in accordance with aspects of the present disclosure. The scatter plot 700 displays data points plotted with supply pressure on the horizontal axis and vapor quantity (e.g., an area 622 corresponding to a pressure versus time curve) on the vertical axis. Each data point corresponds to a determined area 622 from a vapor delivery line 126 operating a specified supply pressure. The line represents an expected area at each supply pressure value.

[0075] The scatter plot 700 also includes an anomalous data point depicted in a position significantly above the linear trend line. The anomalous data point indicates that the measured vapor quantity exceeds the expected value for the given supply pressure, suggesting that condensate has evaporated and contributed extra vapor to the system. In some embodiments, the model 190 can use the linear relationship between vapor quantity and supply pressure to normalize vapor quantities across different supply pressures and identify condensate by detecting vapor quantities that deviate significantly from the linear trend.

[0076] FIG. 8 depicts a block diagram of an illustrative computer system 800 operating in accordance with one or more aspects of the present disclosure. In alternative embodiments, the machine can be connected (e.g., networked) to other machines in a Local Area Network (LAN), an intranet, an extranet, or the Internet. The machine can operate in the capacity of a server or a client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine can be a personal computer (PC), a tabletAttomey Docket No.: 08090.1223 (L1056PCT)computer, a set-top box (STB), a Personal Digital Assistant (PDA), a cellular telephone, a web appliance, a server, a network router, switch orbridge, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term "machine" shall also be taken to include any collection of machines (e.g., computers) that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein. In embodiments, computing device 800 can correspond to predictive server 112 of FIG.1 or another processing device of system 100. In other or similar embodiments, computing device 800 can correspond to computing system 150 of system 100.

[0077] The example computing device 800 includes a processing device 802, a main memory 804 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM), etc.), a static memory 806 (e.g., flash memory, static random access memory (SRAM), etc.), and a secondary memory (e.g., a data storage device 828), which communicate with each other via a bus 808.

[0078] Processing device 802 can represent one or more general-purpose processors such as a microprocessor, central processing unit, or the like. More particularly, the processing device 802 can be a complex instruction set computing (CISC) microprocessor, reduced instruction set computing (RISC) microprocessor, very long instruction word (VLIW) microprocessor, processor implementing other instruction sets, or processors implementing a combination of instruction sets. Processing device 802 can also be one or more special-purpose processing devices such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), network processor, or the like. Processing device 802 can also be or include a system on a chip (SoC), programmable logic controller (PLC), or other type of processing device. Processing device 802 is configured to execute the processing logic for performing operations and steps discussed herein.

[0079] The computing device 800 can further include a network interface device 822 for communicating with a network 864. The computing device 800 also can include a video display 810 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alpha-numeric input device 812 (e.g., a keyboard), a cursor control device 814 (e.g., a mouse), and a signal generation device 820 (e.g., a speaker).

[0080] The data storage device 828 can include a machine-readable storage medium (or more specifically a non-transitory computer-readable storage medium) 824 on which is stored one or more sets of instructions 826 embodying any one or more of the methodologies or functions described herein. A non-transitory storage medium refers to a storage medium otherAttorney Docket No.: 08090.1223 (L1056PCT)than a carrier wave. The instructions 826 can also reside, completely or at least partially, within the main memory 804 and / or within the processing device 802 during execution thereof by the computer device 800, the main memory 804 and the processing device 802 also constituting computer-readable storage media.

[0081] The computer-readable storage medium 824 can also be used to store model 190 and data used to train model 190. The computer readable storage medium 824 can also store a software library containing methods that call model 190. While the computer-readable storage medium 824 is shown in an example embodiment to be a single medium, the term "computer- readable storage medium" should be taken to include a single medium or multiple media (e g, a centralized or distributed database, and / or associated caches and servers) that store the one or more sets of instructions. The term "computer-readable storage medium" shall also be taken to include any medium that is capable of storing or encoding a set of instructions for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present disclosure. The term "computer-readable storage medium" shall accordingly be taken to include, but not be limited to, solid-state memories, and optical and magnetic media.

[0082] The preceding description sets forth numerous specific details such as examples of specific systems, components, methods, and so forth in order to provide a good understanding of several embodiments of the present disclosure. It will be apparent to one skilled in the art, however, that at least some embodiments of the present disclosure can be practiced without these specific details. In other instances, well-known components or methods are not described in detail or are presented in simple block diagram format in order to avoid unnecessarily obscuring the present disclosure. Thus, the specific details set forth are merely exemplary'. Particular implementations can vary’ from these exemplary details and still be contemplated to be within the scope of the present disclosure.

[0083] Reference throughout this specification to “one embodiment” or “an embodiment ’ means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, the appearances of the phrase “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same em bodiment. In addition, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” When the term “about” or “approximately” is used herein, this is intended to mean that the nominal value presented is precise within ± 10%.

[0084] Although the operations of the methods herein are shown and described in a particular order, the order of operation s of each method can be altered so that certain operationsAttorney Docket No.: 08090.1223 (L1056PCT)can be performed in an inverse order so that certain operations can be performed, at least in part, concurrently with other operations. In another embodiment, instructions or sub -op erations of distinct operations can be in an intermittent and / or alternating manner.

[0085] It is understood that the above description is intended to be illustrative, and not restrictive. Many other embodiments will be apparent to those of skill in the art upon reading and understanding the above description. The scope of the disclosure should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

Attorney Docket No.: 08090.1223 (L1056PCT)CLAIMSWhat is claimed is:

1. A method comprising:terminating a flow of a vapor in a vapor delivery line;obtaining one or more pressure measurements of the vapor delivery line; determining, based on the one or more pressure measurements, a vapor qu antity in the vapor delivery line;processing the vapor quantity and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery' line in view of the vapor quantity and the supply pressure; andresponsive to determining that the condensation risk level satisfies a condensation criterion, generating an alert indicating detection of a condensate in the vapor delivery line.

2. The method of claim I, wherein the alert further comprises an indication of a region of the vapor delivery' line that includes the condensate based on a position of a pressure sensor in the vapor delivery line.

3. The method of claim 1, wherein the model normalizes the vapor quantity' using the supply pressure.

4. The method of claim 1, wherein determining the vapor quantity comprises calculating an area under a pressure versus time curve derived from the one or more pressure measurements.

5. The method of claim 1. wherein determining that the condensation risk level satisfies the condensation criterion comprises determining that the condensation risk level exceeds a condensation threshold.

6. The method of claim 1, wherein the one or more pressure measurements are obtained from a pressure sensor located in the vapor delivery line.

7. The method of claim 6, wherein the pressure sensor is located between an injection valve and a carrier gas mass flow controller.Attorney Docket No.: 08090.1223 (L1056PCT)8. The method of claim 1, wherein the supply pressure is an operating pressure of the vapor delivery line before the flow of the vapor is terminated.

9. The method of claim 1, wherein the model comprises at least one of an artificial intelligence (Al) model or a physics-based model.

10. A system comprising:a processing chamber;a vapor delivery line; anda computing device configured to perform operations comprising:terminating a flow of a vapor in the vapor deliver}' line;obtaining one or more pressure measurements of the vapor delivery line; determining, based on the one or more pressuremeasurements, a vapor quantity in the vapor delivery line;processing the vapor quantity and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery' line in view of the vapor quantity and the supply pressure; and responsive to determining that the condensation risk level satisfies a condensation criterion, generating an alert indicating detection of a condensate in the vapor delivery' line.

11. The system of claim 10, further comprising a pressure sensor in the vapor delivery line, wherein the one or more pressure measurements are obtained from the pressure sensor.

12. The system of claim 10, wherein the alert further comprises an indication of a region of the vapor delivery line that includes the condensate based on a position of a pressure sensor in the vapor delivery line.

13. The system of claim 10, wherein determining the vapor quantity comprises calculating an area under a pressure versus time curve derived from the one or more pressure measurements.Attorney Docket No.: 08090.1223 (L1056PCT)14. The system of claim 10, wherein determining that the condensation risk level satisfies the condensation criterion comprises determining that the condensation risk level exceeds a condensation threshold.

15. A non-transitory computer readable medium comprising instructions that, when executed by a set of one or more processing devices, cause the set of one or more processing devices to:terminate a flow of a vapor in a vapor delivery line;obtain one or more pressure measurements of the vapor delivery line; determine, based on the one or more pressure measurements, a vapor quantity in the vapor delivery' line;process the vapor quantity and a supply pressure of the vapor delivery line using a model that outputs an indication of a condensation risk level in the vapor delivery line in view of the vapor quantity and the supply pressure; andresponsive to determining that the condensation risk level satisfies a condensation criterion, generate an alert indicating detection of a condensate in the vapor delivery line.

16. The non-transitory computer readable medium of claim 15, wherein the alert further comprises an indication of a region of the vapor delivery line that includes the condensate based on a position of a pressure sensor in the vapor delivery line.

17. The non-transitory computer readable medium of claim 15, wherein determining the vapor quantity comprises calculating an area under a pressure versus time curve derived from the one or more pressure measurements.

18. The non-transitory computer readable medium of claim 15, wherein determining that the condensation risk level satisfies the condensation criterion comprises determining that the condensation risk level exceeds a condensation threshold.

19. The non-transitory computer readable medium of claim 15, wherein the model normalizes the vapor quantity using the supply pressure.

20. The non-transitory computer readable medium of claim 15, wherein the one or more pressure measurements are obtained from a pressure sensor located in the vapor delivery line.

Images