Load locking with diagnostic and repair capabilities

The load lock system, with its self-diagnostic and automatic repair capabilities, monitors and automatically cleans the load lock in real time, resolving the issue of defective products caused by contamination of the load lock system, thereby improving production efficiency and reducing maintenance costs.

CN123126087APending Publication Date: 2026-11-17APPLIED MATERIALS INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202580028738.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-09-04
Filing Date
2025-01-06
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing load-locking systems can become dirty or contaminated if left unattended, leading to substandard substrate processing. Regular cleaning and particle testing also increase tool downtime and costs.

Method used

The load locking system employs self-diagnostic and automatic repair capabilities, including sensors, a computing subsystem, and a complex atom system. It can monitor and automatically execute cleaning cycles in real time, cleaning the load lock through high-speed airflow and a vacuum pump, reducing manual intervention and downtime.

Benefits of technology

The system automates cleaning and maintenance of the load-locking system, reducing the production of defective products, improving production efficiency, and lowering maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123126087A_ABST
    Figure CN123126087A_ABST
Patent Text Reader

Abstract

A load lock includes a sensing and re-atom system to repair measured conditions within the load lock. The sensing subsystem can use various sensors to measure conditions within a processing chamber, and a computing subsystem can selectively activate either the re-atom system or the repair subsystem of the re-atom system to repair one aspect of the measured condition, such as chamber contamination deemed necessary based on the measured condition. The repair subsystem may include several mechanisms, including gas purging of the chamber. The entire system can adjust for chamber contamination, wafer contamination, or chamber component integrity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of this disclosure generally relate to a load lock for transferring a substrate or wafer to further processing devices and chambers, and particularly to a load lock system capable of performing self-diagnosis and repair. Background Technology

[0002] An electronic component manufacturing system may include a factory interface (which may be, for example, an Equipment Front End Module (EFEM)) configured to receive a substrate on which electronic components can be manufactured, a transfer chamber for transferring the substrate to and from a processing chamber, and one or more load locks that separate the transfer chamber from the factory interface.

[0003] The internal condition of a load lock can deteriorate over time. For example, the load lock may become dirty and / or contaminated, and mechanical degradation or failure may occur within or around the load lock. To address the gradual buildup of contaminants and particles within the load lock, engineers typically schedule regular maintenance and cleaning. For instance, a load lock may be scheduled for cleaning every three months without actually knowing its internal condition.

[0004] In some cases, engineers will discontinue the use of load locks to perform particle testing on test substrates within the load lock. The test substrates can then be removed from the load lock and measured using external metrology equipment to determine the number of particles on the test substrates. This periodic testing can be used to determine when to perform load lock cleaning. However, because discontinuing the use of load locks and running particle tests is costly in terms of tool downtime and engineer time, this test is not performed frequently (e.g., once a day). Therefore, multiple product substrates are typically processed in dirty or contaminated load locks before it is determined that the load lock is dirty or contaminated. Furthermore, most particle tests performed show uncontaminated load locks, unnecessarily reducing tooling time.

[0005] Unattended and unattended unsatisfactory processing conditions can jeopardize the substrate (including the final product and equipment), reduce throughput, and increase costs due to system downtime and component maintenance. Summary of the Invention

[0006] In one aspect of this disclosure, a load lock is provided for an electronic component manufacturing system. The load lock includes: a substrate support configured to hold a substrate; at least one sensor for generating a measurement reflecting a condition within the load lock; a computing subsystem configured to: process the measurement to determine the condition within the load lock, and determine whether to perform a repair action based on the condition of the load lock; and a re-atom system configured to perform a repair action in response to the determination to perform a repair action.

[0007] In one aspect of this disclosure, a load lockout system for electronic component manufacturing is provided. The load lockout includes: an inlet gas line including a first valve; an outlet gas line; a vacuum pump coupled to the outlet gas line; and a controller operatively coupled to the first valve and the vacuum pump, wherein the controller performs an automatic cleaning cycle for the load lockout, and wherein, in order to perform the automatic cleaning cycle, the controller sets a target pressure for the vacuum pump between 1 mTorr and 10 Torr and actuates the first valve to cause an initial flow rate of gas entering the load lockout through the inlet line greater than 100 sccm / L.

[0008] In one aspect of this disclosure, a load lock is provided for an electronic component manufacturing system, and a method for performing a repair action within the load lock. The method for performing a repair action within the load lock includes: generating a measurement of the condition within the load lock via at least one sensor; processing the measurement via a computational subsystem to determine the condition within the load lock; determining, via the computational subsystem, whether to perform a repair action based on the condition of the load lock; and performing the repair action via a complex atomic system in response to the determination to perform the repair action. Attached Figure Description

[0009] The accompanying drawings illustrate this disclosure by way of example rather than limitation, wherein the same reference numerals denote similar elements. It should be noted that different references to "a" or "an" embodiment in this disclosure do not necessarily refer to the same embodiment, and such references imply at least one.

[0010] The accompanying drawings are for illustrative purposes only and are not necessarily drawn to scale. The drawings are not intended to limit the scope of this disclosure in any way.

[0011] Figure 1A A top view of an exemplary embodiment of a substrate manufacturing system including a load-locking system and placement, as well as other processing chambers and equipment in a factory environment, is shown.

[0012] Figure 1B A cross-sectional side view of an exemplary embodiment of a load locking system with associated sensing and complex atom systems is shown.

[0013] Figure 2An exemplary implementation of a production process associated with the use of a load locking system is shown.

[0014] Figure 3 An exemplary implementation of a computing subsystem associated with the use of a load locking system is shown.

[0015] Figure 4 An exemplary implementation of the hardware and control algorithm associated with the use of a load locking system is shown.

[0016] Figure 5A An embodiment of a standard gas purification process is illustrated in the characteristic diagram.

[0017] Figure 5B One embodiment of the purification and cleaning process is illustrated in the diagram.

[0018] Figure 6 An embodiment of a computing device associated with a substrate manufacturing system is shown in graphical representation. Detailed Implementation

[0019] The embodiments described herein relate to a load-locking system and load lock capable of self-diagnosis and automatic recovery. The load-locking system and load lock can be used in processing or manufacturing systems, such as in substrate processing or manufacturing systems.

[0020] The embodiments described herein relate to a load-locking system that includes one or more sensors capable of performing self-diagnostics and / or one or more automatic recovery and / or prevention systems capable of performing automatic repair actions (e.g., a purification cleaning cycle). In embodiments, the load-locking system may further include one or more contamination and / or degradation prevention systems. The load-locking system can periodically or continuously generate measurements of the substrate, load-locked components, the internal environment of the load-lock, etc. For example, measurements can be taken of each substrate passing through the load-lock, such as before the substrate enters the load-lock, while the substrate is under load-lock, and / or after the substrate leaves the load-lock. The load-locking system can process the measurements to determine one or more conditions of the substrate, one or more components under load-lock, the environment of the load-lock, etc. The determined conditions can be compared with one or more criteria. If the determined conditions meet one or more criteria associated with a repair action, a repair action can be automatically suggested, scheduled, and / or initiated. Therefore, problematic conditions within the load lock can be automatically detected and / or predicted without halting the load lock operation or running test processes on the test substrate within the load lock. Consequently, the operating time of the load lock can be significantly longer than that of a conventional load lock. Furthermore, any problematic conditions within the load lock can be predicted or detected immediately after a single substrate may have been exposed to problematic conditions (e.g., high particulate contamination, high humidity, organic contamination, etc.) and before any additional substrates are exposed to problematic conditions.

[0021] In one embodiment, the load lockout system includes an automatic recovery and / or prevention system that can be activated in response to the load lockout system detecting a problematic condition (e.g., in response to determining that the load lockout has a particle count higher than a particle count threshold) or predicting a future problematic condition. In one embodiment, the automatic recovery system includes a valve connected to the load lockout gas inlet, a vacuum pump connected to the load lockout gas outlet, and a controller configured to perform an automatic cleanup process on the load lockout. To perform the automatic cleanup process, the controller can open the valve at the load lockout gas inlet to allow gas to flow into the load lock at high speed when the load lock is under vacuum. This high-speed gas inflow may interfere with particles accumulated on various surfaces inside the load lock and cause those particles to propagate through the air. The cleanup process may further include setting the vacuum pump coupled to the load lockout outlet to a low vacuum pressure, which is lower than the load lockout's standard operating vacuum pressure. The valve at the inlet may close after a period of time to allow the load lock to reach vacuum pressure again, followed by an additional cleanup cycle in which the valve reopens to allow the high-speed gas flow to re-enter the load lock. In one embodiment, multiple cleanup cycles may be performed according to the cleanup formulation. By automatically initiating a cleaning process in response to the detection of a dirty load lock, the load lock system can clean the load lock on demand, rather than relying on periodic scheduled maintenance that disconnects it from the actual current state of the load lock. Furthermore, the load lock system can clean the load lock without the need for technical personnel or by opening the load lock and exposing it to the external environment (which typically results in a lengthy requalification process). Therefore, the load lock system described in this embodiment offers several advantages over conventional load locks.

[0022] In this implementation, load locking is either autonomous or semi-autonomous, meaning the decision can be made locally without transmitting data to a remote computing device. In this implementation, intelligent load locking (also known as autonomous or semi-autonomous load locking) can determine when to perform maintenance (such as a cleaning process) and can further perform this maintenance automatically without technician intervention. In this implementation, intelligent load locking uses a trained machine learning model to process sensor data, where the trained machine learning model produces outputs such as predictions of future problems and / or suggestions for performing maintenance and / or other corrective actions.

[0023] In some implementations, the trained machine learning model is an edge-based model executed on a load-locked and / or load-locked substrate processing system (e.g., a platform, transfer chamber, host, fab interface, and / or tool cluster). Alternatively, the trained machine learning model may be executed on a manufacturing facility (“semiconductor foundry (fab)”) level computing device or a remote server computing device. In some implementations, training of the machine learning model may be performed remotely, after which the trained machine learning model may be transferred to a load-locked controller and / or substrate processing system.

[0024] In one example, a decision can be made regarding whether to schedule cleaning (e.g., automatic cleaning) of the load lock chamber after the substrate has circulated through the load lock chamber and before subsequent substrates have circulated through it. For example, the decision to perform a cleanup process on the smart load lock can be made within 1 to 5 minutes of the substrate circulating through the load lock chamber, within less than 1 minute of the substrate circulating through the load lock chamber, or even within a few seconds or less of the substrate circulating through the load lock chamber. This rapid response time reduces the exposure of the product substrate (the substrate that will produce the device product to be sold to consumers) to non-compliant load locks, which could lead to substrate contamination and / or failure of the final manufactured product.

[0025] The components of the embodiments described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, as shown in the drawings, the detailed description of various embodiments is not intended to limit the scope of this disclosure, but merely to illustrate various embodiments. Although various aspects of the embodiments are presented in the drawings, the drawings are not necessarily drawn to scale unless specifically indicated otherwise. The term “coupled to” has a broad meaning sufficient to indicate any suitable coupling or other form of interaction between two or more entities, including direct and / or indirect mechanical, fluid, and thermal interactions. Thus, two components may be coupled to each other even if they are not directly coupled. The term “attached to” or “directly attached to” refers to an interaction between two or more entities that are in direct contact with each other and / or separated from each other only by any suitable type of fastener (e.g., mounting hardware or adhesive). The term “fluidly connected” is used in its common sense and has a broad meaning sufficient to indicate an arrangement in which fluid (e.g., gas or liquid) can flow from one component to another when the components are in fluid communication with each other.

[0026] Please refer to the attached diagram. Figure 1AThis is a schematic diagram of a cluster tool 100 (also referred to as a system, substrate processing system, or manufacturing system) configured for substrate manufacturing according to at least some embodiments of this disclosure. In one exemplary embodiment, the manufacturing system 100 may include a processing section 104, a transfer chamber 110, a load lock 120, a factory interface 106, and a substrate carrier 122 or a front-opening unified pod (FOUP). The processing section 104 may include a plurality of processing chambers 114, 116, and 118 in which specific and controlled substrate manufacturing processes occur. The transfer chamber 110 may house a transfer robot 112, which includes a substrate transfer mechanism or end effector capable of transferring substrates 102 (in this disclosure, the substrate transfer mechanism and end effector are interchangeable for further operation). The transfer chamber 110 may be located within a transfer chamber housing 108. The load lock 120 may be connected to the processing section 104 and the factory interface 106. Factory interface 106 may include a factory interface robot 126 for transferring substrates to and from carrier 122 and load lock 120. The factory interface may further include a plurality of loading ports 124 for receiving carrier 122 carrying one or more substrates. Transfer chamber 110 is typically maintained at a vacuum pressure level, while factory interface 106 is typically maintained at atmospheric pressure.

[0027] In some embodiments, the transfer chamber 110, processing chambers 114, 116, and 118, and load lock 120 can be maintained at a vacuum level. The vacuum level of the transfer chamber 110 can range from, for example, about 1 mTorr (or about 5 mTorr, 10 mTorr, 15 mTorr, 20 mTorr, 50 mTorr, 100 mTorr, etc.) to about 80 Torr (or about 1 Torr, 5 Torr, 20 Torr, 50 Torr, etc.). Other vacuum levels can be used.

[0028] A factory interface robot 126 is configured to transfer a substrate from FOUP 122 through a load lock door to a load lock 120. The number of load locks may be more or less than two, but for illustrative purposes only, two load locks 120 are shown, each load lock having a door (e.g., a slit valve) connecting the load lock to the factory interface 106 and a door connecting the load lock to the transfer chamber 110. The load locks 120 may or may not be batch load locks. In this embodiment, the load locks are intelligent load locks capable of performing self-diagnosis and / or automatic prevention and / or recovery.

[0029] Under the control of the controller 150, the load lock 120 can be maintained in an atmospheric pressure environment or a vacuum pressure environment and serves as an intermediate or temporary holding space for substrates being transported back and forth to the transfer chamber 110. The transfer chamber includes a robotic arm 112 configured to transport the substrate from the load lock 120 to one or more of a plurality of processing chambers 114, 116, 118 (also referred to as processing chambers), or to one or more passage chambers (also referred to as through holes), without disrupting the vacuum, i.e., simultaneously maintaining the vacuum pressure environment in the transfer chamber 110 and the plurality of processing chambers 114, 116, 118.

[0030] A door (e.g., a slit valve) connects each corresponding load lock 120 to the transfer chamber 110. Multiple processing chambers 114, 116, and 118 are configured to perform one or more processes. Examples of processes that can be performed by one or more of the processing chambers 114, 116, and 118 include cleaning processes (e.g., pre-cleaning processes to remove surface oxides from a substrate), annealing processes, deposition processes (e.g., for depositing capping layers, hard mold layers, barrier layers, bit line metal layers, barrier metal layers, etc.), etching processes, etc. Examples of deposition processes that can be performed by one or more of the processing chambers include physical vapor deposition (PVD), chemical vapor deposition (CVD), atomic layer deposition (ALD), etc. Examples of etching processes that can be performed by one or more processing chambers include plasma etching processes.

[0031] Controller 150 (e.g., tool and equipment controller, tool cluster controller, etc.) controls various aspects of the cluster tool 100, such as gas pressure in the processing chambers, individual gas flow rates, space flow ratios, plasma power in various processing chambers, temperature of various chamber components, radio frequency (RF) or electrical status of the processing chambers, etc. Controller 150 can receive signals from and send commands to any component of the cluster tool 100, such as robotic arms 112, 126, processing chambers 114, 116, 118, load lock 120, slit valves, and / or one or more sensors and / or other processing components of the cluster tool 100. Controller 150 can therefore control the start and stop of processing, adjust the deposition rate and / or target layer thickness, adjust the processing temperature, adjust the type or mixing of deposition components, adjust the etching rate, and initiate automatic prevention and / or recovery processes on load lock 120, etc. The controller 150 can further receive sensor measurement data from various sensors (e.g., optical measurement data, vibration data, spectral data, particle detection data, temperature data, etc.) and make decisions based on such measurement data.

[0032] In various embodiments, controller 150 may be and / or include computing devices, such as personal computers, server computers, programmable logic controllers (PLCs), microcontrollers, etc. Controller 150 may include (or) one or more processing devices, which may be general-purpose processing devices, such as microprocessors, central processing units, or similar devices. More specifically, the processing devices may be complex instruction set computing (CISC) microprocessors, reduced instruction set computing (RISC) microprocessors, very long instruction word (VLIW) microprocessors, or processors implementing other instruction sets, or processors implementing combinations of instruction sets. The processing devices may also be one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, or similar devices. Controller 150 may include data storage devices (e.g., one or more disk drives and / or solid-state drives), main memory, static memory, a network interface, and / or other components. The processing means of controller 150 may execute instructions to perform any one or more of the methods and / or embodiments described herein. The instructions may be stored on a computer-readable storage medium, which may include main memory, static memory, secondary storage devices, and / or processing means (during instruction execution). In some embodiments, controller 150 is a dedicated controller for load locking 120.

[0033] In this implementation, the controller 150 has increased processing power and memory capacity compared to conventional controllers used for clustering tools. In this implementation, the processing power and memory are sufficient to handle the parallel execution and use of multiple trained machine learning models, as well as the training of machine learning models. For example, the memory and processing power may be sufficient to handle the parallel execution of 2 to 15 (e.g., 3, 4, 5, 6, 7, 8, 9, 10, etc.) different machine learning models (e.g., one or more machine learning models per processing chamber 114, 116, 118 and / or load lock 120).

[0034] In one embodiment, controller 120 includes an autonomous load-locking engine 152. The autonomous load-locking engine 152 may be implemented using hardware, firmware, software, or a combination thereof. The autonomous load-locking engine 152 may be configured to receive and process measurement data generated by one or more sensors of load-locking 120 during and / or after the substrate undergoes a load-locking cycle. Sensor measurements may include temperature measurements, pressure measurements, particle measurements, spectral measurements, vibration measurements, accelerometer measurements, voltage measurements, current measurements, resistance measurements, time measurements, optical measurements (e.g., emission spectrum measurements and / or reflection measurements), position measurements, humidity measurements, component health measurements, and / or other types of measurements. Some example measurements include chamber pressure (e.g., pressure that may be measured in millitors), OES spectral measurements at one or more wavelengths or frequencies (e.g., wavelengths of 3870 nm, 7035 nm, 775 nm, etc.), one or more substrate support / heater temperatures, one or more substrate temperatures, etc. Some or all of these measurements are combined to generate a feature vector, which is then fed into a trained machine learning model of the autonomous tool engine 121.

[0035] The autonomous load locking engine 152 running on controller 150 may include one or more rule-based engines and / or trained machine learning models for controlling one or more load locks and / or making decisions for one or more load locks. The one or more trained machine learning models may have been trained to receive sensor measurements from load lock 120 and / or receive sensor measurements associated with load lock 120, and to predict, classify, or determine load locks. In one implementation, each trained machine learning model may be associated with a different decision-making process for load locking. Alternatively, in another implementation, one or more trained machine learning models may be associated with multiple decision-making processes for load locking.

[0036] In one implementation, one or more of the trained machine learning models are regression models trained using regression. Examples of regression models are those trained using linear regression or Gaussian regression. Given known values ​​of the variable X, the regression model predicts the value of Y. Regression analysis can be used to train the regression model, and this analysis may include interpolation and / or extrapolation. In one implementation, least squares is used to estimate the parameters of the regression model. Alternatively, Bayesian linear regression, percentage regression, least absolute deviation, nonparametric regression, contextual optimization, and / or distance metric learning can be performed to train the regression model.

[0037] In one implementation, one or more of the trained machine learning models are decision trees, random forests, support vector machines, or other types of machine learning models.

[0038] In one implementation, one or more of the trained machine learning models are artificial neural networks (also simply referred to as neural networks). Artificial neural networks may be, for example, convolutional neural networks (CNNs) or deep neural networks. In one implementation, the processing logic performs supervised machine learning to train the neural network.

[0039] Artificial neural networks typically include feature representation components with classifier or regression layers that map features to a target output space. For example, a convolutional neural network (CNN) manages multiple layers of convolutional filters. Pooling is performed at lower layers, which can handle non-linearity. These lower layers are typically topped with multiple perceptrons to map the top-level features extracted by the convolutional layers to a decision (e.g., a classification output). Neural networks can be deep networks with multiple hidden layers or shallow networks with zero or a few (e.g., one to two) hidden layers. Deep learning is a class of machine learning algorithms that use cascades of multiple non-linear processing units for feature extraction and transformation. Each subsequent layer uses the output from the previous layer as input. Neural networks can learn in a supervised (e.g., classification) and / or unsupervised (e.g., pattern analysis) manner. Some neural networks (e.g., deep neural networks) include hierarchical structures where different layers learn different levels of representation corresponding to different levels of abstraction. In deep learning, each level learns to transform its input data into a slightly more abstract and complex representation.

[0040] One or more of the trained machine learning models can be recurrent neural networks (RNNs). An RNN is a type of neural network that includes memory, enabling it to capture temporal dependencies. RNNs learn input-output mappings that depend on current and past inputs. An RNN processes past and future measurements and makes predictions based on this continuous measurement information. For example, sensor measurements can be continuously taken during a process, and these sets of measurements can be sequentially fed into an RNN. Current and previous sensor measurements can influence the current output of the trained machine learning model. One type of RNN that can be used is a long short-term memory (LSTM) neural network.

[0041] Some trained machine learning models of the autonomous load locking engine 152 use all sensor measurements generated by load locking. Some trained machine learning models of the autonomous load locking engine 152 use a subset of the generated sensor measurements.

[0042] In one implementation, the autonomous load lockout engine 152 includes an automatic prevention and / or recovery manager 154. The automatic prevention and / or recovery manager 154 may include one or more rule-based systems and / or one or more trained machine learning models trained to receive sensor measurements of load lockout and output decisions on whether preventive or recovery actions (such as maintenance) should be performed on the load lockout.

[0043] Controller 150 is operatively connected to a server (not shown). The server may be or include a computing device operating as a factory floor server, connected to some or all of the tools in the manufacturing facility. The server can perform training to generate a trained machine learning model and can send the trained machine learning model to an autonomous load-locking engine 152 on controller 150. Alternatively, the machine learning model can be trained on controller 150.

[0044] Neural networks can be trained using supervised learning, which involves feeding the network a training dataset consisting of labeled inputs, observing its output, defining the error (by measuring the difference between the output and the labeled value), and using techniques such as deep gradient descent and backpropagation to tune the weights of all layers and nodes of the network to minimize the error. In many applications, repeating this process with many labeled inputs in the training dataset produces a network that can produce the correct output when given inputs different from those in the training dataset. This generalization is achieved in high-dimensional settings, such as large images, when sufficiently large and diverse training datasets are available.

[0045] In another exemplary embodiment, Figure 1B A load locking system 1000 associated with a load lock 1100 or a load lock chamber 1100 is shown (in this disclosure, the load lock and the load lock chamber are interchangeable for further operation), which may be associated with a substrate processing system. The load locking system 1000 may include a computing subsystem 1300 (discussed in further detail with respect to Figures 3 and 4), a sensing subsystem 1600, a complex atom system 1700, and / or the load lock 1100. The load lock 1100 may further include a first gate portion 1110, a second gate portion 1120, and a substrate support device 1130. In embodiments, the load locking system 1000 corresponds to a system consisting of… Figure 1A The autonomous load locking engine 152 controls the load locking 120.

[0046] In some embodiments, the substrate 1140 may be placed on the support device 1130. In some embodiments, the substrate support device 1130 may not support the substrate. In some embodiments, the substrate support device 1130 may support more than one substrate (not shown in the figures). The substrate support device 1130 may include a heating support, a cooling support, a chuck (e.g., an electrostatic chuck or a vacuum chuck), etc.

[0047] In some embodiments, the substrate support device 1130 may be configured to move vertically upward or downward within the load locking chamber 1100. In some embodiments, the substrate support device 1130 may be configured to rotate within the chamber.

[0048] In further embodiments, the vertical movement of the substrate support device 1130 within the load locking chamber 1100 can be achieved by a mechanical actuator (not shown in the figures). Those skilled in the art, benefiting from this disclosure, will be able to design various methods and systems to achieve the above-mentioned functions using various types of mechanical actuators.

[0049] The load-locking chamber in a vacuum processing system is used to load and unload substrates, such as silicon wafers or other substrates, without disrupting the vacuum environment of the main processing chamber or transfer chamber.

[0050] A substrate support for a load-locking chamber typically refers to a structure or device for securing a substrate in place. This structure or device is designed to securely hold the substrate while ensuring easy and undamaged movement of the substrate into and out of the load-locking chamber. Some substrate supports are platforms or trays for placing substrates thereon. These substrate supports may be static or include mechanisms for rotation or other movement, such as vertical movement. Some substrate supports may also include clamping or other securing mechanisms to hold the substrate in place, particularly during any movement. In some embodiments, the substrate support has thermal control capabilities. For example, the substrate support may be heated or cooled to heat or cool the supported substrate and / or maintain the substrate at a specific temperature. The substrate support may include embedded heating elements that apply resistance heating in one or more areas, including optical heating, etc. The substrate support may additionally or alternatively include cooling mechanisms, such as channels through which coolant flows, to provide liquid cooling to the supported substrate.

[0051] Substrate 1140 can be conveyed through first door portion 1110 to load locking chamber 1100, and then conveyed out through second door portion 1120 for processing. Similarly, substrate can be conveyed through second door portion 1120 to load locking chamber 1100 and conveyed out through first door portion 1110. The first and second door portions 1110, 1120 may be ports including slit valves, which can be opened to allow a robotic arm to pick up or place substrates from / on the substrate support, and can be closed to seal the interior of the load locking chamber. The pressure in the load locking chamber can then be adjusted. The first and / or second substrate (not shown in the figures) can be placed on support device 1130. Those skilled in the art, benefiting from this disclosure, will be able to conceive of various mechanisms and structures for support device 1130 to support one or more substrates.

[0052] Door portions 1110 and 11120 may include any type of sealable portion that provides a pressure seal against one or more external pressures of the load lock. In some embodiments, the load lock can control the ambient pressure of the chamber and raise or lower the pressure according to the overall needs of the electronic component or substrate manufacturing system.

[0053] In some embodiments, more than two door portions may be used. In some embodiments, door portions 1110 and 1120 may be adjacent to each other or inserted into the orthogonal sidewalls of chamber 1100.

[0054] In some embodiments, the sensing subsystem 1600 of the load-locking system 1000 may include a critical component health monitor 1322, which may include sensors on the first and second door portions 1110 and 1120, such sensors being physical distortion sensors 1632 and 1634 located near the door portions. These sensors may provide measurements to the sensing subsystem 1600 and, on the same system, to a control algorithm 1316 to sense the integrity of the mechanism associated with the first and second door portions 1110 and 1120. In this way, sensors 1632 and 1634 can monitor the component health of the load-locking 1100 door mechanism and allow the control algorithm to signal to the user of the manufacturing system when the door mechanism requires maintenance.

[0055] In some embodiments, the critical component health monitor 1322 may alternatively or additionally include a vibration sensor and / or accelerometer (not shown in the figures) coupled to the substrate support 1130 or a mechanical actuator (not shown in the figures) for moving the substrate support. In some embodiments, this vibration sensor and / or accelerometer may be used to transmit information about the integrity or malfunction of the actuator coupled to the substrate support. In this way, the critical component health monitor 1322 can monitor the component health of the substrate support 1130 and the attached actuator, and allow the control algorithm to signal to the user of the manufacturing system when the substrate support 1130 or the attached actuator requires maintenance.

[0056] In some embodiments, sensors 1632 and 1634 may include one of a capacitive sensor, a piezoresistive sensor, a strain gauge, or any type of distortion sensor commonly used in the manufacture of electronic components. In some embodiments

[0057] In some embodiments, the sensing subsystem 1600 may include a particle sensor 1320, which includes sensor elements 1602 and 1604. In some embodiments, the particle sensor 1320 may be sensor elements 1602 and 1604, which may include one or more, or any combination of one or more of the following: light scattering sensor, impact sensor, aerosol electrometer, mass spectrometer, residual gas analyzer (RGA) sensor, weight sensor, surface acoustic wave (SAW) sensor, corona discharge sensor, or any other particle sensor commonly used in electronic manufacturing systems.

[0058] In some embodiments, the light scattering sensor may include one or more imaging sensors. In some embodiments, one or more imaging sensors may generate images of different regions within the load lock 1110. In some embodiments, images of different regions within the load lock 1110 from one or more imaging sensors may be stitched together to produce a stitched image of the entire inner surface of the chamber. In some embodiments, the imaging sensor captures the visible spectrum. In some embodiments, the imaging sensor captures thermal radiation (e.g., infrared and / or near-infrared light), ultraviolet light, and / or visible light.

[0059] although Figure 1B Only two physical spaces for placing elements 1602 and 1604 are shown, but those skilled in the art who benefit from the content of this disclosure can conceive of many such configurations for placing one or any number of such particle sensors and combinations thereof within the load lock 1100.

[0060] In some embodiments, particle sensor 1320 may include corona discharge sensor 1624, which includes a cathode 1626 and an anode 1628. In some embodiments, the cathode 1626 and anode 1628 may be integrally placed with the sidewall of load lock 1100. In other embodiments, the cathode 1626 and anode 1628 may protrude into the load lock chamber 1100. In some embodiments, corona discharge sensor 1624 may be used in conjunction with an electrometer within the chamber to perform particle detection. The corona discharge sensor (commonly referred to as a corona discharge ionization detector) may be combined with an electrometer for particle detection, such as air quality monitoring and aerosol particle detection. Corona discharge is a process in which current flows from electrodes with a high potential into a neutral fluid (typically air), ionizing the fluid and creating a plasma region around the electrodes. The electrodes are typically a thin metal wire to which a high voltage is applied. The high electric field near the metal wire causes ionization and results in current flow in the air. As particles pass through the corona discharge region, they become ionized or charged (e.g., they gain or lose electrons, thus acquiring a net positive or negative charge). Once charged, the particles move towards a collecting electrode under the influence of an electric field. This collecting electrode is connected to an electrometer.

[0061] An electrometer is a highly sensitive measuring instrument that measures electric charge or potential difference. When charged particles strike the collecting electrode, they generate a small current that is measured by the electrometer. By measuring the charge collected over time, the number of particles that have passed through the detector can be inferred. This is because each particle carries a certain amount of charge, so by measuring the total collected charge and knowing the charge of each particle, the total number of particles can be calculated. Furthermore, larger particles tend to carry more charge, making it possible to determine the size distribution of the particles. This makes it possible to count particles in a gas or air sample, and sometimes determine their size. Subsequently, these measurements can be used to determine when to perform one or more remedial actions, such as automated cleaning operations. For example, measurements from any of the aforementioned particle sensors can be used to determine when to invoke a complex atom system.

[0062] although Figure 1B Two physical spaces for placing elements 1626 and 1628 are illustrated, but those skilled in the art who benefit from the content of this disclosure will be able to conceive of many such configurations for the cathode 1626 and anode 1628 to be placed within the load lock 1100 or for any additional cathode and anode.

[0063] In some embodiments, the particle sensor 1320 can be used to detect foreign materials or contaminants within the load-locking chamber 1100. In some embodiments, the contaminant may be a solid or particulate matter; in other embodiments, such solid or particulate matter may be a gas or vapor.

[0064] In some embodiments, the environmental sensor 1640, which includes sensor elements 1642 and 1644, may be one or more of the following, or any combination of one or more of the following: an ambient temperature sensor, a pressure sensor, a relative humidity (RH) sensor, a surface acoustic wave (SAW) sensor, an electrometer with or without a corona discharge sensor, or any other type of environmental sensor commonly used in electronic component manufacturing systems.

[0065] In some embodiments, the environmental sensor 1640 may be placed in the top plate, bottom plate, or side wall of the load locking chamber 1100 and sense one or more environmental characteristics associated with the load locking chamber 1100. Although Figure 1B Two physical spaces for placing elements 1642 and 1644 are illustrated, but those skilled in the art, upon receiving this disclosure, will be able to envision many such configurations and locations for placing one or any number of such environmental sensors within the load lock 1100. In embodiments, measurements from the environmental sensors can be used to determine the state of the load lock chamber and whether one or more repair actions need to be performed.

[0066] In some embodiments, the substrate sensor 1670 may be disposed within or on the load-locking chamber 1100. In some embodiments, the substrate sensor 1670 may be a mass spectrometer placed above a substrate. In some embodiments, the substrate sensor 1670 may also be a particle detector, such as sensor element 1602 or 1604. Although Figure 1B A physical space for placing sensor 1670 is illustrated, but those skilled in the art who benefit from this disclosure will be able to envision many such configurations and locations for placing one or any number of such substrate sensors within load lock 1100.

[0067] exist Figure 1B In one exemplary embodiment, the load lockout system 1000 includes a complex atom system 1700 associated with the load lockout system 1100. In some embodiments, the complex atom system 1700 includes a repair subsystem 1704 and a prevention subsystem 1706. In some embodiments, the repair subsystem 1704 may include a purification subsystem 1326, a thermal subsystem 1720, an aerosol subsystem 1730, a UV subsystem 1328, and / or a user instruction subsystem 1760. In some embodiments, the purification subsystem 1326 may further include an exhaust line 1342 and a pump line 1344. In some embodiments, the user instruction subsystem 1760 may include notifications to the user transmitted via server 1310 to the user interface 1306.

[0068] In some embodiments, the prevention subsystem 1706 may include a trap subsystem 1332, which includes a cold trap 1752 and / or an electrostatic trap 1754. In some embodiments, the cold trap 1752 may utilize exhaust and pump lines 1342 and 1344. In some embodiments, the electrostatic trap further includes a cathode 1756 and an anode 1758.

[0069] The cold trap 1752, also known as a cryogenic trap, is a device used in scientific and industrial applications to condense all vapors except permanent gases into a liquid or solid. Cold traps are used in vacuum systems to prevent contaminants from reaching the vacuum pump, or in cryogenic systems to remove residual gases. Cold traps operate by typically using liquid nitrogen or mechanical refrigeration systems to cool the chamber to extremely low temperatures. The temperature should be low enough to condense the specific vapors the trap is designed to capture. When the vacuum system is operating, vapor travels from the vacuum chamber toward the vacuum pump. As the vapor passes through the cold trap, it impacts the cold surface of the cold trap 1752. Due to the low temperature, the vapor condenses on the surface, changing from a gas to a solid or liquid. This effectively removes and traps vapors from the gas stream.

[0070] Once the vapors are condensed and captured, they are retained in the cold trap 1752, thus preventing them from reaching the vacuum pump. This helps protect the pump and maintain the quality of the vacuum. Furthermore, this prevents vapors and particles from reaching the substrate in the load-locking chamber. When there is no substrate in the load-locking chamber, once the captured material is heated and restored to a less volatile state, it can be safely disposed of later (e.g., as part of an automated cleaning process).

[0071] The effectiveness of a cold trap depends on the type of vapor present (because different substances condense at different temperatures) and the temperature of the cold trap (lower temperatures allow for the condensation of more types of vapor). The trap's design and surface area also affect its performance.

[0072] An electrostatic trap 1754, also known as an ion trap, is a device that uses an electric field to confine charged particles (usually ions). The electrostatic trap 1754 operates by ionizing particles (unless the particles are already ionized). Ionization can be achieved through several methods, such as electron bombardment, photoionization, or chemical ionization. The electrostatic trap 1754 generates an electrostatic field with a geometry that confines the charged particles. This is typically accomplished using a set of electrodes to which a voltage is applied. Ions are trapped in this electric field. The plasma is subjected to an electric current that propels the ions toward regions of minimum (for positively charged ions) or maximum (for negatively charged ions) potential within the trap.

[0073] The stability of the electrostatic trap 1754 (i.e., its ability to retain trapped ions) depends on the specific geometry of the electric field and the specific voltage applied to the electrodes. Ion traps come in various designs, including Penning traps (using both electric and magnetic fields) and Paul traps (using time-varying electric fields). Once trapped, ions can be manipulated for a variety of purposes. For example, ions may be cooled to extremely low temperatures, excited to high energy states, or undergo chemical reactions with other ions or neutral particles.

[0074] In some embodiments, the purification subsystem may use exhaust lines and pump lines 1342 and 1344 to purify the load locking chamber 1100 with gas, liquid, vapor or similar substances.

[0075] In some implementations, the purification subsystem may use a purification medium containing nitrogen, carbon dioxide, dry air, or any other purification medium commonly used in electronics manufacturing.

[0076] In some embodiments, the maximum chamber pressure, fluid flow rate, and other characteristics of the purification method used in the purification subsystem 1326 may vary depending on the application of the purification subsystem 1326. Figure 5A The standard purification method shown is 5000 or Figure 5B The purification and cleaning method 5010 is shown. Each of these purification methods will be described in further detail below.

[0077] In some embodiments, the purification subsystem 1326 can perform a purification cleaning cycle that can be repeated any number of times and is repeated automatically, but does not involve opening the load-locked chamber to the external environment. In embodiments, the purification cleaning cycle can be performed when no substrate is placed in the load-locked chamber, or when a test wafer is placed in the load-locked chamber.

[0078] In some embodiments, the thermal subsystem 1720 may undergo thermal cycling to remove moisture or other materials by utilizing the high heat within the load-locking chamber 1100. In some embodiments, thermal cycling may include raising the temperature of the load lock 1100 to 150%, 200%, or 300% of the normal operating temperature of the load lock chamber 1100, and then lowering the temperature back to the normal operating temperature. In some embodiments, thermal cycling may be repeated any number of times. In some embodiments, one or more thermal cycles are performed when one or more substrates are disposed in the load lock chamber. In some embodiments, one or more thermal cycles are performed when no substrates are disposed in the load lock chamber.

[0079] In some embodiments, the aerosol subsystem 1730 may be a cryogenic aerosol subsystem designed to freeze the particles so that the particles can subsequently be removed by a purification system.

[0080] In some implementations, the cryogenic aerosol subsystem may utilize cryogenic fluids, including liquid nitrogen, liquid helium, liquid oxygen, liquid argon, liquid CO2, or any other type of cryogenic fluid commonly used in electronics manufacturing.

[0081] Cryo-aerosol systems, also known as cryo-aerosol cleaning systems, are used for particle removal or contamination control in highly sensitive manufacturing environments such as semiconductor wafer fabrication, optical component manufacturing, and high-precision manufacturing processes. To perform cryo-aerosol cleaning, a cryo-aerosol is generated. This is typically accomplished by using a cryogenic liquid (such as liquid nitrogen or liquid argon) and rapidly expanding it to form a cold gas. As the gas expands, it cools sufficiently to freeze small amounts of moisture in the air, forming a clump of tiny ice particles. This clump of particles forms the cleaning aerosol. The cryo-aerosol is then directed to the surface to be cleaned (e.g., into the interior of a load-locking chamber). The tiny ice particles in the aerosol collide with contaminant particles on the surface of the load-locking chamber, effectively removing the contaminant particles. Due to the extremely low temperature of the aerosol, the kinetic energy and thermal shock of the aerosol particles, as well as the contraction of the contaminant particles, contribute to the removal process. After the ice particles remove the contaminant particles, both the ice and the contaminants sublimate or evaporate (because the temperature of the aerosol is much lower than the ambient temperature, the aerosol quickly turns back into a gas once it is no longer cooled by contact with the cryogenic liquid). This process may leave little or no residue on the surface, and gaseous contaminants can be removed from the environment using appropriate filtration or extraction systems.

[0082] Low-temperature cleaning processes are dry, non-abrasive, and residue-free, making them suitable for applications where conventional cleaning methods may be inapplicable or ineffective. This process is particularly useful for removing small particles (micron and submicron in size) and cleans surfaces without damaging sensitive components or leaving any residue that could interfere with the performance of the cleaning device.

[0083] In some implementations, the UV subsystem 1328 may emit short-wavelength UV light to decompose organic materials present within the chamber, after which the purification subsystem 1326 may be used to remove the organic materials. Ultraviolet (UV) cleaning systems use UV light, particularly UV-C light, to disinfect surfaces and air. UV-C light is bactericidal, meaning it can inactivate microorganisms such as bacteria, viruses, and fungi. Cleaning systems include a UV-C light source, typically a lamp or LED that generates UV-C light, which is electromagnetic radiation with wavelengths in the range of 200 to 280 nanometers. The system exposes a target area—whether a surface, liquid, or air—to UV-C light. UV-C light can be absorbed by the DNA and RNA of microorganisms. This absorption leads to a process called photodimerization, in which adjacent pyrimidine bases in the DNA or RNA form covalent bonds. This disrupts the structure of the genetic material and prevents the microorganisms from multiplying and infecting. Essentially, this inactivates the microorganisms, rendering them harmless.

[0084] UV cleaning systems offer several advantages, including the absence of chemicals, the absence of residues, and the ability to disinfect air, water, and surfaces. The effectiveness of UV cleaning depends on various factors, such as the intensity and duration of UV-C exposure, the distance from the light source, and the presence of any particles that could block light for microorganisms. Therefore, UV cleaning systems are often used as part of a broader suite of cleaning and disinfection strategies. For example, the UV subsystem 1328 can be used in conjunction with an aerosol subsystem and / or a purification subsystem.

[0085] In some implementations, the repair subsystem may be triggered by control algorithm 1316 (see below). Figure 3 The control algorithm 1316 can be executed by the automatic prevention / recovery manager 154 of the autonomous load locking engine 152, and the prevention subsystem (such as the trap subsystem 1332 including the cold trap 1752 and the electrostatic trap 1754 including the cathode 1756 and the anode 1758) can be selectively triggered by the control algorithm 1316, or can otherwise be always enabled. In some embodiments, the electrostatic trap 1754 may include the cathode 1756 and the anode 1758. In some embodiments, the cathode 1756 and the anode 1758 may be integrally placed with the sidewall of the load lock 1100. In other embodiments, the cathode 1756 and the anode 1758 may protrude into the load lock chamber 1100. Although Figure 1B Two physical spaces for placing elements 1756 and 1758 are illustrated, but those skilled in the art who benefit from this disclosure will be able to conceive of many such configurations for the cathode 1756 and anode 1758 in the load lock 1100, or any additional cathodes and anodes.

[0086] In some embodiments, the electrostatic trap 1754 subsystem is designed to attract charged particles during use load locking 1100. During gas purging of the system, the charge on the electrostatic trap 1754 can be turned off, allowing particles normally held and attracted by the trap 1754 subsystem to be removed via gas purging 1326. In other embodiments, the electrostatic trap 1754 may remain active during gas purging.

[0087] In some embodiments, cold trap 1752 may introduce a cold fluid via exhaust and pump lines 1342 and 1344, which is designed to cool the surface of load lock 1100 or any other surface in the chamber. When cold trap 1752 begins operation, gases or substances within the chamber may condense onto the chamber's cold surfaces, including the sidewalls, top plate, and bottom surface. The condensed material can then be removed from the chamber using the purification method of subsystem 1326.

[0088] Figure 2 An exemplary “smart” and unconventional production process 2000 is illustrated, which utilizes sensing and complex atom systems 1600 and 1700. This “smart” and unconventional production process 2000 may be the opposite of conventional or known processes used for load-locking preventative maintenance.

[0089] In one example of a conventional process for load-locked preventative maintenance, the conventional production process may include production board operation, routine component identification, time-based scheduling identification, component maintenance, and post-maintenance identification.

[0090] In a conventional system, problems within the load-locking chamber or substrate may not be discovered until routine particle identification or time-based scheduled identification. These routines and schedules can be spaced out, and maintaining the load-locking system and chamber requires strict punctuality and care. Furthermore, the inspection and identification process is susceptible to human error and may be performed inconsistently or unsatisfactorily.

[0091] If a problem is identified with the substrate or load-locking system, intensive component maintenance can be performed. As mentioned above, in addition to the negative attributes of this process, the manufacturing system itself may be shut down, and production may be suspended when component maintenance is performed on the load-locking system.

[0092] Once component maintenance is complete, post-maintenance qualification must be performed and passed before production can continue. These qualifications only highlight some issues in conventional processes for load-locking preventative maintenance used in production, including inspecting and maintaining load-locking systems in a manufacturing environment.

[0093] In a single instance, and in some implementations, a non-traditional manufacturing process may be described, which includes a production substrate operation 2100, a smart sensing stage 2200, and a smart recovery stage 2300.

[0094] exist Figure 2 In at least one of the embodiments shown, a non-invasive smart sensing phase 2200 is performed during and / or after each production substrate run 2100. Production substrate run 2100 may include inserting the production substrate into a load lock through a first port and placing the production substrate onto a substrate support within the load lock by a first robot. The production substrate can be placed into the load lock when it is in a vacuum. Gas inlets or outlets (e.g., purge lines) may be opened to a minimum to allow for a slow pressure increase within the load lock. Furthermore, the production substrate may be cooled within the load lock. Once the production substrate has been cooled and / or the load lock has reached a target pressure (e.g., atmospheric pressure), a second port may be opened, and a second robot may retrieve the production substrate from the load lock.

[0095] The non-invasive sensing phase 2200 may be performed concurrently with the production board operation 2100, or after the production board operation 2100 has completed. The non-invasive sensing phase 2200 may include performing one or more non-invasive measurements using any of the sensors described above. A rule-based engine and / or one or more trained machine learning models may be used to process the measurements to determine whether any corrective action should be performed.

[0096] The use of the intelligent sensing phase 2200 allows for more frequent investigations and measurement degradation within the load lockout system 1000. Unlike conventional process flows with long intervals between inspections, process flow 2000 can continuously and autonomously scan for errors in the load lockout system 1000.

[0097] When an error occurs and / or the processing logic determines that a repair action should be performed, the system 2000 immediately detects such an error and / or notifies the user of recommended maintenance or the use of the intelligent recovery phase 2300 and the re-atomic system 1700 to resolve the problem.

[0098] In some implementations, the intelligent restoration phase 2300 may include activating at least one or any combination of subsystems corresponding to the complex atom system 1700 (such as in...). Figure 1B (As discussed in detail below). This may include enabling any one or any combination of the subsystems of the repair subsystem 1704 or the prevention subsystem 1706. This may include enabling any one or more of the purification subsystem 1326, the thermal subsystem 1720, the aerosol subsystem 1738, the UV subsystem 1328, the user indication subsystem 1760, and the trap subsystem 1332, including the cold trap 1752 or the electrostatic trap 1754.

[0099] The non-invasive smart recovery phase 2300 can be executed simultaneously with the production substrate operation 2100, or after the production substrate operation 2100 has been completed. The non-invasive smart recovery phase 2300 can be executed when one or more substrates are present in the chamber 1100, or when there are no substrates in the chamber. The non-invasive smart recovery phase can be executed automatically and autonomously, or it can be enabled by the system user.

[0100] The production substrate operation 2100, the smart sensing phase 2200 and / or the smart recovery phase 2300 can be executed by the controller in the embodiment, such as by the autonomous load locking engine 152 of FIG1.

[0101] In one exemplary embodiment, Figure 3 A computing subsystem 1300 is shown (e.g., this computing subsystem may correspond to the controller 150 of FIG. 1), which can be used to control sensing and re-atomic systems 1600 and 1700. The computing subsystem 1300 may include a computing device 1310, which includes a processor 1312 that electronically communicates with memory 1314 including a control algorithm 1316. The processor 1312 may correspond to the autonomous load locking engine 152 and / or the automatic prevention / recovery manager 154 of FIG. 1. The computing device 1310 can electronically communicate with subsystems of the load locking system 1000 via a communication channel 1352, which includes a communication channel 1360 for receiving input from the sensing subsystem 1600, a communication channel 1370 for transmitting instructions to the re-atomic system 1700, and a communication channel 1380 for transmitting and receiving information to and from the user interface 1306.

[0102] In some embodiments, computing device 1310 may continuously receive data from communication channel 1360. In some embodiments, server processor 1312 may execute rule-based control algorithm 1316 to select whether to activate restoration system 1700 via communication channel 1370. In some embodiments, control algorithm 1316 may communicate with a pre-trained machine learning model that can acquire specific sensor data from communication channel 1360 and output specific instructions regarding which repair subsystems 1704 of restoration system 1700 should be activated, when to activate repair subsystems 1704, etc.

[0103] In some embodiments, detecting foreign particles within the load-locking chamber 1100 may cause the control algorithm 1316 to trigger the purification subsystem 1326, the electrostatic trap subsystem 1754, the cryogenic aerosol subsystem 1732, or any other combination of such mechanisms to remove the foreign particles from the load-locking chamber 1100 (e.g., release captured particles, followed by purification of the captured particles). Those skilled in the art, benefiting from the content of this disclosure, will be able to conceive of numerous additional subsystems for removing foreign particles from the load-locking chamber 1100.

[0104] In some embodiments, detecting organic contaminants within the load-locking chamber 1100 may cause the control algorithm 1316 to trigger the UV light emitter 1740, the cold trap 1752, the gas purification system 1326, or any combination of such mechanisms to remove the organic contaminants. Those skilled in the art, benefiting from the content of this disclosure, will be able to conceive of numerous additional subsystems for removing organic material from within the load-locking chamber 1100.

[0105] In some embodiments, detecting moisture within the load-locking chamber 1100 may cause the control algorithm 1316 to trigger the thermal subsystem 1720 or heater within the load-locking chamber 1100 to reduce humidity. Additionally, a purification subsystem 1326 may be used to remove moisture from the load-locking chamber 1100. Those skilled in the art, benefiting from the content of this disclosure, will be able to conceive of numerous additional subsystems for removing moisture from the load-locking chamber 1100.

[0106] In some implementations, detecting component failure can cause the control algorithm 1316 to trigger an instruction to the staff or user of the system 1000, indicating that the component needs to be replaced and / or maintained.

[0107] In one exemplary embodiment, Figure 4 This illustrates how control algorithm 1316 can communicate with various associated subsystems 1315, including sensing subsystem 1600, a complex atom system 1700 including repair subsystem 1704 and prevention subsystem 1706, and a chamber subsystem 1780 including vacuum subsystem 1336 and mechanical subsystem 1342, as shown above. Figure 1B The following are descriptions of the implementation details: In some embodiments, the sensing subsystem 1600 may include a particle sensor 1320 and a critical component health monitor 1322. In some embodiments, the repair subsystem 1704 may include a purification subsystem 1326 or a UV organic matter removal subsystem 1328. In some embodiments, the prevention subsystem 1706 may include a particle trap subsystem 1332. In some embodiments, the vacuum subsystem 1336 may include an exhaust port 1338 and pump 1340 hardware. In some embodiments, the mechanical subsystem 1342 may include a substrate transfer mechanism 102.

[0108] like Figure 4 As shown in the embodiment, the control algorithm 1316 can communicate directly with multiple associated subsystems 1315 to perform detection and recovery in a smooth and efficient manner.

[0109] In one embodiment, the purification subsystem 1326 can be controlled to perform one or more purification cleaning cycles to remove particulate buildup from inside the load-locked chamber. The purification cleaning cycle may include actively venting air and / or gas into the load-locked chamber for a short period and pumping the load-locked chamber to extremely low pressures, which in this embodiment may be lower than the standard operating pressure of the load lock.

[0110] exist Figure 5A In one exemplary embodiment shown, the purification subsystem 1326 can perform a purification method as a standard purification method 5000. Purification feature diagram 5100 illustrates a feature diagram of one embodiment of the standard purification method 5000. Purification feature diagram 5100 illustrates an exhaust circulation and a pump circulation, in which gas and / or air are slowly discharged into a load-locking chamber, while in the pump circulation, the load-locking chamber is pumped down to a target pressure. Purification feature diagram 5100 illustrates the pressure, flow rate, and pump status of each in the exhaust circulation and pump circulation. Purification method 5000 can be performed, for example, when the substrate is in the load-locking chamber and when the load-locking chamber is considered "clean" (e.g., the autonomous load-locking engine 152 has not yet indicated that an automatic cleaning action should be performed).

[0111] In some embodiments, feature graph 5100 may include the start of an exhaust “operation time” 5112 during which the purging medium is introduced into the load-locked chamber 1100. In some embodiments, purging method 5000 may include a maximum flow rate. In some embodiments, feature graph 5100 may include a chamber pressure feature line 5150, which includes a maximum pressure represented by 5158. In some embodiments, purging method 5000 may include a waiting time between the exhaust and pump suction operation times. In some embodiments, purging method 5000 may include a pump operation time, which indicates the amount of time the pump is started. As shown, after the exhaust “operation time” 5112, the pressure rises slowly, indicating that the pressure begins to rise gradually.

[0112] In some embodiments, the slope of the pressure 5150 can represent the rate of pressure change within the chamber. In some embodiments, a steeper slope can indicate a more rapid or aggressive introduction of the purification medium into the chamber. Therefore, in some embodiments, a steeper slope indicates a purification method that is more effective in removing material from the chamber due to the faster introduction of the purification medium. A shallower slope can indicate a more gradual or subtle introduction of fluid into the chamber. While maximum flow rate and maximum pressure are some characteristics of a gas purification system, the rate of change of the flow rate and pressure of the gas purification medium and the chamber are also useful characteristics for measuring the effectiveness of a gas purification method.

[0113] In some embodiments, the decreasing slope of pressure 5150 may represent the rate at which the pressure in the chamber decreases during pump operation. In some embodiments, a steeper slope may indicate a more rapid or aggressive extraction of the purification medium from the chamber. A shallower slope may indicate a gentler or more gradual extraction of fluid from the chamber. The rate of change of chamber pressure and the minimum pressure reached are useful characteristics for measuring the effectiveness of the purification method.

[0114] exist Figure 5B In one exemplary embodiment shown, the purification subsystem 1326 may include a more aggressive purification cleaning method 5010 than the standard purification method 5000. In this embodiment, the purification cleaning method 5010 may be performed as a remedial action to clean a load-locked interior. A purification feature chart 5110 is a feature chart of one embodiment of the purification cleaning method 5010. In some embodiments, the purification feature chart 5110 may include the start of an exhaust “operation time” 5114. In some embodiments, the purification cleaning method feature chart 5110 may include a pressure feature line 5160, which includes a maximum pressure 5165. In some embodiments, the purification cleaning method 5010 may include a waiting time between the exhaust “operation time” and the pump “operation time.” In some embodiments, the purification cleaning method 5010 may include a pump “operation time.” In some embodiments, the purification cleaning method feature chart region 5110 may include a flow feature line 5160 containing a slope. As shown, after the start of the exhaust “operation time” 5114, the pressure may immediately begin to increase rapidly, without an initial gradual pressure increase. This may agitate the particles and cause them to become suspended in the air. Subsequently, the airborne particles can be expelled from the chamber, thus effectively cleaning the chamber.

[0115] In some embodiments, the slope of the pressure may represent the rate at which the pressure changes within the chamber. In some embodiments, a steeper slope may indicate a more rapid or aggressive introduction of the purification medium into the chamber. Thus, in some embodiments, a steeper slope indicates a purification method that is more effective in removing material from the chamber due to the faster introduction of the purification medium. A shallower slope may indicate a more gradual or subtle introduction of fluid into the chamber.

[0116] In some embodiments, method 5010 may include pump “operation time”. In some embodiments, feature graph region 5110 may include pressure feature line 5160, which includes a decreasing pressure slope of pump operation time.

[0117] In some implementations, the slope of the pressure decrease can represent the rate of pressure change within the chamber. In some implementations, a steeper slope can indicate a more rapid or aggressive extraction of the purification medium from the chamber. A shallower slope can indicate a gentler or more gradual extraction of fluid from the chamber. The rate of change of chamber pressure and the minimum pressure reached are useful characteristics for measuring the effectiveness of the purification method.

[0118] Through comparison Figures 5A to 5B As can be seen from purification methods 5000 and 5010, in one embodiment of the purification method, the initial flow rate and initial flow velocity during purification cleaning method 5010 are both higher than those of standard purification method 5000. In some embodiments, standard purification method 5000 starts with an initial purification flow rate between 0 and 1 m / s, while purification cleaning method 5010 starts with an initial purification flow rate greater than 2 m / s (e.g., greater than 3 m / s, greater than 4 m / s, greater than 5 m / s, greater than 6 m / s, greater than 7 m / s, greater than 8 m / s, etc.). In some embodiments, the standard purification method 5000 begins with an initial purification flow rate of less than 100 sccm / L, while the purification method 5010 begins with an initial purification flow rate greater than 100 sccm / L (e.g., greater than 200 sccm / L, greater than 300 sccm / L, greater than 400 sccm / L, greater than 500 sccm / L, greater than 600 sccm / L, greater than 700 sccm / L, greater than 800 sccm / L). In some embodiments, the initial time period may be defined as from the start of the purification method until the chamber pressure reaches the target pressure (e.g., 10 torts, 20 torts, 30 torts, 40 torts, 50 torts, etc.).

[0119] From comparison Figures 5A to 5BAs can be seen from purification methods 5000 and 5010, in one embodiment of the purification method, the maximum flow rate during the standard purification method 5000 is lower than the maximum flow rate during the purification and cleaning method 5010.

[0120] Through comparison Figures 5A to 5B As can be seen in purification methods 5000 and 5010, in one embodiment of the purification method, the maximum chamber pressure during the standard purification method 5000 is higher than the maximum chamber pressure during the purification method 5010. In some embodiments, the maximum chamber pressure can be sustained for a period of time, while the maximum chamber pressure can be sustained for a very short period of time, or not at all.

[0121] Through comparison Figures 5A to 5B As can be seen from purification methods 5000 and 5010, in one embodiment of the purification method, the duration of the exhaust "operation time" during the standard purification method 5000 is much longer than the duration of the exhaust "operation time" during the purification and cleaning method 5010.

[0122] Through comparison Figures 5A to 5B As can be seen from purification methods 5000 and 5010, in one embodiment of the purification method, the waiting time between the exhaust phase and the pump phase during the standard purification method 5000 is much longer than the waiting time between the exhaust phase and the pump phase during the purification and cleaning method 5010.

[0123] Through comparison Figures 5A to 5B As can be seen from purification methods 5000 and 5010, in one embodiment of the purification method, the duration of the pump "operation time" during the standard purification method 5000 is much shorter than the duration of the pump "operation time" during the purification and cleaning method 5010.

[0124] Through comparison Figures 5A to 5B As can be seen from purification methods 5000 and 5010, in one embodiment of the purification method, a lower load-locking chamber pressure is achieved at the end of the purification cleaning method 5010 compared to the standard purification method 5000.

[0125] Figure 6A graphical representation of a machine in an example form of computing device 600 is shown, in which a set of instructions is executable to cause the machine to perform any one or more of the methods described herein. In alternative embodiments, the machine may be connected (e.g., networked) to other machines in a Local Area Network (LAN), an internal network, an external network, or the Internet. The machine may operate as a server or client machine in a client-server network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), tablet computer, set-top box (STB), personal digital assistant (PDA), cellular phone, web device, server, network router, switch, or bridge, or any machine capable of executing a set of instructions (sequentially or otherwise) specifying the actions the machine will take. Furthermore, although only a single machine is shown, the term "machine" should also be understood to include any collection of machines (e.g., computers) that individually or jointly execute one (or more) sets of instructions to perform any one or more of the methods discussed herein. In an embodiment, computing device 600 may correspond to a controller for a tool cluster and / or load locking of a substrate processing system. For example, computing device 600 may correspond to... Figure 1A The controller 150.

[0126] Example computing device 600 includes processing device 602, main memory 604 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) (such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM)), static memory 606 (e.g., flash memory, static random access memory (SRAM), hard disk (magnetic storage device)), and auxiliary memory (e.g., data storage device 618), all of which communicate with each other via bus 630.

[0127] Processing device 602 represents one or more general-purpose processors, such as microprocessors, central processing units, or similar processors. More specifically, processing device 602 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing combinations of instruction sets. Processing device 602 may also be one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, or similar processors. Processing device 602 is configured to execute processing logic (instruction 622) for performing the operations and steps discussed herein.

[0128] The computing device 600 may further include a network interface device 608. The computing device 600 may also include a video display unit 610 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 612 (e.g., a keyboard), a cursor control device 614 (e.g., a mouse), and a signal generation device 616 (e.g., a speaker).

[0129] Data storage device 618 may include machine-readable storage medium (or more specifically, computer-readable storage medium) 628 on which one or more instruction sets 622 are stored, which embody any one or more of the methods or functions described herein. During execution of instructions by computer system 600, instructions 622 may also reside wholly or at least partially in main memory 604 and / or processing device 602, which also constitute computer-readable storage medium.

[0130] Computer-readable storage medium 628 may also be used to store autonomous load locking engine 152 and / or software programs containing methods for calling autonomous load locking engine 152. Although computer-readable storage medium 628 is shown as a single medium in the example embodiment, the term "computer-readable storage medium" should be understood to include a single medium or multiple media (e.g., a centralized or distributed database, and / or associated cache memory and server) storing one or more instruction sets. The term "computer-readable storage medium" should also be understood to include any medium capable of storing or encoding a set of instructions executable by a machine and causing the machine to perform any one or more of the methods described herein. Therefore, the term "computer-readable storage medium" should include, but is not limited to, non-transitory computer-readable media such as solid-state memory, optical and magnetic media.

[0131] The modules, components, and other features described herein (e.g., for...) Figures 1A to 2 These modules can be implemented as discrete hardware components or integrated into the functionality of hardware components such as ASICs, FPGAs, DSPs, or similar devices. Furthermore, modules can be implemented as firmware or functional circuitry systems within a hardware device. Additionally, such modules can be implemented using any combination of hardware devices and software components, or solely using software.

[0132] Some parts of the detailed description of algorithms and symbolic representations for manipulating data bits within computer memory have been presented. These algorithmic descriptions and representations are a means by which those skilled in the art of data processing most effectively communicate the essence of their work to others skilled in the art. Algorithms are generally considered herein as a self-consistent sequence of steps leading to a target result. These steps are those requiring physical manipulation of physical quantities. Typically, although not always necessary, these physical quantities take the form of electrical or magnetic signals that can be stored, transmitted, combined, compared, and otherwise manipulated. Primarily for common purposes, it has proven convenient to sometimes refer to these signals as bits, values, elements, symbols, characters, terms, numbers, etc.

[0133] However, it should be remembered that all such and similar terms are associated with appropriate physical quantities and are merely convenient labels applied to those quantities. Unless otherwise stated, it will be apparent from the following discussion that throughout the description, the use of terms such as “receive,” “identify,” “determine,” “select,” “provide,” “store,” or the like refers to the actions and processes of a computer system or similar electronic computing device that manipulates and converts data representing physical (electronic) quantities in the computer system’s caches and memory into other data representing physical quantities similarly represented in the computer system’s memory or caches, or other such information storage, transmission, or display devices.

[0134] Embodiments of this disclosure also relate to an apparatus for performing the operations described herein. This apparatus may be specifically constructed for the purposes discussed, or it may comprise a general-purpose computer system selectively programmed by a computer program stored in a computer system. Such a computer program may be stored in a computer-readable storage medium, such as, but not limited to, any type of disk, including floppy disks, optical disks, CD-ROMs and magneto-optical disks, read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic disk storage media, optical storage media, flash memory devices, other types of machine-accessible storage media, or any type of medium suitable for storing electronic instructions, each medium being coupled to a computer system bus.

[0135] The foregoing description sets forth numerous specific details, such as examples of specific systems, components, and methods, to facilitate a better understanding of the various embodiments of this disclosure. However, it will be apparent to those skilled in the art that at least some embodiments of this disclosure can be implemented without such specific details. In other instances, well-known components or methods have not been described in detail or presented in a simple block diagram format to avoid unnecessarily obscuring this disclosure. Therefore, the specific details set forth are merely exemplary. Specific embodiments may differ from these exemplary details and are still considered to be within the scope of this disclosure.

[0136] Throughout this specification, the phrase "one embodiment" or "an embodiment" means that at least one embodiment includes a particular feature, structure, or characteristic described for that embodiment. Therefore, the appearance of the phrase "in one embodiment" or "in an embodiment" throughout this specification does not necessarily refer to the same embodiment. Furthermore, the term "or" indicates inclusive "or," not exclusive "or." When the terms "about" or "approximately" are used herein, this is intended to mean that the given nominal values ​​are accurate to within ±10%.

[0137] Although the operations of the methods described herein are shown and described in a specific order, the order of operations for each method can be changed so that some operations can be performed in reverse order, thereby allowing some operations to be performed at least partially concurrently with other operations. In another embodiment, instructions or sub-operations of different operations can be performed intermittently and / or alternately.

[0138] It should be understood that the above description is intended to illustrate, not limit. Many other embodiments will be apparent to those skilled in the art upon reading and understanding the above description. Therefore, the scope of this disclosure should be determined by referring to the full scope of the appended claims and their equivalents.

Claims

1. A load lock, the load lock comprising: A substrate support member configured to hold a substrate; At least one sensor, said at least one sensor being used to generate a measurement reflecting the condition within the load lock; A computing subsystem, configured to: The measured values ​​are processed to determine the condition within the load lock; and Determine whether to perform a repair action based on the state of the load lock; and A complex atom system configured to perform the repair action in response to a determination to perform the repair action.

2. The load lock as claimed in claim 1, wherein the condition includes at least one of the following: the contamination state of the load lock, the environmental state of the load lock, the component state of the load lock, the presence of the substrate on the substrate support, or the condition of the substrate.

3. The load lockout as claimed in claim 1, wherein the computing subsystem further determines whether the condition within the load lockout meets a threshold value, wherein the computing subsystem determines to perform the repair action in response to the determination that the condition meets the threshold value.

4. The load locking of claim 1, wherein the computational subsystem uses a trained machine learning model to process the measurement, wherein the trained machine learning model a) outputs a suggestion to perform the repair action, or b) initiates the execution of the repair action.

5. The load lockout as claimed in claim 1, wherein the condition associated with the measured value includes at least one of the amount of foreign particles, the amount of moisture, or the mass of organic matter.

6. The load lock as claimed in claim 1, wherein the at least one sensor includes a temperature sensor, and the measured value includes a measured value of the temperature within the load lock.

7. The load lock as claimed in claim 1, wherein the at least one sensor includes at least one of a vibration sensor or an accelerometer, and wherein the condition is associated with at least one of the load lock, a structural failure, surface damage, or vibration of at least one of the substrate or a component of the load lock.

8. The load lockout of claim 1, wherein the condition is associated with at least one of the surface charge, warpage, back-side cleanliness, degassing, or static charge of the substrate.

9. The load locking as claimed in claim 1, wherein the at least one sensor is selected from the group consisting of: microelectromechanical systems (MEMS) sensors, light scattering sensors, impact sensors, aerosol electrometers, relative humidity (RH) sensors, mass spectrometers, residual gas analyzer (RGA) sensors, pressure sensors, solid distortion sensors, or surface acoustic wave (SAW) sensors.

10. The load locking as claimed in claim 1, wherein the complex atom system further comprises at least one of a purification and cleaning subsystem, a thermal cycling subsystem, an electrostatic trap subsystem, a cryogenic aerosol subsystem, an ultraviolet (UV) removal subsystem, or a cold trap.

11. The load lock as claimed in claim 1, wherein the repair action includes scheduled maintenance of the load lock.

12. The load lock as claimed in claim 1, wherein the repair action includes an automatic cleaning cycle for the load lock, wherein the automatic cleaning cycle: Reduce the pressure in the load-locked chamber to between 1 mTorr and 10 mTorr; and This results in an initial flow rate of more than 100 sccm / liter for the gas entering the load lock through the inlet line.

13. The load lock as claimed in claim 1, wherein the repair action includes an automatic cleaning cycle of the load lock, wherein the automatic cleaning cycle includes introducing gas into the load lock through an inlet line at an initial velocity of greater than 2 m / s.

14. The load lock as claimed in claim 1, wherein the computing subsystem processes the measured values ​​to determine the condition within the load lock, and based on the condition of the load lock, determines whether to perform a repair action after the production board has been run, or periodically.

15. The load locking of claim 1, wherein the complex atom system further comprises a trap system configured to passively trap particles using thermal changes or electrostatic mechanisms.

16. The load lock of claim 15, wherein the trap subsystem includes an electrostatic trap configured to capture charged particles, wherein the electrostatic trap includes an anode and a cathode embedded in the sidewall of the load lock.

17. A load lock, the load lock comprising: An inlet gas line, the inlet gas line including a first valve; Outlet gas pipeline; A vacuum pump, coupled to the outlet gas line; and A controller operably coupled to the first valve and the vacuum pump, wherein the controller performs the load-locked automatic cleaning cycle, and wherein, in order to perform the automatic cleaning cycle, the controller will: Set a target pressure between 1 mTorr and 10 Torr for the vacuum pump; and The first valve is actuated to make the initial flow rate of the gas entering the load lock through the inlet gas line greater than 100 sccm / liter.

18. The load locking as claimed in claim 17, wherein the load locking further comprises: At least one sensor is used to generate a measurement reflecting the condition within the load lock.

19. The load lockout of claim 18, wherein the controller responds to the measurement value of the at least one sensor within the load lockout to determine the execution of the automatic cleaning cycle, wherein the measured condition includes the contamination state within the load lockout.

20. The load lock as claimed in claim 17, wherein the automatic cleaning cycle includes introducing gas into the load lock through the inlet gas line at an initial velocity of greater than 2 m / s.

21. The load lockout of claim 17, wherein the vacuum pump is configured to return the load lockout to a base pressure below 1 Torr.

22. A method for performing a repair action in a load lockout, the method comprising: The system generates measurements of the conditions within the load lockout using at least one sensor. The measured values ​​are processed by a computational subsystem to determine the condition within the load lock. Based on the load lock status, the computing subsystem determines whether to perform a repair action. and In response to the determination to perform the repair action, the repair action is performed through the complex atom system.