Carrying panel substrate

Abstract

An apparatus and method for processing a substrate are provided herein. In some embodiments disclosed herein, a processing system includes: a processing chamber including one or more of: a rotational base to receive a panel substrate, wherein the rotational base is capable of adjusting an orientation of the panel substrate; and a set of sensors; a transport mechanism capable of transporting the panel substrate into or out of the processing chamber or adjusting a translational position of the panel substrate; and a computing device capable of receiving sensor data from the set of sensors and calculating, based on the sensor data: a rotational correction to be performed by the rotational base to adjust the orientation of the panel substrate; or a translational correction to be performed by the transport mechanism to adjust the translational position of the panel substrate.

Description

Background Technology field

[0001] The embodiments of the present invention generally relate to an apparatus and method for handling substrates in an electronic device manufacturing system, and more particularly, to an apparatus and method for transporting panel substrates. Related technical descriptions

[0002] In electronic device manufacturing, substrate handling often involves performing deposition processes on one side of a substrate. However, processing chambers are typically designed to handle wafers (e.g., disk-shaped substrates). When processing substantially rectangular substrates (such as panels), specific challenges arise. For example, unlike wafers, panel orientation is critical during processing (e.g., to ensure uniform deposition on the panel). Even a slight deviation from the desired orientation can negatively impact the deposition results, leading to a significant decrease in panel yield.

[0003] Attempts have been made to alter the position and orientation of the panel substrate within the processing chamber. However, such methods often rely on a single transport mechanism (e.g., a robotic arm) to perform multiple iterative corrections of the panel position and orientation, which can be very time-consuming. Furthermore, these methods may not depend on sensors specifically designed to detect the panel's position and orientation, leading to inefficiencies during processing.

[0004] Accordingly, there is a need in the art for an apparatus and method for handling panel substrates. Summary of the Invention

[0005] The embodiments described herein generally relate to physical vapor deposition (PVD) of substrates in electronic device manufacturing processes. More specifically, the embodiments described herein provide an apparatus and method for handling panel substrates.

[0006] In one embodiment, a processing system includes: a processing chamber comprising one or more of the following: a rotating base for receiving a panel substrate, wherein the rotating base is capable of adjusting the orientation of the panel substrate; a set of sensors; a transport mechanism capable of transporting the panel substrate into or out of the processing chamber or adjusting the translational position of the panel substrate; and a computing device capable of receiving sensor data from the set of sensors and calculating, based on the sensor data: that rotational correction will be performed by the rotating base to adjust the orientation of the panel substrate, or that translational correction will be performed by the transport mechanism to adjust the translational position of the panel substrate.

[0007] In another embodiment, a method includes: receiving sensor data indicating a panel substrate from a set of sensors; calculating the position and orientation of the panel substrate based on the sensor data; calculating a rotation correction or a translation correction to be performed by a rotating base to adjust the orientation of the panel substrate, the translation correction to be performed by a transport mechanism to adjust the translational position of the panel substrate; and commanding the rotating base to perform the rotation correction or the transport mechanism to perform the translation correction. Attached Figure Description

[0008] To gain a more detailed understanding of the features set forth above in this disclosure, a more specific description of the disclosure, which has been briefly outlined above, can be obtained with reference to the embodiments illustrated in the accompanying drawings. However, it should be noted that the drawings illustrate exemplary embodiments only and should therefore not be construed as limiting the scope of this disclosure, as other equally effective embodiments are acknowledged.

[0009] Figure 1A An example substrate handling system for transporting a panel substrate is described according to certain embodiments.

[0010] Figure 1B According to certain embodiments Figure 1A A schematic isometric top view of a portion of a substrate processing system, which includes a flipper module.

[0011] Figure 1C According to certain embodiments Figure 1A A schematic isometric top view of a portion of a substrate processing system, which includes a flipper module.

[0012] Figure 1D An example of a transfer section of a substrate handling system for transporting panel substrates according to certain embodiments is depicted.

[0013] Figures 2A to 2B An example sensor array for transporting a panel substrate is depicted according to certain embodiments.

[0014] Figures 3A to 3B An example scenario for a correction calculation for transporting a panel substrate, according to certain embodiments, is described.

[0015] Figure 4 This is a flowchart of an example operation for handling a panel substrate according to certain embodiments.

[0016] To facilitate understanding, the same element symbols have been used to denote common elements across the figures, where possible. It is contemplated that elements and features of one embodiment may be advantageously incorporated into other embodiments without further elaboration. Detailed Implementation

[0017] The embodiments described herein generally relate to an electronic device manufacturing system configured to perform one or more processes on a panel substrate. Some embodiments disclosed herein relate to physical vapor deposition (PVD) processes performed on a panel substrate within the electronic device manufacturing system. More specifically, the embodiments described herein provide an apparatus and method for handling a panel substrate.

[0018] Compared to conventional methods that use only one mechanism (such as a robotic arm) to adjust the position and orientation of a panel substrate, the embodiments described herein enable the panel substrate to be moved to a target position and orientation using a transport mechanism and a rotating base separately. Using multiple separate mechanisms to correct the position and orientation of the panel substrate allows for rapid and accurate correction, thereby increasing output.

[0019] The embodiments described herein incorporate multiple sensors specifically designed to detect the position of a panel substrate via its corners. Identifying the corners of the panel substrate allows for rapid identification of the panel substrate's position and orientation, which in turn allows for accurate calculation of the translational and / or rotational corrections required to move the panel substrate to a target position and orientation to ensure successful processing. Example substrate processing system

[0020] Figure 1A An example substrate processing system 100 according to certain embodiments is illustrated. Figures 1B to 1D Examples of portions of a substrate processing system 100 according to certain embodiments are illustrated. In some embodiments, the substrate processing system 100 is specifically configured to process substantially rectangular or square substrates, such as panel 110. As used herein, the term "panel" generally refers to a substantially rectangular substrate, which may be largely rectangular or square and includes small cutouts on one or more sides. Panels can be used for device packaging or the formation of large-panel displays. In some device packaging examples, the panel may include a large surface area substrate comprising a polymer material disposed above a structural core.

[0021] The substrate processing system 100 may include a transport mechanism 120 (or a transfer robot 120), a substrate support base 122 (also referred to as a rotating base 122 or base 122), and a sensor array 124 disposed above the base 122. In some examples, the transport mechanism 120 is located in a transfer chamber 106, while the base 122 and the sensor array 124 are located in a processing chamber 130. Figures 1B to 1C In this configuration, processing chamber 130 is coupled to transfer chamber 106. Sensor array 124 will include one or more sensors, such as... Figure 1BSensors 124A and 124B are shown. Processing chamber 130 can be positioned in one or more of processing chamber locations 108, 110, 112, 114, 116, and 118 in the substrate processing system 100, such as... Figure 1A As shown in the diagram. Generally, the processing chamber 130 may include an atomic layer deposition (ALD) chamber, a chemical vapor deposition (CVD) chamber, a physical vapor deposition (PVD) chamber, an etching chamber, a degassing chamber, an ion implantation chamber, an ashing chamber, a cleaning chamber, a thermal processing chamber (e.g., rapid thermal processing, annealing, or cooling), or other types of substrate processing chambers. In some embodiments, the processing chamber 130 includes a PVD deposition chamber adapted to deposit a film on one or more sides 111 of the panel 110 via a "sputtering" process.

[0022] Figure 1A This is a schematic top view of an exemplary substrate processing system 100 (also referred to as a "processing platform") according to certain embodiments. In some embodiments, the substrate processing system 100 is specifically configured to process large-area substrates. As used herein, the term "panel" may refer to a large-area substrate that can be used for device packaging or the formation of a large-panel display. In some device packaging examples, a "panel" may include a large surface area substrate comprising a polymer material disposed above a structural core. The substrate processing system 100 generally includes: a device front-end module (EFEM) 102 for loading substrates into the processing system 100; a first loading lock chamber 104 coupled to the EFEM 102; a transfer chamber 106 coupled to the first loading lock chamber 104; and a plurality of other chambers coupled to the transfer chamber 106, as detailed below. The transfer chamber 106 may include a transport mechanism, such as... Figures 1A to 1D The transport unit 120 shown.

[0023] Moving counterclockwise from the first loading lock chamber 104 around the transfer chamber 106, the processing system 100 includes a first dedicated degassing chamber 108, a first pre-cleaning chamber 110, a first deposition chamber 112, a second pre-cleaning chamber 114, a second deposition chamber 116, a second dedicated degassing chamber 118, and a second loading lock chamber 105. Chambers 110, 112, 114, and 116 may each include a rotating base and a sensor array, such as... Figures 1A to 1D The base 122 and sensor array 124 are shown.

[0024] In some embodiments, the transfer chamber 106 and the chambers coupled to the transfer chamber 106 are under vacuum. As used herein, the term "vacuum" may refer to a pressure below about 10... -2 Pa. However, some high vacuum systems can operate at 10 Pa. -7The operation is below Pa. In some embodiments, the vacuum is generated using a turbomolecular pump coupled to the loading lock chambers 104, 105. However, other types of vacuum pumps are also conceivable.

[0025] In some embodiments, system controller 126 (also referred to herein as a “processing chamber controller”) includes a central processing unit (CPU) 127, memory 128, and support circuitry 129. System controller 126 controls the sequence of processes during the processing of panel 110, including substrate transfer methods and substrate position and orientation correction methods as described above. CPU 127 is a general-purpose computer processor configured for use in an industrial environment to control processing system 100 and its associated subprocessors. Memory 128, as described herein, is typically non-volatile memory and may include random access memory, read-only memory, floppy disk drives or hard disk drives, or other suitable forms of digital memory, local or remote. Support circuitry 129 is conventionally coupled to CPU 127 and may include caches, frequency circuitry, input / output subsystems, power supplies, and the like and combinations thereof. Software instructions (i.e., software programs) and data may be decoded and stored in memory 128 for instructing the processor within CPU 127. A software program (which can be read by the CPU 127 in the system controller 126) determines which task can be performed by a component in the processing system 100. Typically, the software program, which can be read by the CPU 127 in the system controller 126, includes program code that, when executed by the processor (CPU 127), performs tasks related to the processing and substrate transport scheme described above. The software program may include instructions for controlling various hardware and electrical components within the processing system 100 such that the methods described herein can be performed. In some embodiments, the program includes instructions for performing one or more of the methods for correcting the position and orientation of the substrate as described above. For example, the CPU 127, memory 128, and support circuitry 129 may be communicatively coupled to a transport mechanism, a rotating base, and a sensor array to correct the position and orientation of the substrate in the chamber, as described above.

[0026] In some embodiments, multiple substrates are loaded into the processing system 100 through a door (also referred to as a "slit valve") in a first loading lock chamber 104 and unloaded from the processing system 100 through a door in a second loading lock chamber 105. In some embodiments, the stack of multiple substrates is supported in a cassette placed in the first loading lock chamber 104. Once the first loading lock chamber 104 is evacuated, a robot located in a transfer chamber 106 retrieves one substrate at a time from the cassette. In some embodiments, after processing has been performed on each side of a single substrate, the second loading lock chamber 105 receives the single substrate and unloads the processed substrate to EFEM 102. However, other loading and unloading configurations are also contemplated.

[0027] Pre-cleaning of the substrate is important for removing impurities (such as oxides) from the substrate surface, ensuring that the metal film deposited in the deposition chamber is not electrically insulating from the substrate. By performing pre-cleaning in a first pre-cleaning chamber 110 and a second pre-cleaning chamber 114 that share the same vacuum environment as the first deposition chamber 112 and the second deposition chamber 116, the substrate can be transferred from the cleaning chamber to the deposition chamber without exposure to the atmosphere. This prevents impurities from forming on the substrate during transfer. Furthermore, since a vacuum is maintained in the substrate processing system 100 during the transfer of the cleaned substrate to the deposition chamber, vacuum pumping cycles are reduced.

[0028] In some embodiments, only one substrate is processed at a time within each pre-cleaning and deposition chamber. Alternatively, multiple substrates, such as four to six substrates, can be processed at once. In such embodiments, the substrates may be disposed on rotatable bases within individual chambers. In some embodiments, cleaning the first pre-cleaning chamber 110 and the second pre-cleaning chamber 114 is a pre-cleaning etching chamber for etching the substrate surface. However, other types of pre-cleaning chambers are also contemplated. In some embodiments, one or both of the pre-cleaning chambers are replaced by deposition chambers for performing reactive sputtering processes, such as depositing silicon nitride, aluminum oxide, or other materials by reactive sputtering. In an inductively coupled plasma (ICP) chamber, a coil located in the upper portion of the chamber is energized with an external RF source to generate an excitation field within the chamber. Argon gas flows through the chamber from an external gas source. Argon atoms in the chamber are ionized (charged) by the RF energy. The substrate is biased by a DC bias source coupled to an aluminum base on which the substrate rests. Charged atoms are attracted to the substrate, resulting in etching of the substrate surface. Besides argon, other gases may be used, depending on the desired etching rate and the material to be etched. The ion energy level used for etching as part of a cleaning process can be relatively low compared to processes used to etch features onto the substrate surface. Lower energies avoid damaging existing circuitry and features on the substrate.

[0029] In some embodiments, the first deposition chamber 112 and the second deposition chamber 116 are PVD chambers. In such embodiments, the PVD chambers may be configured to deposit copper, titanium, aluminum, gold, nickel, nickel-vanadium, silver, and / or tantalum. However, other types of deposition processes and materials are also contemplated. In the PVD chamber, the entire back side of the substrate is in electrical and thermal contact with the substrate. Temperature control of the substrate is important during the sputtering process to obtain predictable and reliable films. The coolant system includes an external cooling source that supplies fluid to cooling lines in the substrate. The cooling source may be replaced or enhanced with a heating source to increase the workpiece temperature outside the sputtering process.

[0030] In some embodiments, an RF bias source is electrically coupled to the substrate to enable the substrate and thus the substrate during the sputtering process. For example, substrate bias (RF bias) may be used when the substrate / panel has features requiring good step coverage. Alternatively, the substrate may be grounded, floating, or biased only by a DC voltage source.

[0031] In operation, the PVD deposition chamber is emptied and backfilled with argon. The gas is energized using a DC source to couple with the electromagnetic field inside the PVD deposition chamber, thereby generating a sustained high-density plasma near the target surface. The plasma, confined near the target surface, contains positive ions (such as Ar+) and free electrons. Ions in the plasma bombard the target surface, sputtering material away from the target. The substrate receives the sputtered material to form a deposited layer on its surface. In one example, up to 20 kW of DC power can be provided on the target, enabling the target to deposit material at a rate of approximately 1 micrometer per minute on the substrate.

[0032] Outside the vacuum, the PVD deposition chamber uses a magnetron sputtering assembly to further control the bombardment of the target by ions formed in the plasma. In some embodiments, a fixed permanent magnet is positioned behind the target (as a deposition source), confining the plasma to the target area. In others, the magnet sweeps across the entire back side of the target to help distribute the magnetic field evenly across the entire target for more uniform target erosion. The resulting magnetic field forms a looping path that acts as an electron trap, reshaping the orbits of secondary electrons emitted from the target into cycloidal paths, thereby significantly increasing the ionization probability of the sputtering gas within the confined area. Inert gases (such as argon) are typically used as sputtering gases because they tend not to react with the target material or combine with any process gases, and their high molecular weight results in high sputtering and deposition rates. Positively charged argon ions from the plasma are accelerated toward the negatively biased target and bombard it, causing the material to be sputtered from the target surface.

[0033] During the processing operation, the walls of the chamber are typically electrically grounded. A bias voltage on the substrate drives a flux of charged species (Ar+ and / or atoms sputtered from the target) to the substrate. This flux can modify the properties of the sputtered material deposited on the substrate surface, such as film density.

[0034] Figure 1B According to certain embodiments Figure 1A A schematic isometric top view of a portion of the illustrated substrate processing system 100. As shown, a processing chamber 130 is configured to receive a panel 110 for processing therein. The processing chamber 130 includes a top chamber 125 and a bottom chamber 131, which are configured to receive the panel 110 and form a processing environment therein. In one example, the top chamber 125 includes a target (not shown) for sputtering a deposited layer on a first side 111 of the panel 110 using a PVD deposition process, and the panel is supported on a base 122.

[0035] like Figure 1B As illustrated, panel 110 is disposed in transfer chamber 106. The edge of panel 110 contacts end effector 120A, which is coupled to the robotic arm 120B of transfer robot 120. Panel 110 and the end effector of transfer robot 120 are aligned with the door of processing chamber 130. According to some embodiments, the first side 111 of panel 110 faces upward. Panel 110 may have a thickness ranging from about 0.1 mm to about 4 mm, for example, from about 0.2 mm to about 3.2 mm. In this example, panel 110 is a panel (also referred to herein as a “substrate”).

[0036] Figure 1B Further illustration shows how the loading lock chamber 130 can also include multiple sensors 124A, 124B. Each of the multiple sensors 124A, 124B can be an optical sensor, a proximity sensor, or any combination thereof, for analyzing the characteristics of the panel 110, as will be discussed further below.

[0037] Figure 1C This illustration shows the situation where panel 110 is transferred from transfer chamber 106 to processing chamber 130 by transfer robot 120. Note that, for clarity, Figure 1C Only the lower portion 131 of the transfer chamber 106 and the processing chamber 130 is shown. (See image below.) Figure 1C As shown, the transfer robot 120 has moved the panel 110 from the transfer chamber 106 into the bottom chamber 131 of the processing chamber 130, so that the panel 110 can then be placed on the base 122 (located below the panel 110), as... Figure 1CAs shown in the diagram. Once the substrate is in the processing chamber 130, the transfer robot can release the panel 110, allowing the processing chamber 130 to provide support to the panel 110.

[0038] The transport mechanism 120 can transport the panel 110 to or away from the base 122. Figure 1D In this example, such as Figure 1D As illustrated, the conveying mechanism 120 includes a robotic arm, but may also include other systems such as a conveyor belt system. The panel 110 can be conveyed via the conveying mechanism 120 to the base 122 for processing. When the panel 110 is to be placed on or received from the base 122, the base 122 can be rotated clockwise or counterclockwise using a rotary actuator, wherein the rotary actuator is coupled to the base to rotate the panel 110 and / or change the orientation of the panel 110 relative to the base 122 during processing.

[0039] In some examples, the transport mechanism 120, base 122, and associated support components (such as actuators, sensors, and other related devices) are communicatively coupled to a computing device (such as a system controller 126). In some embodiments, the computing device includes one or more processors, one or more storage devices, one or more communication interfaces, and / or other computing components. The computing device can perform calculations and command the transport mechanism 120 and base 122 to perform actions to adjust the position and orientation of the panel 110. In some examples, the computing device is integrated within the transport mechanism 120. Details regarding how the computing device performs calculations are provided below.

[0040] After the panel 110 is placed on the base 122, one or more sensors (such as...) within the sensor array 124 can be used. Figure 1B The illustrated sensors 124A and 124B are used to detect panel 110. One or more sensors in sensor array 124 may include one or more pairs of light emitters (e.g., laser emitters) and light receivers (e.g., laser receivers). In some examples, additionally or alternatively, sensor array 124 includes one or more cameras (e.g., pinhole cameras, stereo cameras, or thermal imaging cameras) positioned above panel 110 to observe at least a portion of the panel, including at least one edge. The term "camera" is intended to broadly describe a device capable of generating a visual image stream based at least on the field of view (FOV) of one or more optical components (e.g., lenses) and an image sensor (e.g., CCD, CMOS sensor, etc.). Sensor array 124 may be communicatively coupled to the computing device described above. Details regarding the arrangement of sensor array 124 can be found in [reference needed]. Figures 2A to 2B See below.

[0041] Sensor data received from one or more sensors 124A, 124B within sensor array 124 can be provided to a computing device to calculate the position and orientation of panel 110 on base 122. The computing device can compare the position and orientation of panel 110 with a target position and target orientation of panel 110 to calculate a set of corrections required to reposition panel 110 to the target position and to reorient panel 110 to the target orientation. The target positions(s) can be stored in the memory of the computing device, and the target positions are derived from previous testing or calibration procedures performed within substrate processing system 100. The set of corrections may include rotational corrections to adjust the orientation of panel 110, or translational corrections to adjust the translational position of panel 110. Rotational corrections can be performed using a rotary actuator (coupled to base 122), while translational corrections can be performed by controlling the movement of transport mechanism 120 relative to base 122.

[0042] In some examples, a computing device (e.g., system controller 126) may use a first machine learning model to help calculate the position or orientation of panel 110. For example, the first machine learning model may be used to calculate the coordinates of the center of panel 110 and some (e.g., at least two) corners. The coordinates of the center of panel 110 may correspond to the position of panel 110, while the coordinates of the center and corners of panel 110 may be used to calculate the orientation of panel 110. The first machine learning model may include a convolutional neural network (CNN), a recurrent neural network (RNN), or a residual neural network (ResNet).

[0043] Accordingly, after calculating the position and orientation of panel 110, the computing device can calculate the deviation (e.g., as a vector) and angular deviation (e.g., as an angle) from that position to the target position, to reflect the corrections to be performed by transport mechanism 120 and base 122, respectively. For example, the deviation from the position to the target position can be decomposed into translational correction and radial correction, while the angular deviation can correspond to rotational correction. Details regarding the decomposition of the corrections can be found in [reference needed]. Figure 3A See below.

[0044] Additionally, in some examples, the computing device may use a second machine learning model (e.g., a reinforcement learning model) to help command the transport mechanism 120 to perform translational corrections or command the base 122 to perform rotational corrections. For example, the second machine learning model may include a deep Q-network (DQN) or a proximal policy optimization (PPO).

[0045] For details regarding this set of corrections, please refer to [link / reference]. Figure 3A See below, and refer to how the computing device calculates the position of panel 110. Figure 3B See below. Example environment of sensor array

[0046] Figure 2A A schematic side view of environment 200 is shown, which is a front side view of a portion of processing chamber 130. Figure 2B This is an isometric top view of environment 200. Figure 2A and Figure 2B Both illustrations depict an environment 200 as viewed from a perspective, where a portion of the processing chamber 130 is viewed from within the transport chamber 106. As shown, the environment 200 includes a chamber 210, a robotic arm 120B, a panel 110, two sensors 124A and 124B (which respectively include light emitters 240a-b and light receivers 242a-b), and viewports 244a-d. The robotic arm 120B may be part of the transport mechanism 120, the panel 110 may be part of the panel 110, and the light emitters 240a-b, light receivers 242a-b, and viewports 244a-d may be part of the sensor array 124, such as... Figure 1D exemplified.

[0047] like Figure 2A As shown, light emitters 240a-b are located in the upper portion of cavity 210 and above panel 110, while light receivers 242a-b are located in the lower portion of cavity 210 and below panel 110. In this example, as illustrated, light emitters 240a-b and light receivers 242a-b are located outside cavity 210. Viewports 244a-d allow light emitted by light emitters 240a-b to pass through and exit cavity 210, so that the light can be received by light receivers 242a-b.

[0048] In this example, as illustrated, light emitted by light emitter 240a enters chamber 210 through viewport 244a, strikes panel 110 (e.g., partially near a corner), and travels from chamber 210 to receiver 242a through viewport 244b. Similarly, light emitted by light emitter 240b enters chamber 210 through viewport 244c, strikes panel 110 (e.g., partially near another corner), and travels from chamber 210 to receiver 242b through viewport 244d.

[0049] Alternatively, in some examples, the light emitters 240a-b and the light receivers 242a-b are located inside the chamber 210 (e.g., in a protective housing), and the viewports 244a-d may also be located in the chamber 210 (e.g., on the protective housing).

[0050] In this example, as illustrated, each light emitter is vertically positioned above its corresponding light receiver. For example, light emitter 240a is vertically positioned above light receiver 242a, while light emitter 240b is vertically positioned above light receiver 242b. Light emitters 240a-b emit curtain-like beams having an elongated shape focused towards a square region indicating and / or used to detect desired positions at the corners of the substrate, and the light receiver detects multiple segments of the emitted diagonal curtain-like beams. Details regarding the shape of the emitted and received beams can be found in [reference needed]. Figure 3B See below.

[0051] like Figure 2B As shown, light emitters 240a-b and light receivers 242a-b can be positioned to cover two corners on the leading edge (i.e., the front edge) of panel 110. Placing sensors (such as light emitters 240a-b and light receivers 242a-b) away from the robotic arm 120B will allow the robotic arm 120B and other transport mechanisms more maneuverability when transferring and correcting the position of panel 110. Alternatively, in another example, the light emitters and light receivers can cover a first corner of panel 110, while another pair can cover a second corner diagonally opposite the first corner.

[0052] Although shown as including two pairs of sensors, the environment 200 may include more pairs of light emitters and light receivers. For example, four pairs of light emitters and light receivers may be used, with each pair covering a separate corner of the square panel 110. This can be useful when the edges of the panel 110 have different cutouts to indicate a specific target orientation of the panel 110.

[0053] In some examples, alternatively or additionally, environment 200 includes other types of sensors, such as cameras, as reference sensors. Figures 1A to 1D As stated above. Example calculations for correcting panel position and orientation

[0054] Figure 3A Example breakdown 300 of the correction is shown. This set of corrections may include: various component translation corrections, such as radial correction 310a (e.g., extension of the robotic arm 120B of the conveying mechanism 120) and θ correction 310b (e.g., rotation of the conveying mechanism 120 about a vertical central axis); and rotation corrections (e.g., rotation of the base 122), such as rotation correction 310c.

[0055] Translational correction can be decomposed into radial correction 310a and θ correction 310b. Radial correction 310a and θ correction 310b can be achieved via a transport mechanism (such as...) Figures 1A to 1DThe illustrated transport mechanism 120 performs the corrections. In this example, radial correction 310a and θ correction 310b are generated based on translation corrections using polar or spherical coordinates, where the reference point is the central axis of rotation of the transport mechanism 120 (e.g., along the z-axis as illustrated).

[0056] As illustrated, during operation of the transport mechanism, radial correction 310a varies as the robotic arm 120B changes the extension of the end effector 120A, and θ correction 310b varies in the direction of angular operation of the transport mechanism about its central axis of rotation (e.g., when the transport mechanism rotates clockwise or counterclockwise).

[0057] Similarly, rotational correction 310c can be performed by rotating the base to correct the orientation of the panel, such as... Figure 1D The base 122 is illustrated. The rotation correction 310c can indicate angular changes, such as indicating a positive angle of counterclockwise rotation or a negative angle of clockwise rotation.

[0058] Figure 3B This is a bird's-eye view of environment 200, illustrating example scenario 320 used for calibration calculations. As illustrated, scenario 320 uses two pairs of light emitters and light receivers configured to detect two corners on the same leading edge of panel 110. A curtain beam strikes panel 110 with two diagonal segments (i.e., segments 340a-b). In some examples, each of segments 340a-b has a length of approximately 10 mm. As illustrated, segments 340a-b are the diagonals of two square regions that partially cover the two corners of panel 110, respectively. As illustrated, each of the square regions is a sensing area, which can be observed by a set of sensors (such as sensor 124).

[0059] In some embodiments, the curtain bundle includes a sensing region that fills a square region (parallel to the main surface of panel 110) and is therefore configured to detect overlap between the square region and the corners of panel 110, such as Figure 3B As illustrated. In some other embodiments, the curtain bundle includes a sensing region that extends across the diagonal of the square region and is thus configured to detect the amount of length of the corner overlap of the curtain bundle (which extends across the diagonal of the square region) of the panel 110.

[0060] When the emitted light from the light emitter 240a strikes a portion of the panel 110, line segment 340a is partially blocked by the panel 110, forming points 350a-b, which become the intersection points between the edge of the panel 110 and the curtain bundle. Points 350a-b can be detected by the light receiver due to the detection of the transition between the light received from the light emitter and the light blocked by the edge of the panel 110. Similarly, line segment 340b intersects the panel 110 at points 350c-d. Points 350a-b and 350c-d can be compared with a set of reference points. The reference points represent the intersection points when the panel 110 is in the desired target position and has the desired target orientation. Accordingly, the directional distances (e.g., vectors) from each point in points 350a-d to its corresponding reference point can be used to calculate the translation and rotation corrections required to move the panel 110 to its target position and target orientation. Example operation of handling panel substrate

[0061] Figure 4 This is a flowchart of an example operation 400 for handling a panel substrate. Operation 400 can be executed by a computing device, such as as shown in the reference. Figures 1A to 1D The aforementioned computing device. In some embodiments, operation 400 is performed under vacuum.

[0062] In operation 410, operation 400 begins by inserting the substrate into the processing chamber 130. During operation 410, the computing device receives sensor data from a set of sensors, indicating the initial position of the panel substrate above the base 122. In this example, the sensor data may be received from a sensor array 124, and the substrate may be a panel 110, such as... Figures 1A to 1D exemplified.

[0063] In some embodiments, although not intended to limit the scope of the invention as provided herein, the set of sensors includes: a light emitter located in the upper portion of the processing chamber and above the panel substrate; and a light receiver located in the lower portion of the processing chamber and below the panel substrate. For example, the light emitter may be one of light emitters 240a-b, and the light receiver may be one of light receivers 242a-b, as illustrated in FIG2.

[0064] In some embodiments, the light emitter is positioned vertically above the light receiver, wherein the light emitter emits a curtain-like beam pointing diagonally toward a square, the square indicating a desired position of a corner of the panel substrate, and wherein the light receiver detects multiple segments of the diagonal curtain-like beam. For example, these segments may be one of line segments 340a-b.

[0065] In some embodiments, the sensor set includes multiple cameras located in the upper portion of the processing chamber and above the panel substrate, as shown in reference. Figure 1D As stated above.

[0066] In 420, the position and orientation of the panel substrate are calculated based on sensor data established by the overlap between light emitted from the light emitter and the area of ​​the panel 110, or sensor data established by the panel 110 entering the field of view of the camera. For example, the position and orientation of the panel substrate can be as described in reference... Figure 1D The calculation is based on the above.

[0067] In step 430, based on the position and orientation of the panel substrate, calculations are performed: rotational correction, which will be executed by the rotating base 122 to adjust the orientation of the panel substrate, or translational correction, which will be executed by the transport mechanism to adjust the translational position of the panel substrate. For example, rotational correction and translational correction can be performed as described in reference... Figure 3B The calculation is based on the above.

[0068] In some embodiments, the computing device decomposes translation correction into radial correction and θ correction. For example, such decomposition may be as described with reference to Figure 3A The above is performed. In such embodiments, commanding the transport mechanism to perform translational correction includes commanding the transport mechanism to perform radial correction and commanding the transport mechanism to perform θ correction.

[0069] In some embodiments, the calculation of rotational or translational correction is performed based on a machine learning model. In such embodiments, the machine learning model includes one or more of the following: convolutional neural networks (CNNs), recurrent neural networks (RNNs), or residual neural networks (ResNets). For example, the machine learning model may learn from historical or simulated samples, where each sample indicates input data (e.g., sensor data indicating the location where the curtain bundle strikes the corner of the panel substrate) and labels (e.g., the desired output), which may indicate translational correction or orientation. Thus, the trained machine learning model can calculate translational correction or orientation to place the panel substrate in a target location and orientation based on the input data.

[0070] In 440, the rotating base is commanded to perform rotational correction, or the transport mechanism is commanded to perform translational correction. For example, the rotating base may be base 122, and the transport mechanism may be transport mechanism 120, such as... Figures 1A to 1D As illustrated, for example, the computing device can first command the transport mechanism to perform translational correction on the panel substrate, so that the center of mass of the panel substrate is aligned with the center of mass of the target location. Subsequently, the computing device can command the rotating base to perform rotational correction to correct the orientation of the panel substrate.

[0071] In some embodiments, the command delivery mechanism performs translational corrections using a reinforcement learning model. In such embodiments, the reinforcement learning model includes a deep Q-network (DQN) or a proximal policy optimization (PPO). For example, the reinforcement learning model may learn from a history or simulated sequence of events related to panel substrate movement, where each event includes a time series of the panel substrate's position and orientation. The trained reinforcement model can then compute the desired translational or rotational corrections at each time step to move the panel substrate to the target position and orientation.

[0072] Although the foregoing description is directed to embodiments of the present invention, other and further embodiments of the present invention may be devised without departing from the basic scope of the present invention, and the scope of the present invention is defined by the appended claims.

Claims

1. A semiconductor processing apparatus, comprising: Treatment chambers, including one or more of the following: A rotating base for receiving a panel substrate, wherein the rotating base is capable of adjusting the rotational orientation of the panel substrate disposed thereon about a first axis; as well as A set of sensors is configured to detect at least a portion of the panel substrate and at least one edge of the panel substrate when the panel substrate is disposed above the rotating base; The transport mechanism is capable of transporting the panel substrate in at least one direction perpendicular to the first axis; as well as The computing device is capable of receiving sensor data from the set of sensors and calculating, based on the sensor data: rotational correction, which will be performed by the rotating base to adjust the rotational orientation of the panel substrate, or translational correction, which will be performed by the transport mechanism to adjust the translational position of the panel substrate.

2. The semiconductor processing apparatus of claim 1, wherein the set of sensors comprises: A light emitter is located in the upper part of the processing chamber, above the panel substrate; And a light receiver, located in the lower part of the processing chamber, below the panel substrate.

3. The semiconductor processing apparatus of claim 2, wherein the light emitter is vertically positioned above the light receiver, wherein the light emitter emits a curtain beam diagonally directed to a square, the square indicating a desired position of a corner of the panel substrate, and wherein the light receiver detects a plurality of segments of the diagonal curtain beam.

4. The semiconductor processing apparatus of claim 1, wherein the set of sensors includes a plurality of cameras located in the upper portion of the processing chamber above the panel substrate.

5. The semiconductor processing apparatus of claim 1, wherein the calculation of the rotation correction or the translation correction is performed based on a machine learning model.

6. The semiconductor processing apparatus of claim 5, wherein the machine learning model includes one or more of the following: convolutional neural network (CNN), recurrent neural network (RNN), or residual neural network (ResNet).

7. The semiconductor processing apparatus of claim 1, wherein the transport mechanism uses a reinforcement learning model to perform the translation correction.

8. The semiconductor processing apparatus of claim 7, wherein the reinforcement learning model comprises a deep Q-network (DQN) or a proximal policy optimization (PPO).

9. The semiconductor processing apparatus of claim 1, wherein the computing device decomposes translation correction into radial correction and θ correction.

10. The semiconductor processing apparatus of claim 9, wherein the translation correction performed by the transport mechanism to adjust the translational position of the panel substrate includes the radial correction and the θ correction.

11. A method comprising: Receive sensor data from a set of sensors, indicating the sensor panel substrate; The position and orientation of the panel substrate are calculated based on the sensor data; Based on the position and orientation of the panel substrate, the following calculations are made: rotational correction, performed by the rotating base to adjust the orientation of the panel substrate, or translational correction, performed by the transport mechanism to adjust the translational position of the panel substrate; and The rotating base is commanded to perform the rotational correction, or the conveying mechanism is commanded to perform the translational correction.

12. The method of claim 11, wherein the set of sensors comprises: A light emitter is located in the upper part of the processing chamber, above the panel substrate; And a light receiver, located in the lower part of the processing chamber, below the panel substrate.

13. The method of claim 12, wherein the light emitter is vertically positioned above the light receiver, wherein the light emitter emits a curtain-like beam diagonally directed to a square, the square indicating a desired position of a corner of the panel substrate, and wherein the light receiver detects a plurality of segments of the diagonal curtain-like beam.

14. The method of claim 11, wherein the set of sensors includes a plurality of cameras disposed above the panel substrate.

15. The method of claim 11, wherein the calculation of the rotation correction or the translation correction is performed based on a machine learning model.

16. The method of claim 15, wherein the machine learning model comprises one or more of the following: convolutional neural network (CNN), recurrent neural network (RNN), or residual neural network (ResNet).

17. The method of claim 11, wherein the transport mechanism is commanded to perform the translation correction using a reinforcement learning model.

18. The method of claim 17, wherein the reinforcement learning model comprises a deep Q-network (DQN) or a proximal policy optimization (PPO).

19. The method of claim 11, wherein the computing device decomposes the translation correction into radial correction and θ correction.

20. The method of claim 19, wherein commanding the transport mechanism to perform the translation correction comprises: The command instructs the transport mechanism to perform the radial correction and the command instructs the transport mechanism to perform the θ correction.

Images