Substrate processing apparatus, substrate processing method, learning data generation method, learning method, learning device, learned model generation method, and learned model

Abstract

A substrate processing apparatus (100) includes a plurality of processing units (1A1, 1B1), a plurality of individual exhaust pipes (41A, 41B), a common exhaust pipe (81A), a plurality of first pressure adjusting mechanisms (73A1, 73B1), a second pressure adjusting mechanism (87A), and a control section (102). Gases discharged from the processing units (1A1, 1B1) flow into the individual exhaust pipes (41A, 41B). Gases discharged from the individual exhaust pipes (41A, 41B) flow into the common exhaust pipe (81A). The first pressure adjusting mechanisms (73A1, 73B1) adjust the pressures of the gases flowing from the processing units (1A1, 1B1) into the individual exhaust pipes (41A, 41B). The second pressure adjusting mechanism (87A) adjusts the pressure of the gases inside the common exhaust pipe (81A). The control section (102) acquires processing-time pressure adjusting information that specifies the actions of the plurality of first pressure adjusting mechanisms (73A1, 73B1) and the action of the second pressure adjusting mechanism (87A) from a learned model (LM).

Description

[0001] This case is a divisional application of the patent application filed on May 12, 2021, with application number 202180038611.3 and invention title "Substrate processing apparatus, substrate processing method, method for generating learning data, learning method, learning apparatus, method for generating learned model and learned model". Technical Field

[0002] This invention relates to a substrate processing apparatus, a substrate processing method, a method for generating learning data, a learning method, a learning apparatus, a method for generating a learned model, and a learned model. Background Technology

[0003] Substrate processing apparatuses with multiple processing units are known. For example, the substrate processing apparatus of Patent Document 1 has multiple processing units and one exhaust flow path (shared exhaust pipe). In the substrate processing apparatus of Patent Document 1, the waste gas (gas) discharged from each processing unit flows into one exhaust flow path and is discharged to the outside of the substrate processing apparatus.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2019-91758 Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] However, in a structure where gases from multiple processing units flow into a single common exhaust pipe, the flow rate of the gas flowing in the common exhaust pipe is prone to variation. Consequently, the exhaust pressure of the common exhaust pipe is also prone to variation. If the exhaust pressure of the common exhaust pipe varies, the exhaust pressure of each processing unit will also vary, potentially leading to the generation of particles.

[0009] For example, the exhaust pressure of the common exhaust pipe varies depending on the number of processing units simultaneously processing the substrate. Furthermore, even when the number of processing units simultaneously processing the substrate is the same, the exhaust pressure of the common exhaust pipe varies if the positions of the processing units are different. This is because the distance from each processing unit to the common exhaust pipe varies for each processing unit, resulting in different pressure losses for each unit. These are the main reasons why the exhaust pressure of the common exhaust pipe varies. Therefore, it is not easy to properly adjust the exhaust pressure of the processing units.

[0010] The present invention was made in view of the above-mentioned problems, and its object is to provide a substrate processing apparatus, a substrate processing method, a method for generating learning data, a learning method, a learning apparatus, a method for generating a learned model, and a learned model that can more appropriately adjust the exhaust pressure of the processing unit.

[0011] Methods for solving problems

[0012] According to one aspect of the present invention, a substrate processing apparatus includes multiple processing units, multiple individual exhaust pipes, a common exhaust pipe, multiple first pressure regulating mechanisms, a second pressure regulating mechanism, and a control unit. The multiple processing units process the substrate by supplying the processing liquid to the substrate based on processing time sequence information indicating the order in which the substrate is processed using the processing liquid. Gas discharged from the multiple processing units flows into the multiple individual exhaust pipes. Gas discharged from the multiple individual exhaust pipes flows into the common exhaust pipe. The multiple first pressure regulating mechanisms regulate the pressure of the gas flowing from the multiple processing units into the multiple individual exhaust pipes. The second pressure regulating mechanisms regulate the pressure of the gas inside the common exhaust pipe. The control unit obtains processing time pressure regulation information, specifying the operation of the multiple first pressure regulating mechanisms and the operation of the second pressure regulating mechanisms, from a learned model based on the processing time sequence information, and controls the multiple first pressure regulating mechanisms and the second pressure regulating mechanisms based on the processing time pressure regulation information. The learned model is constructed by machine learning of learning data that associates processing result information with learning object information, the processing result information indicating the result of processing the learning object substrate using the processing liquid. The learning object information includes learning pressure adjustment information and learning sequence information. The learning pressure adjustment information indicates the operation of multiple first pressure adjustment mechanisms and the operation of second pressure adjustment mechanisms when processing the learning object substrate. The learning sequence information indicates the order in which the learning object substrate is processed using the processing liquid.

[0013] In one embodiment, the processing liquid comprises multiple processing liquids, and the gas discharged from the processing unit comprises multiple gases corresponding to the multiple processing liquids. There are multiple common exhaust pipes, each corresponding to one of the multiple gases. There are multiple second pressure regulating mechanisms, each corresponding to one of the multiple common exhaust pipes.

[0014] In one embodiment, the plurality of individual exhaust pipes each include a plurality of individual exhaust pipes corresponding to a plurality of the plurality of gases.

[0015] In one embodiment, the substrate processing apparatus further includes a switching mechanism. The switching mechanism switches the destination of the gas discharged from the corresponding processing unit among the various individual exhaust pipes.

[0016] In one embodiment, the first pressure regulating mechanism is disposed on the switching mechanism.

[0017] In one embodiment, the substrate processing apparatus further includes a storage unit for storing the learned model.

[0018] In one embodiment, the learning object information further includes at least one of the following: concentration information, representing the concentration of the treatment liquid; temperature information, representing the internal temperature of each of the plurality of treatment units; position information, representing the position of the plurality of treatment units; and distance information, representing the distance from which the gas discharged from the treatment unit is discharged from the downstream end of the common exhaust pipe.

[0019] According to another aspect of the present invention, a substrate processing method is a method of processing a substrate in a substrate processing apparatus. The substrate processing apparatus includes multiple processing units, multiple individual exhaust pipes, a common exhaust pipe, multiple first pressure regulating mechanisms, a second pressure regulating mechanism, and a control unit. The multiple processing units process the substrate by supplying processing liquid to the substrate based on processing time sequence information indicating the order in which the substrate is processed using processing liquid. Gas discharged from the multiple processing units flows into the multiple individual exhaust pipes. Gas discharged from the multiple individual exhaust pipes flows into the common exhaust pipe. The multiple first pressure regulating mechanisms regulate the pressure of the gas flowing from the multiple processing units into the multiple individual exhaust pipes. The second pressure regulating mechanisms regulate the pressure of the gas inside the common exhaust pipe. The control unit controls the multiple first pressure regulating mechanisms and the second pressure regulating mechanisms. The substrate processing method includes the following steps: obtaining processing time pressure regulating information from a learned model, which specifies the operation of the multiple first pressure regulating mechanisms and the operation of the second pressure regulating mechanisms, based on the processing time sequence information; and controlling the multiple first pressure regulating mechanisms and the second pressure regulating mechanisms based on the processing time pressure regulating information. The learned model is constructed by performing machine learning on learning data that associates processing result information with learning object information. The processing result information represents the result of processing the learning object substrate using the processing fluid. The learning object information includes learning pressure adjustment information and learning sequence information. The learning pressure adjustment information represents the actions of multiple first pressure adjustment mechanisms and second pressure adjustment mechanisms during the processing of the learning object substrate. The learning sequence information represents the order in which the learning object substrate is processed using the processing fluid.

[0020] According to another aspect of the present invention, a method for generating learning data includes the following steps: obtaining processing result information representing the result of a substrate processing apparatus processing a learning target substrate; obtaining learning target information; and associating the processing result information with the learning target information and storing it as learning data in a storage unit. The substrate processing apparatus includes multiple processing units, multiple individual exhaust pipes, a common exhaust pipe, multiple first pressure regulating mechanisms, a second pressure regulating mechanism, and a control unit. The multiple processing units process the learning target substrate by supplying the processing liquid to the learning target substrate based on learning time sequence information representing the order in which the learning target substrate is processed using a processing liquid. Gas discharged from the multiple processing units flows into the multiple individual exhaust pipes. Gas discharged from the multiple individual exhaust pipes flows into the common exhaust pipe. The multiple first pressure regulating mechanisms regulate the pressure of the gas flowing from the multiple processing units into the multiple individual exhaust pipes. The second pressure regulating mechanism regulates the pressure of the gas inside the common exhaust pipe. The control unit controls the multiple first pressure regulating mechanisms and the second pressure regulating mechanism. The learning object information includes learning pressure adjustment information and learning sequence information. The learning pressure adjustment information indicates the actions of multiple first pressure adjustment mechanisms and second pressure adjustment mechanisms when processing the learning object substrate.

[0021] According to another aspect of the present invention, the learning method includes the following steps: obtaining learning data generated according to the above-described method for generating learning data; and inputting the learning data into a learning program to perform machine learning on the learning data.

[0022] According to another aspect of the present invention, a learning apparatus includes a storage unit and a learning unit. The storage unit stores learning data generated according to the method for generating learning data. The learning unit inputs the learning data into a learning program to perform machine learning on the learning data.

[0023] According to another aspect of the present invention, a method for generating a learned model includes the following steps: obtaining learning data generated according to the above-described method for generating learning data; and generating a learned model by performing machine learning on the learning data.

[0024] According to another aspect of the present invention, the learned model is constructed by performing machine learning on learning data generated according to the above-described method for generating learning data.

[0025] Invention Effects

[0026] According to the substrate processing apparatus, substrate processing method, method for generating learning data, learning method, learning apparatus, method for generating learned model, and learned model of the present invention, the exhaust pressure of the processing unit can be adjusted more appropriately. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of a substrate processing learning system equipped with a substrate processing apparatus according to an embodiment of the present invention.

[0028] Figure 2 This is a top view showing the interior of a substrate processing apparatus according to an embodiment of the present invention.

[0029] Figure 3 This is a right-side view showing the right-side structure of the substrate processing apparatus according to an embodiment of the present invention.

[0030] Figure 4 This is a right-side view showing the left-side structure of a substrate processing apparatus according to an embodiment of the present invention.

[0031] Figure 5 This is a right-side view showing the central structure of the substrate processing apparatus according to an embodiment of the present invention.

[0032] Figure 6 This is a front view showing the first, second, and third individual exhaust pipes.

[0033] Figure 7 This is a side view showing the interior of the processing unit.

[0034] Figure 8 This is a top view showing the processing unit, the switching mechanism, and the first to third individual exhaust pipes.

[0035] Figure 9 This is the main view for switching mechanisms.

[0036] Figure 10 This is a system diagram of the exhaust path originating from multiple processing units.

[0037] Figure 11 This is a top view of a substrate processing apparatus according to an embodiment of the present invention.

[0038] Figure 12 This is a block diagram illustrating a substrate processing apparatus according to an embodiment of the present invention.

[0039] Figure 13 This is a flowchart illustrating the process of obtaining damper opening adjustment information.

[0040] Figure 14 It is a diagram that represents device information.

[0041] Figure 15 This is a diagram representing an example of sequential information during processing.

[0042] Figure 16 This is a diagram illustrating an example of the operation of a substrate processing apparatus according to an embodiment of the present invention.

[0043] Figure 17 This is a block diagram representing a substrate processing apparatus.

[0044] Figure 18 It is a graph representing test process data.

[0045] Figure 19 This is a diagram representing an example of sequential information during learning.

[0046] Figure 20 This is a block diagram representing a data generation device for learning.

[0047] Figure 21 This is a flowchart of a learning data generation method according to an embodiment of the present invention.

[0048] Figure 22 This is a block diagram illustrating a learning device according to an embodiment of the present invention.

[0049] Figure 23 This is a flowchart illustrating the learning method and the method for generating the learned model according to embodiments of the present invention.

[0050] Figure 24 This is a system diagram of the exhaust path of a substrate processing apparatus according to another embodiment of the present invention. Detailed Implementation

[0051] The following refers to the attached diagram ( Figures 1 to 24 This section describes the substrate processing apparatus, substrate processing method, method for generating learning data, learning method, learning apparatus, method for generating a learned model, and embodiments of the learned model according to the present invention. However, the present invention is not limited to the following embodiments. Furthermore, descriptions of repeated parts are sometimes appropriately omitted. Additionally, in the figures, the same or equivalent parts are labeled with the same reference numerals without being described repeatedly.

[0052] First, refer to Figure 1 This describes a substrate processing learning system 200 equipped with the substrate processing apparatus 100 of this embodiment. Figure 1 This is a schematic diagram of the substrate processing learning system 200. (As shown...) Figure 1 As shown, the substrate processing learning system 200 includes a substrate processing apparatus 100, a substrate processing apparatus 100L, an inspection apparatus 300, a learning data generation apparatus 400, and a learning apparatus 500.

[0053] The substrate processing apparatus 100 processes the substrate to be processed using a processing liquid. The substrate processing apparatus 100 is a single-piece type apparatus that processes each substrate individually. Typically, the substrate to be processed is approximately circular.

[0054] The substrate processing apparatus 100L processes the learning target substrate using a processing liquid. Furthermore, the structure of the learning target substrate is the same as that of the processing target substrate. Typically, the learning target substrate is approximately circular. The substrate processing apparatus 100L is a monolithic device that processes learning target substrates one by one. The structure of the substrate processing apparatus 100L is substantially the same as that of the substrate processing apparatus 100. The substrate processing apparatus 100L may also be the same object as the substrate processing apparatus 100. For example, the same substrate processing apparatus may have previously processed a learning target substrate and then processed the processing target substrate. Alternatively, the substrate processing apparatus 100L may also be another product having a structure substantially the same as that of the substrate processing apparatus 100.

[0055] In the following description of this specification, the learning target substrate will sometimes be referred to as "learning target substrate WL" and the processing target substrate will sometimes be referred to as "processing target substrate Wp". In addition, when it is not necessary to distinguish between the learning target substrate WL and the processing target substrate Wp in the description, the learning target substrate WL and the processing target substrate Wp will sometimes be referred to as "substrate W".

[0056] The substrate W is, for example, a semiconductor wafer, a substrate for a liquid crystal display device, a substrate for a plasma display, a substrate for a field emission display (FED), a substrate for an optical disc, a substrate for a magnetic disk, a substrate for an optical disk, a substrate for a photomask, a ceramic substrate, or a substrate for a solar cell.

[0057] The substrate processing apparatus 100L obtains learning object information LTD by processing the learning object substrate WL. For example, the substrate processing apparatus 100L may obtain the learning object information LTD each time it processes the learning object substrate WL. Alternatively, the substrate processing apparatus 100L may obtain the learning object information LTD each time it processes a predetermined number of learning object substrates WL. Specifically, the learning object information LTD may also be obtained at the end of each processing of all multiple learning object substrates WL constituting a batch. The learning object information LTD, for example, includes information obtained from the processing unit 1 (refer to...). Figure 8 Time series data of the pressure (exhaust pressure) of the gas discharged. Hereinafter, the learning object substrate WL processed by the substrate processing apparatus 100L will sometimes be referred to as "processed substrate WLa".

[0058] The learning object information LTD is input to the learning data generation device 400. For example, the learning object information LTD can also be input to the learning data generation device 400 via communication. Alternatively, the learning object information LTD can also be input to the learning data generation device 400 via a storage medium. The storage medium can be, for example, an optical disc such as a CD (Compact Disc) or a storage device such as a USB (Universal Serial Bus) memory. Alternatively, the learning object information LTD can also be input by an operator using an input device such as a keyboard, mouse, or touch sensor.

[0059] In detail, the learning object information LTD is input to the learning data generation apparatus 400 in association with the substrate information. The substrate information includes, for example, batch identification information, substrate identification information, and processing order information. Batch identification information is information used to identify batches (e.g., batch number). A batch represents a processing unit of substrate W. One batch consists of a predetermined number of substrates W. Substrate identification information is information used to identify substrate W. Processing order information is information indicating the order in which each learning object substrate WL is processed within a batch.

[0060] The inspection device 300 inspects the processed substrates WLa to generate processing result information RS. The processing result information RS can be generated for each processed substrate WLa, or for a predetermined number of processed substrates WLa. For example, it can also be generated for each batch.

[0061] The processing result information RS indicates the result of treating the learning object substrate WLa with the processing solution. Specifically, the inspection device 300 is a particle measuring device, and the processing result information RS indicates the number of particles generated in the substrate WLa after treatment. Alternatively, the inspection device 300 may also determine whether the number of particles exceeds a predetermined value. In this case, the processing result information RS indicates the result of determining whether the number of particles exceeds the predetermined value. Alternatively, the operator may determine whether the number of particles exceeds the predetermined value by referring to the particle count measured by the inspection device 300. In this case, the processing result information RS is generated by the operator.

[0062] The processing result information RS is input to the learning data generation device 400. For example, the processing result information RS can also be input to the learning data generation device 400 via communication. Alternatively, the processing result information RS can also be input to the learning data generation device 400 via a storage medium. Or, the processing result information RS can also be input by an operator using an input device.

[0063] In detail, the processing result information RS is input to the learning data generation apparatus 400 in association with the substrate information. When the processing result information RS is input to the learning data generation apparatus 400 via communication, the substrate information is also input to the learning data generation apparatus 400 together with the processing result information RS via communication. In this case, the substrate information is input to the inspection apparatus 300. The same applies when the processing result information RS is input to the learning data generation apparatus 400 via a storage medium. When an operator inputs the processing result information RS to the learning data generation apparatus 400, the operator operates the input device to input the substrate information and the processing result information RS together into the learning data generation apparatus 400.

[0064] The learning data generation apparatus 400 associates the learning object information LTD with the processing result information RS to generate learning data LFD. Specifically, the learning data generation apparatus 400 uses substrate information to associate the learning object information LTD with the processing result information RS. The learning data LFD can be generated for each learning object substrate WL, or for each predetermined number of learning object substrates WL. For example, it can also be generated for each batch.

[0065] The learning device 500 generates (builds) a learned model LM by performing machine learning on multiple learning datasets (LFDs). In detail, the learning device 500 identifies certain rules from the multiple learning datasets (LFDs) and generates a model (learned model LM) to represent those rules.

[0066] The substrate processing apparatus 100 inputs apparatus information MD into the learned model LM. As a result, damper opening adjustment information PA is output from the learned model LM. The apparatus information MD contains processing sequence information indicating the order in which the substrate processing apparatus 100 processes the substrate Wp to be processed using the processing liquid. The damper opening adjustment information PA is used to adjust the damper opening from the processing unit 1 (refer to...). Figure 8 The information referenced when determining the pressure of the discharged gas. Specifically, it specifies the first baffle 73 (refer to...). Figure 8 The time series data of the opening degree of ).

[0067] Next, refer to Figures 2 to 15 Further explanation of the substrate processing apparatus 100. Figure 2 This is a top view showing the interior of the substrate processing apparatus 100. (Example) Figure 2 As shown, the substrate processing apparatus 100 includes multiple (e.g., four) carrier placement units 150, a conveying mechanism 151, a substrate placement unit 152A, a conveying mechanism 153A, and processing units 1A1, 1B1, 1C1, and 1D1.

[0068] Multiple carrier placement portions 150 are disposed at the end of the substrate processing apparatus 100. In this specification, for convenience, the embodiments are sometimes described using the front-back direction and the left-right direction. Here, the side where the multiple carrier placement portions 150 are disposed is the front side of the substrate processing apparatus 100.

[0069] A conveying mechanism 151 is disposed on the rear side of a plurality of carrier placement sections 150. A substrate placement section 152A is disposed on the rear side of the conveying mechanism 151. A conveying mechanism 153A is disposed on the rear side of the substrate placement section 152A.

[0070] Processing units 1A1, 1B1, 1C1, and 1D1 are arranged around the conveying mechanism 153A. Specifically, processing units 1A1 and 1B1 are arranged to the right of the conveying mechanism 153A, and processing units 1C1 and 1D1 are arranged to the left of the conveying mechanism 153A. Furthermore, processing unit 1B1 is arranged behind processing unit 1A1, and processing unit 1D1 is arranged behind processing unit 1C1.

[0071] Multiple carrier mounting sections 150 are arranged in a left-right direction. Each carrier mounting section 150 has a carrier C mounted on it. The carrier C contains multiple substrates W. The carrier C is, for example, a FOUP (Front Opening Unift Pod).

[0072] The substrate W is transported into the processing units 1A1, 1B1, 1C1, and 1D1 via the transport mechanism 153A. After holding the substrate W transported in by the transport mechanism 153A, the processing units 1A1, 1B1, 1C1, and 1D1 respectively supply processing liquid to the substrate W to process it.

[0073] like Figure 2 As shown, the substrate processing apparatus 100 also includes a first separate exhaust pipe 41A, a second separate exhaust pipe 42A, a third separate exhaust pipe 43A, a switching mechanism 51A1, and a control device 101.

[0074] The first separate exhaust pipe 41A, the second separate exhaust pipe 42A, the third separate exhaust pipe 43A, and the switching mechanism 51A1 are disposed near the processing unit 1A1. In this embodiment, the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, the third separate exhaust pipe 43A, and the switching mechanism 51A1 are adjacent to the processing unit 1A1. The first separate exhaust pipe 41A, the second separate exhaust pipe 42A, and the third separate exhaust pipe 43A extend in the vertical direction.

[0075] Gas discharged from processing unit 1A1 flows into one of the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, and the third separate exhaust pipe 43A. Specifically, switching mechanism 51A1 switches the destination of the gas discharged from processing unit 1A1 between the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, and the third separate exhaust pipe 43A. Switching mechanism 51A1 is controlled by control device 101.

[0076] In detail, the processing unit 1A1 is capable of supplying various processing liquids to the substrate W. Therefore, the gas discharged from the processing unit 1A1 includes various gases corresponding to the various processing liquids. In this embodiment, the various processing liquids include a first processing liquid, a second processing liquid, and a third processing liquid. The processing unit 1A1 is capable of supplying the first processing liquid, the second processing liquid, and the third processing liquid to the substrate W.

[0077] When the processing unit 1A1 supplies the first processing liquid to the substrate W, the switching mechanism 51A1 causes the gas discharged from the processing unit 1A1 to flow into the first separate exhaust pipe 41A. When the processing unit 1A1 supplies the second processing liquid to the substrate W, the switching mechanism 51A1 causes the gas discharged from the processing unit 1A1 to flow into the second separate exhaust pipe 42A. When the processing unit 1A1 supplies the third processing liquid to the substrate W, the switching mechanism 51A1 causes the gas discharged from the processing unit 1A1 to flow into the third separate exhaust pipe 43A.

[0078] like Figure 2 As shown, the substrate processing apparatus 100 further includes a first separate exhaust pipe 41B, a second separate exhaust pipe 42B, a third separate exhaust pipe 43B, and a switching mechanism 51B1. The first separate exhaust pipe 41B, the second separate exhaust pipe 42B, the third separate exhaust pipe 43B, and the switching mechanism 51B1 are disposed near the processing unit 1B1. The switching mechanism 51B1 is controlled by the control device 101. The structures of the first separate exhaust pipe 41B, the second separate exhaust pipe 42B, the third separate exhaust pipe 43B, and the switching mechanism 51B1 are substantially the same as the structures of the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, the third separate exhaust pipe 43A, and the switching mechanism 51A1, therefore, the description here is omitted.

[0079] like Figure 2As shown, the substrate processing apparatus 100 also includes a first separate exhaust pipe 41C, a second separate exhaust pipe 42C, a third separate exhaust pipe 43C, and a switching mechanism 51C1. The first separate exhaust pipe 41C, the second separate exhaust pipe 42C, the third separate exhaust pipe 43C, and the switching mechanism 51C1 are disposed near the processing unit 1C1. The switching mechanism 51C1 is controlled by the control device 101. The structure of the first separate exhaust pipe 41C, the second separate exhaust pipe 42C, the third separate exhaust pipe 43C, and the switching mechanism 51C1 is substantially the same as the structure of the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, the third separate exhaust pipe 43A, and the switching mechanism 51A1, therefore, the description here is omitted.

[0080] like Figure 2 As shown, the substrate processing apparatus 100 further includes a first separate exhaust pipe 41D, a second separate exhaust pipe 42D, a third separate exhaust pipe 43D, and a switching mechanism 51D1. The first separate exhaust pipe 41D, the second separate exhaust pipe 42D, the third separate exhaust pipe 43D, and the switching mechanism 51D1 are disposed near the processing unit 1D1. The switching mechanism 51D1 is controlled by the control device 101. The structure of the first separate exhaust pipe 41D, the second separate exhaust pipe 42D, the third separate exhaust pipe 43D, and the switching mechanism 51D1 is substantially the same as the structure of the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, the third separate exhaust pipe 43A, and the switching mechanism 51A1, therefore, the description here is omitted.

[0081] Furthermore, when it is not necessary to distinguish between the first individual exhaust pipes 41A to 41D in the description, the first individual exhaust pipes 41A to 41D are sometimes referred to as "first individual exhaust pipe 41". Similarly, when it is not necessary to distinguish between the second individual exhaust pipes 42A to 42D in the description, the second individual exhaust pipes 42A to 42D are sometimes referred to as "second individual exhaust pipe 42". In addition, when it is not necessary to distinguish between the third individual exhaust pipes 43A to 43D in the description, the third individual exhaust pipes 43A to 43D are sometimes referred to as "third individual exhaust pipe 43".

[0082] like Figure 2 As shown, the substrate processing apparatus 100 also includes air supply pipes 44A to 44D. Air supply pipe 44A is disposed near processing unit 1A1. Air supply pipe 44B is disposed near processing unit 1B1. Air supply pipe 44C is disposed near processing unit 1C1. Air supply pipe 44D is disposed near processing unit 1D1. Air supply pipes 44A to 44D extend in the vertical direction.

[0083] Next, refer to Figure 3 and Figure 4 Further explanation of the substrate processing apparatus 100. Figure 3 This is a right-side view showing the right-side structure of the substrate processing apparatus 100. (Example) Figure 3 As shown, the substrate processing apparatus 100 further includes processing units 1A2 to 1A6 and processing units 1B2 to 1B6.

[0084] Processing units 1A2 to 1A6 are disposed below processing unit 1A1. Specifically, processing units 1A1 to 1A6 are arranged in a row along the vertical direction. More specifically, processing units 1A1 to 1A6 are stacked from top to bottom in this order. Processing units 1A2 to 1A6 have a structure that is substantially the same as that of processing unit 1A1.

[0085] Processing units 1B2 to 1B6 are disposed below processing unit 1B1. Specifically, processing units 1B1 to 1B6 are arranged in a row along the vertical direction. More specifically, processing units 1B1 to 1B6 are stacked from top to bottom in this order. Processing units 1B2 to 1B6 have a structure that is substantially the same as that of processing unit 1B1.

[0086] In addition, when it is not necessary to distinguish between processing units 1A1 to 1A6 in the description, processing units 1A1 to 1A6 are sometimes referred to as "processing unit 1A". Similarly, when it is not necessary to distinguish between processing units 1B1 to 1B6 in the description, processing units 1B1 to 1B6 are sometimes referred to as "processing unit 1B".

[0087] Air supply pipe 44A supplies clean air to processing units 1A1 to 1A6. Similarly, air supply pipe 44B supplies clean air to processing units 1B1 to 1B6.

[0088] like Figure 3 As shown, the substrate processing apparatus 100 also includes switching mechanisms 51A2 to 51A6. Switching mechanisms 51A2 to 51A6 are disposed below switching mechanism 51A1. Switching mechanisms 51A2 to 51A6 correspond to processing units 1A2 to 1A6, respectively. Switching mechanisms 51A2 to 51A6 have a structure substantially the same as switching mechanism 51A1. Switching mechanisms 51A2 to 51A6 are defined by reference... Figure 2 The control device 101 described above controls the system.

[0089] like Figure 3 As shown, the substrate processing apparatus 100 also includes switching mechanisms 51B2 to 51B6. Switching mechanisms 51B2 to 51B6 are disposed below switching mechanism 51B1. Switching mechanisms 51B2 to 51B6 correspond to processing units 1B2 to 1B6, respectively. Switching mechanisms 51B2 to 51B6 have a structure substantially the same as that of switching mechanism 51B1. Switching mechanisms 51B2 to 51B6 are defined by reference... Figure 2 The control device 101 described above controls the system.

[0090] In addition, when it is not necessary to distinguish between switching mechanisms 51A1 to 51A6 in the description, switching mechanisms 51A1 to 51A6 are sometimes referred to as "switching mechanism 51A". Similarly, when it is not necessary to distinguish between switching mechanisms 51B1 to 51B6 in the description, switching mechanisms 51B1 to 51B6 are sometimes referred to as "switching mechanism 51B".

[0091] Figure 4 This is a right-side view showing the left-side structure of the substrate processing apparatus 100. (Example) Figure 4 As shown, the substrate processing apparatus 100 further includes processing units 1C2 to 1C6 and processing units 1D2 to 1D6.

[0092] Processing units 1C2 to 1C6 are disposed below processing unit 1C1. Specifically, processing units 1C1 to 1C6 are arranged in a row along the vertical direction. More specifically, processing units 1C1 to 1C6 are stacked from top to bottom in this order. Processing units 1C2 to 1C6 have a structure that is substantially the same as that of processing unit 1C1.

[0093] Processing units 1D2 to 1D6 are positioned below processing unit 1D1. Specifically, processing units 1D1 to 1D6 are arranged in a vertical row. More specifically, processing units 1D1 to 1D6 are stacked from top to bottom in this order. Processing units 1D2 to 1D6 have a structure substantially the same as processing unit 1D1.

[0094] Furthermore, when it is not necessary to distinguish between processing units 1C1 to 1C6 in the description, processing units 1C1 to 1C6 are sometimes referred to as "processing unit 1C". Similarly, when it is not necessary to distinguish between processing units 1D1 to 1D6 in the description, processing units 1D1 to 1D6 are sometimes referred to as "processing unit 1D". Moreover, when it is not necessary to distinguish between processing units 1A to 1D in the description, processing units 1A to 1D are sometimes referred to as "processing unit 1".

[0095] Air supply pipe 44C supplies clean air to processing units 1C1 to 1C6. Similarly, air supply pipe 44D supplies clean air to processing units 1D1 to 1D6. In addition, when it is not necessary to distinguish between air supply pipes 44A to 44D in the description, air supply pipes 44A to 44D are sometimes referred to as "air supply pipe 44".

[0096] like Figure 4 As shown, the substrate processing apparatus 100 further includes switching mechanisms 51C2 to 51C6. Switching mechanisms 51C2 to 51C6 are disposed below switching mechanism 51C1. Switching mechanisms 51C2 to 51C6 correspond to processing units 1C2 to 1C6, respectively. Switching mechanisms 51C2 to 51C6 have a structure substantially the same as that of switching mechanism 51C1. Switching mechanisms 51C2 to 51C6 are defined by reference... Figure 2The control device 101 described above controls the system.

[0097] like Figure 4 As shown, the substrate processing apparatus 100 also includes switching mechanisms 51D2 to 51D6. Switching mechanisms 51D2 to 51D6 are disposed below switching mechanism 51D1. Switching mechanisms 51D2 to 51D6 correspond to processing units 1D2 to 1D6, respectively. Switching mechanisms 51D2 to 51D6 have a structure substantially the same as switching mechanism 51D1. Switching mechanisms 51D2 to 51D6 are defined by reference... Figure 2 The control device 101 described above controls the system.

[0098] Furthermore, when it is not necessary to distinguish between switching mechanisms 51C1 to 51C6 in the description, switching mechanisms 51C1 to 51C6 are sometimes referred to as "switching mechanism 51C". Similarly, when it is not necessary to distinguish between switching mechanisms 51D1 to 51D6 in the description, switching mechanisms 51D1 to 51D6 are sometimes referred to as "switching mechanism 51D". Moreover, when it is not necessary to distinguish between switching mechanisms 51A to 51D in the description, switching mechanisms 51A to 51D are sometimes referred to as "switching mechanism 51".

[0099] Next, refer to Figures 2-5 Further explanation of the substrate processing apparatus 100. Figure 5 This is a right-side view showing the central structure of the substrate processing apparatus 100. (Example) Figure 5 As shown, the substrate processing apparatus 100 also includes a substrate mounting section 152B and a conveying mechanism 153B.

[0100] The substrate mounting section 152B is disposed below the substrate mounting section 152A. The conveying mechanism 151 transfers the substrate W between the plurality of carrier mounting sections 150 and the substrate mounting section 152A, and between the plurality of carrier mounting sections 150 and the substrate mounting section 152B. The conveying mechanism 151 is controlled by the control device 101.

[0101] Specifically, the conveying mechanism 151 accesses any one of the plurality of carrier placement sections 150 and holds the substrate W. Then, the conveying mechanism 151 conveys the substrate W to one of the substrate placement sections 152A and 152B. For example, the conveying mechanism 151 conveys the substrate W to the substrate placement section 152A and the substrate placement section 152B alternately.

[0102] Furthermore, when the substrate W is placed in the substrate placement section 152A by the transfer mechanism 153A, the transfer mechanism 151 accesses the substrate placement section 152A to hold the substrate W. Then, the transfer mechanism 151 transfers the substrate W to any one of the plurality of carrier placement sections 150. Specifically, the substrate W is received in any one of the plurality of carrier placement sections 150.

[0103] Similarly, when the substrate W is placed on the substrate placement section 152B by the transfer mechanism 153B, the transfer mechanism 151 accesses the substrate placement section 152B to hold the substrate W. Then, the transfer mechanism 151 transfers the substrate W to any one of the plurality of carrier placement sections 150.

[0104] The conveying mechanism 153A is located at the substrate mounting section 152A and the reference. Figure 3 and Figure 4 The processing units 1A1-1A3, 1B1-1B3, 1C1-1C3, and 1D1-1D3 are described, and the transfer substrate W is between them. The conveying mechanism 153A is controlled by the control device 101.

[0105] Specifically, the transfer mechanism 153A accesses the substrate placement section 152A to hold the substrate W placed in the substrate placement section 152A. Then, the transfer mechanism 153A moves the substrate W into any one of the processing units 1A1-1A3, 1B1-1B3, 1C1-1C3, and 1D1-1D3. Additionally, the transfer mechanism 153A accesses any one of the processing units 1A1-1A3, 1B1-1B3, 1C1-1C3, and 1D1-1D3 to remove the substrate W. Finally, the transfer mechanism 153A transfers the substrate W to the substrate placement section 152A.

[0106] The conveying mechanism 153B is positioned below the conveying mechanism 153A. The conveying mechanism 153B is located between the substrate mounting section 152B and the reference section. Figure 3 and Figure 4 The processing units 1A4-1A6, 1B4-1B6, 1C4-1C6, and 1D4-1D6 are described, and the transfer substrate W is between them. The conveying mechanism 153B is controlled by the control device 101.

[0107] Specifically, the transfer mechanism 153B accesses the substrate placement section 152B to hold the substrate W placed in the substrate placement section 152B. Then, the transfer mechanism 153B transfers the substrate W into any one of the processing units 1A4-1A6, 1B4-1B6, 1C4-1C6, and 1D4-1D6. Additionally, the transfer mechanism 153B accesses any one of the processing units 1A4-1A6, 1B4-1B6, 1C4-1C6, and 1D4-1D6 to remove the substrate W. Afterward, the transfer mechanism 153B transfers the substrate W to the substrate placement section 152B.

[0108] Furthermore, when a substrate W is being processed in a certain processing unit 1, the conveying mechanisms 151, 153A, and 153B convey the substrate W to another processing unit 1 that is not currently processing the substrate W. Therefore, the substrate processing apparatus 100 can process multiple substrates W simultaneously. Hereinafter, unless it is necessary to distinguish between the conveying mechanisms 153A and 153B in the description, the conveying mechanisms 153A and 153B will sometimes be referred to as "conveying mechanism 153".

[0109] Next, refer to Figure 6 Further explanation of the substrate processing apparatus 100. Figure 6 This is a front view showing the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, and the third separate exhaust pipe 43A.

[0110] like Figure 6 As shown, the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, and the third separate exhaust pipe 43A extend from the processing unit 1A6 to the upper wall of the substrate processing apparatus 100. The switching mechanisms 51A1 to 51A6 are connected to the first separate exhaust pipe 41A, the second separate exhaust pipe 42A, and the third separate exhaust pipe 43A.

[0111] Furthermore, although not shown in the diagram, the first individual exhaust pipes 41B-41D, the second individual exhaust pipes 42B-42D, and the third individual exhaust pipes 43B-43D also extend from the lowest processing unit 1 (processing units 1B6-1D6) to the upper wall of the substrate processing apparatus 100. Additionally, switching mechanisms 51B1-51B6 are connected to the first individual exhaust pipe 41B, the second individual exhaust pipe 42B, and the third individual exhaust pipe 43B. Switching mechanisms 51C1-51C6 are connected to the first individual exhaust pipe 41C, the second individual exhaust pipe 42C, and the third individual exhaust pipe 43C. Switching mechanisms 51D1-51D6 are connected to the first individual exhaust pipe 41D, the second individual exhaust pipe 42D, and the third individual exhaust pipe 43D.

[0112] Next, refer to Figure 7 The processing unit 1 will be described. Figure 7 This is a side view showing the interior of processing unit 1. (Example) Figure 7 As shown, the processing unit 1 includes a processing housing 2, a substrate holding part 31, a rotation drive part 32, a first nozzle 33, a first nozzle moving part 33a, a second nozzle 34, a second nozzle moving part 34a, a third nozzle 35, a third nozzle moving part 35a, a fourth nozzle 36, a fourth nozzle moving part 36a, a cup part 37, and a blowing unit 38. Additionally, the substrate processing apparatus 100 also includes a first processing liquid supply pipe 333, a second processing liquid supply pipe 343, a third processing liquid supply pipe 353, a rinsing liquid supply pipe 363, and an air supply pipe 39.

[0113] The processing housing 2 has a box shape. The processing housing 2 houses a substrate holding part 31, a rotation drive part 32, a first nozzle 33, a first nozzle moving part 33a, a second nozzle 34, a second nozzle moving part 34a, a third nozzle 35, a third nozzle moving part 35a, a fourth nozzle 36, a fourth nozzle moving part 36a, a cup part 37, and a blowing unit 38. Additionally, the processing housing 2 houses a portion of a first processing liquid supply pipe 333, a portion of a second processing liquid supply pipe 343, a portion of a third processing liquid supply pipe 353, a portion of a rinsing liquid supply pipe 363, and a portion of an air supply pipe 39. The substrate W, which is moved into the processing unit 1, is housed inside the processing housing 2.

[0114] The substrate holding portion 31 holds the substrate W horizontally. The substrate holding portion 31 is, for example, a vacuum rotary chuck. However, the method by which the substrate holding portion 31 holds the substrate W is not limited to vacuum. The substrate holding portion 31 can also hold the substrate W in a clamping or Bernoulli manner, for example.

[0115] The rotation drive unit 32 causes the substrate holding unit 31 to rotate about the rotation axis AX. As a result, the substrate W and the substrate holding unit 31 rotate together about the rotation axis AX. The rotation axis AX extends in the vertical direction.

[0116] The first nozzle 33 supplies the first processing liquid to the substrate W from above. Specifically, the first nozzle 33 sprays the first processing liquid onto the rotating substrate W. The first nozzle moving part 33a moves the first nozzle 33 between a processing position and a retracted position. When the first nozzle 33 moves to the processing position, it is opposite to the substrate W in a top-view view. When the first nozzle 33 moves to the retracted position, it is not opposite to the substrate W in a top-view view. Specifically, when the first nozzle 33 moves to the retracted position, it is retracted to the periphery of the substrate W in a top-view view.

[0117] Specifically, the first nozzle moving part 33a has a first nozzle arm 331 and a first nozzle drive part 332. The first nozzle arm 331 extends in a generally horizontal direction. A first nozzle 33 is disposed at the front end of the first nozzle arm 331. The first nozzle drive part 332 rotates the first nozzle arm 331 in a generally horizontal plane about a rotation axis extending in the vertical direction. As a result, the first nozzle 33 moves circumferentially about the rotation axis extending in the vertical direction. The first nozzle drive part 332 includes an electric motor capable of rotating in both directions.

[0118] The first processing liquid supply pipe 333 supplies the first processing liquid to the first nozzle 33. The first processing liquid supply pipe 333 is a tubular component for the flow of the first processing liquid. In this embodiment, the first processing liquid is an acidic liquid. For example, the first processing liquid is hydrofluoric acid, hydrochloric acid-hydrogen peroxide solution (SPM), sulfuric acid, sulfuric acid-hydrogen peroxide solution, fluoronitric acid (a mixture of hydrofluoric acid and nitric acid), or hydrochloric acid.

[0119] The second nozzle 34 supplies the second processing liquid to the substrate W from above. Specifically, the second nozzle 34 sprays the second processing liquid onto the rotating substrate W. Similar to the first nozzle moving part 33a, the second nozzle moving part 34a moves the second nozzle 34 between a processing position and a retracted position. Specifically, similar to the first nozzle moving part 33a, the second nozzle moving part 34a has a second nozzle arm 341 and a second nozzle drive part 342. The structures of the second nozzle arm 341 and the second nozzle drive part 342 are the same as those of the first nozzle arm 331 and the first nozzle drive part 332, therefore, descriptions are omitted.

[0120] The second processing fluid supply pipe 343 supplies the second processing fluid to the second nozzle 34. The second processing fluid supply pipe 343 is a tubular component for the flow of the second processing fluid. In this embodiment, the second processing fluid is an alkaline liquid. For example, the second processing fluid is ammonia peroxide water (SC1), ammonia water, ammonium fluoride solution, or tetramethylammonium hydroxide (TMAH).

[0121] The third nozzle 35 supplies the third processing liquid to the substrate W from above. Specifically, the third nozzle 35 sprays the third processing liquid onto the rotating substrate W. Similar to the first nozzle moving part 33a, the third nozzle moving part 35a moves the third nozzle 35 between a processing position and a retracted position. Specifically, similar to the first nozzle moving part 33a, the third nozzle moving part 35a has a third nozzle arm 351 and a third nozzle drive part 352. The structures of the third nozzle arm 351 and the third nozzle drive part 352 are the same as those of the first nozzle arm 331 and the first nozzle drive part 332, therefore, descriptions are omitted.

[0122] The third processing fluid supply pipe 353 supplies the third processing fluid to the third nozzle 35. The third processing fluid supply pipe 353 is a tubular component for the flow of the third processing fluid. In this embodiment, the third processing fluid is an organic solvent. For example, the third processing fluid is isopropanol (IPA), methanol, ethanol, hydrofluoroether (HFE), or acetone.

[0123] The fourth nozzle 36 supplies rinsing fluid to the substrate W from above. Specifically, the fourth nozzle 36 sprays rinsing fluid onto the rotating substrate W. Similar to the first nozzle moving part 33a, the fourth nozzle moving part 36a moves the fourth nozzle 36 between a processing position and a retracted position. Specifically, similar to the first nozzle moving part 33a, the fourth nozzle moving part 36a has a fourth nozzle arm 361 and a fourth nozzle drive part 362. The structures of the fourth nozzle arm 361 and the fourth nozzle drive part 362 are the same as those of the first nozzle arm 331 and the first nozzle drive part 332, therefore, descriptions are omitted.

[0124] The flushing fluid supply pipe 363 supplies flushing fluid to the fourth nozzle 36. The flushing fluid supply pipe 363 is a tubular component for the flow of flushing fluid. For example, the flushing fluid is pure water (deionized water), carbonated water, electrolyzed water, hydrogen water, ozone water, or diluted hydrochloric acid water.

[0125] The cup portion 37 is disposed around the substrate holding portion 31. The cup portion 37 surrounds the side of the substrate W held by the substrate holding portion 31. The cup portion 37 receives the first to third processing liquids and rinsing liquid that are scattered from the rotating substrate W.

[0126] Air supply piping 39 connects air supply pipe 44 to blowing unit 38. Gas (e.g., clean air) flowing in air supply pipe 44 is supplied to blowing unit 38 via air supply piping 39. Blowing unit 38 supplies the gas supplied from air supply pipe 44 into the interior of processing housing 2. Blowing unit 38 is mounted on upper plate 21 of processing housing 2. Blowing unit 38 blows gas downwards.

[0127] Next, refer to Figure 8 The processing unit 1, the switching mechanism 51, and the first separate exhaust pipe 41 to the third separate exhaust pipe 43 will be described. Figure 8 This is a top view showing the processing unit 1, the switching mechanism 51, and the first individual exhaust pipe 41 to the third individual exhaust pipe 43.

[0128] like Figure 8 As shown, the switching mechanism 51 has a switching housing 52. The switching housing 52 has an internal space 54. The internal space 54 of the switching housing 52 is a flow space for gas flow. Hereinafter, the internal space 54 will sometimes be referred to as "flow space 54".

[0129] The switching housing 52 has an inlet opening 53. The inlet opening 53 communicates with the flow space 54. The inlet opening 53 of the switching housing 52 communicates with the internal space of the processing unit 1 (processing housing 2). Therefore, gas flows from the processing unit 1 (processing housing 2) into the flow space 54. The gas flowing from the processing unit 1 (processing housing 2) into the flow space 54 flows within the flow space 54.

[0130] The switching housing 52 also has a first outlet opening 55 to a third outlet opening 57. The first outlet opening 55 to the third outlet opening 57 are connected to the flow space 54.

[0131] The internal space of the first separate exhaust pipe 41 is connected to the flow space 54 via the first outlet opening 55. Specifically, the first separate exhaust pipe 41 has an opening 41k opposite to the first outlet opening 55.

[0132] Similarly, the internal space of the second separate exhaust pipe 42 is connected to the flow space 54 via the second outlet opening 56. Specifically, the second separate exhaust pipe 42 has an opening 42k opposite to the second outlet opening 56.

[0133] Additionally, the internal space of the third separate exhaust pipe 43 is connected to the flow space 54 via the third outlet opening 57. Specifically, the third separate exhaust pipe 43 has an opening 43k opposite to the third outlet opening 57.

[0134] like Figure 8 As shown, the switching mechanism 51 also has a first opening / closing part 61 to a third opening / closing part 63. The first opening / closing part 61 to the third opening / closing part 63 are disposed in the flow space 54.

[0135] The first opening / closing part 61 to the third opening / closing part 63 correspond to the first outlet opening 55 to the third outlet opening 57, respectively. In other words, the first opening / closing part 61 to the third opening / closing part 63 correspond to the opening 41k of the first individual exhaust pipe 41 to the opening 43k of the third individual exhaust pipe 43, respectively.

[0136] The first opening / closing part 61 is movable between an open position and a closed position. The open position is the position where the first opening / closing part 61 opens the first outlet opening 55. The closed position is the position where the first opening / closing part 61 closes the first outlet opening 55. In this embodiment, the first opening / closing part 61 is oscillating about a rotation axis A1 extending in the vertical direction. Similarly, the second opening / closing part 62 and the third opening / closing part 63 are also movable between the open position and the closed position.

[0137] The switching mechanism 51 allows gas flowing from the internal space of the processing unit 1 (processing housing 2) into the circulation space 54 to flow into any one of the first individual exhaust pipes 41 to the third individual exhaust pipe 43. For example, when gas flows into the first individual exhaust pipe 41, the first opening / closing part 61 opens the first outlet opening 55, and the second opening / closing part 62 and the third opening / closing part 63 close the second outlet opening 56 and the third outlet opening 57. As a result, of the openings 41k of the first individual exhaust pipe 41 to 43k of the third individual exhaust pipe 43, only the opening 41k of the first individual exhaust pipe 41 communicates with the circulation space 54. Therefore, the gas flowing from the processing unit 1 (processing housing 2) into the circulation space 54 flows into the first individual exhaust pipe 41.

[0138] Next, refer to Figure 8 Further explanation of the substrate processing apparatus 100. For example... Figure 8 As shown, the substrate processing apparatus 100 also includes a first pressure sensor 71 and a first baffle 73. The first baffle 73 is an example of a first pressure regulating mechanism.

[0139] The first pressure sensor 71 measures the pressure of the gas flowing from the processing unit 1 (processing housing 2) into the first separate exhaust pipe 41 to the third separate exhaust pipe 43. In other words, it measures the exhaust pressure of the processing unit 1. For example, the first pressure sensor 71 is disposed at or near the inlet opening 53 of the switching housing 52 to measure the pressure of the gas flowing into the inlet opening 53 of the switching housing 52.

[0140] The first baffle 73 regulates the pressure of the gas flowing from the processing unit 1 (processing housing 2) into the first separate exhaust pipe 41 to the third separate exhaust pipe 43. In other words, it regulates the exhaust pressure of the processing unit 1. Specifically, the exhaust pressure of the processing unit 1 is regulated by adjusting the opening of the first baffle 73. In this embodiment, the first baffle 73 is provided in the switching mechanism 51. For example, the first baffle 73 is disposed at or near the inlet opening 53 of the switching housing 52 to regulate the pressure of the gas flowing into the inlet opening 53 of the switching housing 52.

[0141] The opening degree of the first baffle 73 is controlled by the control device 101. Specifically, the control device 101 adjusts the opening degree of the first baffle 73 based on the value measured by the first pressure sensor 71. More specifically, the opening degree of the first baffle 73 is adjusted so that the value measured by the first pressure sensor 71 becomes a predetermined value.

[0142] Next, refer to Figure 9 Further explanation of switching mechanism 51. Figure 9 This is the main view of the switching mechanism 51. (Example) Figure 9 As shown, the substrate processing apparatus 100 also includes a first opening / closing drive unit 64 to a third opening / closing drive unit 66.

[0143] The first opening / closing drive unit 64 moves the first opening / closing part 61 between an open position and a closed position. In this embodiment, the first opening / closing drive unit 64 causes the first opening / closing part 61 to swing about a rotation axis A1. The first opening / closing drive unit 64 is controlled by the control device 101. The first opening / closing drive unit 64 includes an electric motor capable of rotating in both directions.

[0144] The second opening / closing drive unit 65 moves the second opening / closing part 62 between an open position and a closed position. In this embodiment, the second opening / closing drive unit 65 causes the second opening / closing part 62 to oscillate about a rotation axis A2 extending in the vertical direction. The second opening / closing drive unit 65 is controlled by the control device 101. The second opening / closing drive unit 65 includes an electric motor capable of rotating in both directions.

[0145] The third opening / closing drive unit 66 moves the third opening / closing part 63 between an open position and a closed position. In this embodiment, the third opening / closing drive unit 66 causes the third opening / closing part 63 to oscillate about a rotation axis A3 extending in the vertical direction. The third opening / closing drive unit 66 is controlled by the control device 101. The third opening / closing drive unit 66 includes an electric motor capable of rotating in both directions.

[0146] Next, refer to Figure 10 Further explanation of the substrate processing apparatus 100. Figure 10 This is a system diagram of the exhaust paths originating from multiple processing units 1 (multiple processing housings 2). For example... Figure 10 As shown, the substrate processing apparatus 100 includes first pressure sensors 71A1-71A6, 71B1-71B6, 71C1-71C6, 71D1-71D6 and first baffles 73A1-73A6, 73B1-73B6, 73C1-73C6, 73D1-73D6. The first baffles 73A1-73A6, 73B1-73B6, 73C1-73C6, and 73D1-73D6 are an example of a first pressure regulating mechanism.

[0147] The first pressure sensors 71A1~71A6, 71B1~71B6, 71C1~71C6, and 71D1~71D6 correspond to processing units 1A1~1A6, 1B1~1B6, 1C1~1C6, and 1D1~1D6, respectively. Similarly, the first baffles 73A1~73A6, 73B1~73B6, 73C1~73C6, and 73D1~73D6 correspond to processing units 1A1~1A6, 1B1~1B6, 1C1~1C6, and 1D1~1D6, respectively.

[0148] Reference Figure 8The first pressure sensor 71 described includes first pressure sensors 71A1-71A6, 71B1-71B6, 71C1-71C6, and 71D1-71D6. The structure and reference of the first pressure sensors 71A1-71A6, 71B1-71B6, 71C1-71C6, and 71D1-71D6 are as follows. Figure 8 The first pressure sensor 71 described herein is the same, therefore its description is omitted. Hereinafter, when there is no need to distinguish between the first pressure sensors 71A1~71A6, 71B1~71B6, 71C1~71C6, and 71D1~71D6 in the description, the first pressure sensors 71A1~71A6, 71B1~71B6, 71C1~71C6, and 71D1~71D6 may sometimes be referred to as "first pressure sensor 71".

[0149] Reference Figure 8 The first baffle 73 described includes first baffles 73A1~73A6, 73B1~73B6, 73C1~73C6, and 73D1~73D6. The structure and reference of the first baffles 73A1~73A6, 73B1~73B6, 73C1~73C6, and 73D1~73D6 are as follows. Figure 8 The first baffle 73 described herein is the same, therefore its description is omitted. Hereinafter, when there is no need to distinguish between the first baffles 73A1~73A6, 73B1~73B6, 73C1~73C6, and 73D1~73D6 in the description, the first baffles 73A1~73A6, 73B1~73B6, 73C1~73C6, and 73D1~73D6 may sometimes be referred to as "first baffle 73".

[0150] like Figure 10 As shown, the substrate processing apparatus 100 includes common exhaust pipes 81A, 82A, 83A, 81B, 82B, and 83B.

[0151] The first individual exhaust pipes 41A and 41B are connected to the common exhaust pipe 81A. Therefore, the gas discharged from the multiple processing units 1A to the first individual exhaust pipe 41A flows into the common exhaust pipe 81A via the first individual exhaust pipe 41A. In addition, the gas discharged from the multiple processing units 1B to the first individual exhaust pipe 41B flows into the common exhaust pipe 81A via the first individual exhaust pipe 41B.

[0152] The second separate exhaust pipes 42A and 42B are connected to the common exhaust pipe 82A. Therefore, the gas discharged from the multiple processing units 1A to the second separate exhaust pipe 42A flows into the common exhaust pipe 82A via the second separate exhaust pipe 42A. In addition, the gas discharged from the multiple processing units 1B to the second separate exhaust pipe 42B flows into the common exhaust pipe 82A via the second separate exhaust pipe 42B.

[0153] The third separate exhaust pipes 43A and 43B are connected to the common exhaust pipe 83A. Therefore, the gas discharged from the multiple processing units 1A to the third separate exhaust pipe 43A flows into the common exhaust pipe 83A via the third separate exhaust pipe 43A. In addition, the gas discharged from the multiple processing units 1B to the third separate exhaust pipe 43B flows into the common exhaust pipe 83A via the third separate exhaust pipe 43B.

[0154] The first individual exhaust pipes 41C and 41D are connected to the common exhaust pipe 81B. Therefore, the gas discharged from the multiple processing units 1C to the first individual exhaust pipe 41C flows into the common exhaust pipe 81B via the first individual exhaust pipe 41C. In addition, the gas discharged from the multiple processing units 1D to the first individual exhaust pipe 41D flows into the common exhaust pipe 81B via the first individual exhaust pipe 41D.

[0155] The second separate exhaust pipes 42C and 42D are connected to the common exhaust pipe 82B. Therefore, the gas discharged from the multiple processing units 1C to the second separate exhaust pipe 42C flows into the common exhaust pipe 82B via the second separate exhaust pipe 42C. In addition, the gas discharged from the multiple processing units 1D to the second separate exhaust pipe 42D flows into the common exhaust pipe 82B via the second separate exhaust pipe 42D.

[0156] The third separate exhaust pipes 43C and 43D are connected to the common exhaust pipe 83B. Therefore, the gas discharged from the multiple processing units 1C to the third separate exhaust pipe 43C flows into the common exhaust pipe 83B via the third separate exhaust pipe 43C. In addition, the gas discharged from the multiple processing units 1D to the third separate exhaust pipe 43D flows into the common exhaust pipe 83B via the third separate exhaust pipe 43D.

[0157] In the following description, when there is no need to distinguish between the common exhaust pipes 81A and 81B, the common exhaust pipes 81A and 81B will sometimes be referred to as "common exhaust pipe 81". When there is no need to distinguish between the common exhaust pipes 82A and 82B, the common exhaust pipes 82A and 82B will sometimes be referred to as "common exhaust pipe 82". When there is no need to distinguish between the common exhaust pipes 83A and 83B, the common exhaust pipes 83A and 83B will sometimes be referred to as "common exhaust pipe 83".

[0158] like Figure 10 As shown, the substrate processing apparatus 100 includes second pressure sensors 84A, 85A, 86A, 84B, 85B, 86B and second baffles 87A, 88A, 89A, 87B, 88B, 89B. The second baffles 87A, 88A, 89A, 87B, 88B, 89B are an example of a second pressure regulating mechanism.

[0159] The second pressure sensors 84A, 85A, 86A, 84B, 85B, and 86B correspond to the common exhaust pipes 81A, 82A, 83A, 81B, 82B, and 83B, respectively. Similarly, the second baffles 87A, 88A, 89A, 87B, 88B, and 89B correspond to the common exhaust pipes 81A, 82A, 83A, 81B, 82B, and 83B, respectively.

[0160] The second pressure sensor 84A measures the pressure of the gas inside the common exhaust pipe 81A. In other words, it measures the exhaust pressure of the common exhaust pipe 81A. Similarly, the second pressure sensors 85A, 86A, 84B, 85B, and 86B measure the pressure (exhaust pressure) of the gas inside the common exhaust pipes 82A, 83A, 81B, 82B, and 83B, respectively.

[0161] The second baffle 87A regulates the pressure of the gas inside the common exhaust pipe 81A. Specifically, the pressure of the gas inside the common exhaust pipe 81A is adjusted by regulating the opening degree of the second baffle 87A. For example, the second baffle 87A is positioned near the inlet of the common exhaust pipe 81A. Similarly, the second baffles 88A, 89A, 87B, 88B, and 89B regulate the pressure of the gas inside the common exhaust pipes 82A, 83A, 81B, 82B, and 83B, respectively.

[0162] The opening degree of the second baffle 87A is controlled by the control device 101. Specifically, the control device 101 adjusts the opening degree of the second baffle 87A based on the value measured by the second pressure sensor 84A. More specifically, the opening degree of the second baffle 87A is adjusted so that the value measured by the second pressure sensor 84A becomes a predetermined value. Similarly, the control device 101 controls the opening degree of each of the second baffles 88A, 89A, 87B, 88B, and 89B based on the values ​​measured by the second pressure sensors 85A, 86A, 84B, 85B, and 86B.

[0163] Hereinafter, when it is not necessary to distinguish between the second pressure sensors 84A and 84B in the description, the second pressure sensors 84A and 84B will sometimes be referred to as "second pressure sensor 84". Similarly, when it is not necessary to distinguish between the second pressure sensors 85A and 85B in the description, the second pressure sensors 85A and 85B will sometimes be referred to as "second pressure sensor 85". When it is not necessary to distinguish between the second pressure sensors 86A and 86B in the description, the second pressure sensors 86A and 86B will sometimes be referred to as "second pressure sensor 86".

[0164] Furthermore, when it is not necessary to distinguish between the second baffles 87A and 87B in the description, the second baffles 87A and 87B are sometimes referred to as "second baffle 87". Similarly, when it is not necessary to distinguish between the second baffles 88A and 88B in the description, the second baffles 88A and 88B are sometimes referred to as "second baffle 88". When it is not necessary to distinguish between the second baffles 89A and 89B in the description, the second baffles 89A and 89B are sometimes referred to as "second baffle 89".

[0165] Next, refer to Figure 11 Further explanation of the shared exhaust pipes 81-83. Figure 11 This is a top view of the substrate processing apparatus 100.

[0166] The common exhaust pipes 81A, 82A, and 83A have front ends 81Af, 82Af, and 83Af, respectively. The common exhaust pipes 81B, 82B, and 83B have front ends 81Bf, 82Bf, and 83Bf, respectively.

[0167] The front ends 81Af, 82Af, 83Af, 81Bf, 82Bf, and 83Bf are respectively equivalent to the upstream ends of the common exhaust pipes 81A, 82A, 83A, 81B, 82B, and 83B.

[0168] In the following explanations, when there is no need to distinguish between front-end 81Af and 81Bf, front-end 81Af and 81Bf will sometimes be referred to as "front-end 81f". Similarly, when there is no need to distinguish between front-end 82Af and 82Bf, front-end 82Af and 82Bf will sometimes be referred to as "front-end 82f". When there is no need to distinguish between front-end 83Af and 83Bf, front-end 83Af and 83Bf will sometimes be referred to as "front-end 83f".

[0169] The downstream end of the common exhaust pipe 81 is connected to a first factory piping. The first factory piping is depressurized to a substantially constant pressure. Specifically, the downstream end of the common exhaust pipe 81 is connected to a first exhaust treatment device via the first factory piping. The first exhaust treatment device recovers gas from the common exhaust pipe 81. The first exhaust treatment device is, for example, installed in a factory equipped with a substrate processing apparatus 100. Similarly, the downstream end of the common exhaust pipe 82 is connected to a second exhaust treatment device via a second factory piping. The downstream end of the common exhaust pipe 83 is connected to a third exhaust treatment device via a third factory piping.

[0170] like Figure 11As shown, the first separate exhaust pipe 41B is connected to the common exhaust pipe 81A on a downstream side compared to the first separate exhaust pipe 41A. The second pressure sensor 84A is located downstream of the connection point between the first separate exhaust pipe 41B and the common exhaust pipe 81A. The second separate exhaust pipe 42B is connected to the common exhaust pipe 82A on a downstream side compared to the second separate exhaust pipe 42A. The second pressure sensor 85A is located downstream of the connection point between the second separate exhaust pipe 42B and the common exhaust pipe 82A. The third separate exhaust pipe 43B is connected to the common exhaust pipe 83A on a downstream side compared to the third separate exhaust pipe 43A. The second pressure sensor 86A is located downstream of the connection point between the third separate exhaust pipe 43B and the common exhaust pipe 83A.

[0171] Additionally, the first separate exhaust pipe 41D is connected to the common exhaust pipe 81B on a downstream side compared to the first separate exhaust pipe 41C. The second pressure sensor 84B is located downstream of the connection point between the first separate exhaust pipe 41D and the common exhaust pipe 81B. The second separate exhaust pipe 42D is connected to the common exhaust pipe 82B on a downstream side compared to the second separate exhaust pipe 42C. The second pressure sensor 85B is located downstream of the connection point between the second separate exhaust pipe 42D and the common exhaust pipe 82B. The third separate exhaust pipe 43D is connected to the common exhaust pipe 83B on a downstream side compared to the third separate exhaust pipe 43C. The second pressure sensor 86B is located downstream of the connection point between the third separate exhaust pipe 43D and the common exhaust pipe 83B.

[0172] Front ends 81f to 83f are open to the outside of the substrate processing apparatus 100. For example, front ends 81f to 83f are open to a cleanroom where the substrate processing apparatus 100 is installed. External gas from the substrate processing apparatus 100 flows into common exhaust pipes 81 to 83f from front ends 81f to 83f. The gas flowing into common exhaust pipes 81 to 83f from front ends 81f to 83f circulates in common exhaust pipes 81 to 83. In addition, it may sometimes flow into any one of the first separate exhaust pipe 41, the second separate exhaust pipe 42, and the third separate exhaust pipe 43.

[0173] The second baffle 87A is positioned upstream of the connection point between the first separate exhaust pipe 41A and the common exhaust pipe 81A. For example, the second baffle 87A is positioned near the front end 81Af. Similarly, the second baffle 88A is positioned upstream of the connection point between the second separate exhaust pipe 42A and the common exhaust pipe 82A. The second baffle 89A is positioned upstream of the connection point between the third separate exhaust pipe 43A and the common exhaust pipe 83A.

[0174] The second baffle 87B is positioned upstream of the connection point between the first separate exhaust pipe 41C and the common exhaust pipe 81B. For example, the second baffle 87B is positioned near the front end 81Bf. Similarly, the second baffle 88B is positioned upstream of the connection point between the second separate exhaust pipe 42C and the common exhaust pipe 82B. The second baffle 89B is positioned upstream of the connection point between the third separate exhaust pipe 43C and the common exhaust pipe 83B.

[0175] The pressure inside the common exhaust pipe 81A varies due to changes in the gas flow rate from the first separate exhaust pipe 41A to the common exhaust pipe 81A and the gas flow rate from the first separate exhaust pipe 41B to the common exhaust pipe 81A. Based on the value measured by the second pressure sensor 84A, the control device 101 controls the opening of the second baffle 87A to bring the pressure inside the common exhaust pipe 81A to a predetermined value. Similarly, the control device 101 adjusts the opening of the second baffles 88A, 89A, 87B, 88B, and 89B.

[0176] Above, refer to Figures 2 to 11 The structure of the substrate processing apparatus 100 has been described. Additionally, refer to... Figure 1 The substrate processing apparatus 100L described herein has the same features as the reference. Figures 2 to 11 The structure is substantially the same as that of the substrate processing apparatus 100, so its description is omitted.

[0177] Next, refer to Figures 1-12 Further explanation of the substrate processing apparatus 100. Figure 12 This is a block diagram representing the substrate processing apparatus 100. (Example) Figure 12 As shown, the control device 101 includes a control unit 102 and a storage unit 103. In addition, the board processing device 100 also includes a display unit 104, an input unit 105, and a communication unit 106.

[0178] The control unit 102 has a processor. The control unit 102 may have, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Alternatively, the control unit 102 may also have a general-purpose arithmetic logic unit (ALU).

[0179] Storage unit 103 stores control program PG1 and data. Control program PG1 is a computer program. Data includes process data RP. Process data RP contains information representing multiple processes. Each process specifies the processing content and processing order of the substrate W.

[0180] Storage unit 103 has a main storage device. The main storage device is, for example, a semiconductor memory. Storage unit 103 may also have an auxiliary storage device. The auxiliary storage device includes, for example, at least one of a semiconductor memory and a hard disk drive. Storage unit 103 may also include a removable medium. Control unit 102 controls the operation of each part of the board processing apparatus 100 based on control program PG1 and data stored in storage unit 103.

[0181] In this embodiment, the storage unit 103 also stores position information MPD, distance information MLD, learned model LM, and baffle opening adjustment information PA. Position information MPD and distance information MLD are referenced... Figure 1 This is part of the device information MD described.

[0182] The position information MPD indicates the position of each of the multiple processing units 1 in the substrate processing apparatus 100. The position information MPD can also indicate the coordinates of each of the multiple processing units 1 in the substrate processing apparatus 100. The distance information MLD indicates the distance of each exhaust path in the substrate processing apparatus 100. Each exhaust path distance includes the distance from which the gas discharged from processing units 1A and 1B is discharged from the downstream end of the common exhaust pipes 81A to 83A, and the distance from which the gas discharged from processing units 1C and 1D is discharged from the downstream end of the common exhaust pipes 81B to 83B.

[0183] In processing units 1A and 1B, the exhaust path distance varies for each processing unit 1. Similarly, in processing units 1C and 1D, the exhaust path distance varies for each processing unit 1. The exhaust path distance is a factor that affects the pressure (exhaust pressure) inside the common exhaust pipes 81-83. In addition, the individual positions of the multiple processing units 1 are also factors that affect the pressure (exhaust pressure) inside the common exhaust pipes 81-83.

[0184] For reference Figure 1 As explained, the learned model LM outputs baffle opening adjustment information PA based on the device information MD. Specifically, when the control unit 102 inputs the device information MD into the learned model LM, it outputs the baffle opening adjustment information PA from the learned model LM.

[0185] The baffle opening adjustment information PA is an example of pressure adjustment information during processing. The baffle opening adjustment information PA specifies the operation of the first baffle 73 and the second baffles 87, 88, and 89. Specifically, the baffle opening adjustment information PA is time-series data representing the opening degree of the first baffle 73 and the opening degrees of the second baffles 87, 88, and 89. When processing the target substrate Wp, the control unit 102 controls the operation of the first baffle 73 (the opening degree of the first baffle 73) and the operation of the second baffles 87, 88, and 89 (the opening degrees of the second baffles 87, 88, and 89) based on the baffle opening adjustment information PA.

[0186] Display unit 104 displays various screens and information. Display unit 104 may be, for example, a liquid crystal display (LCD) or an organic EL display. Display unit 104 may also display, for example, a setting screen for process data (RP), a setting screen for position information (MPD), and a setting screen for distance information (MLD).

[0187] The input unit 105 accepts input from the operator and outputs various information to the control unit 102. For example, the input unit 105 may also accept process data RP, position information MPD, and distance information MLD. The input unit 105 may include, for example, a touch panel and an indicator device. The touch panel may be disposed, for example, on the display surface of the display unit 104. The input unit 105 and the display unit 104 may constitute a graphical user interface.

[0188] The communication unit 106 is connected to a network and communicates with external devices. Networks include, for example, the Internet, a LAN (Local Area Network), a public telephone network, and a short-range wireless network. The communication unit 106 is a communication device, such as a network interface controller. In this embodiment, the communication unit 106 performs communication with the learning device 500. The communication unit 106 receives the learned model LM from the learning device 500. The control unit 102 stores the learned model LM received by the communication unit 106 in the storage unit 103.

[0189] Next, refer to Figure 12 and Figure 13 The process of obtaining the baffle opening adjustment information PA is explained. Figure 13 This is a flowchart illustrating the process of obtaining baffle opening adjustment information PA. The process of obtaining baffle opening adjustment information PA is executed by the control unit 102. The process begins upon receiving a processing start instruction from the operator regarding the substrate Wp to be processed, via the input unit 105. Figure 13 The processing shown.

[0190] When the input unit 105 receives an instruction from the operator to begin processing the target substrate Wp, the control unit 102 acquires device information MD (step S1). Upon acquiring device information MD, the control unit 102 inputs device information MD into the learned model LM (step S2). As a result, baffle opening adjustment information PA is output from the learned model LM. The control unit 102 acquires the baffle opening adjustment information PA output from the learned model LM (step S3). As a result, Figure 13 The processing shown has ended.

[0191] Next, refer to Figure 14 Description of device information (MD). Figure 14 This is a diagram representing device information (MD). For example... Figure 14 As shown, the device information MD includes processing sequence information MPR, number of processed object pieces information MC, location information MPD, and distance information MLD.

[0192] The processing sequence information MPR indicates the order in which the substrate Wp is processed using the processing liquid. The control unit 102 can also obtain the processing sequence information MPR from the process data RP.

[0193] The processing target wafer number information MC represents the unit of the number of wafers processed on the processing target substrate Wp. As already explained, the baffle opening adjustment information PA is time-series data, and its content varies depending on the number of wafers processed on the processing target substrate Wp. The number of wafers processed on the processing target substrate Wp is expressed, for example, in batches. For example, the processing target wafer number information MC represents 1 batch. When the processing target wafer number information MC represents 1 batch, the baffle opening adjustment information PA represents, in time sequence, the opening of the first baffle 73 and the openings of the second baffles 87, 88, and 89 when processing multiple processing target substrates Wp constituting 1 batch. The control unit 102 can also obtain the processing target wafer number information MC from the process data RP.

[0194] For reference Figure 12 As explained, the location information MPD and distance information MLD are stored in the storage unit 103. The control unit 102 retrieves the location information MPD and distance information MLD from the storage unit 103.

[0195] Next, refer to Figure 15 Explain the sequence information MPR during processing. Figure 15 This is a diagram representing an example of the processing sequence information MPR. Figure 15 This is an example illustrating the order in which the substrate Wp is processed. For example... Figure 15 As shown, the substrate Wp to be processed can also be treated with a first processing solution and a second processing solution.

[0196] Next, refer to Figure 16 An example of the operation of the substrate processing apparatus 100 will be described. Figure 16 This diagram illustrates an example of the operation of the substrate processing apparatus 100. In detail, Figure 16 This describes the operation of the substrate processing apparatus 100 when processing one substrate Wp. Furthermore, before processing the substrate Wp, the operation is performed by referring to... Figure 13 The description states that the baffle opening adjustment information PA is stored in the storage unit 103.

[0197] like Figure 16 As shown, in step S11, the substrate Wp to be processed is moved into the processing unit 1. The moved substrate Wp is held by the substrate holding part 31. Specifically, the substrate Wp to be processed is moved into the processing unit 1 by the conveying mechanism 153.

[0198] In step S12, a first processing liquid is supplied to the substrate Wp to be processed. Specifically, the first processing liquid is sprayed from the first nozzle 33 onto the upper surface of the substrate Wp. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 81 via the first separate exhaust pipe 41. In addition, the control unit 102 adjusts the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 with reference to the baffle opening adjustment information PA.

[0199] After the processing based on the first processing liquid is completed, in step S13, rinsing liquid is supplied to the substrate Wp to be processed. Specifically, rinsing liquid is sprayed onto the upper surface of the substrate Wp from the fourth nozzle 36. As a result, the first processing liquid is removed from the substrate Wp. In detail, the liquid film of the first processing liquid is replaced by the liquid film of the rinsing liquid. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 81 via the first separate exhaust pipe 41. In addition, the control unit 102 adjusts the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 with reference to the baffle opening degree adjustment information PA.

[0200] After rinsing, in step S14, a second processing liquid is supplied to the substrate Wp to be processed. Specifically, the second processing liquid is sprayed onto the upper surface of the substrate Wp from the second nozzle 34. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 82 via the second separate exhaust pipe 42. In addition, the control unit 102 adjusts the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 with reference to the baffle opening adjustment information PA.

[0201] After the processing based on the second processing liquid is completed, in step S15, rinsing liquid is supplied to the substrate Wp to be processed. As a result, the second processing liquid is removed from the substrate Wp to be processed. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 82 via the second separate exhaust pipe 42. In addition, the control unit 102 adjusts the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 with reference to the baffle opening degree adjustment information PA.

[0202] After rinsing, in step S16, a third processing liquid is supplied to the substrate Wp to be processed. Specifically, the third processing liquid is sprayed from the third nozzle 35 onto the upper surface of the substrate Wp. As a result, the rinsing liquid is removed from the substrate Wp. In detail, the liquid film of the rinsing liquid is replaced by the liquid film of the third processing liquid. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 83 via the third separate exhaust pipe 43. In addition, the control unit 102 adjusts the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 with reference to the baffle opening adjustment information PA.

[0203] After the rinsing fluid is replaced with the third processing fluid, in step S17, the control unit 102 controls the rotation drive unit 32 to dry the substrate Wp to be processed. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 83 via the third separate exhaust pipe 43. In addition, the control unit 102 adjusts the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 with reference to the baffle opening adjustment information PA.

[0204] Specifically, the control unit 102 increases the rotational speed of the substrate Wp to be processed, thereby applying a greater centrifugal force to the third processing liquid on the substrate Wp. As a result, the third processing liquid adhering to the substrate Wp is thrown to the surrounding area of ​​the substrate Wp. In this way, the third processing liquid is removed from the substrate Wp, and the substrate Wp is dried.

[0205] In addition, the control unit 102, for example, stops the rotation of the target substrate Wp based on the rotation drive unit 32 after a predetermined time has elapsed since the high-speed rotation of the target substrate Wp began to be processed.

[0206] After the drying process of the substrate Wp is completed, in step S18, the substrate Wp is removed from the processing unit 1, and the process ends. Figure 16 The process is shown. Specifically, the substrate Wp to be processed is removed from the processing unit 1 by the transfer mechanism 153. Furthermore, this process is repeated. Figure 16 The processing continues until the number of substrates Wp specified by the processing target number information MC has been processed.

[0207] Above, refer to Figures 1 to 16The substrate processing apparatus 100 of this embodiment has been described. According to this embodiment, the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 can be adjusted by referring to the baffle opening degree adjustment information PA. Therefore, compared with the case where the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 are adjusted only based on the measurement value of the first pressure sensor 71 and the measurement value of the second pressure sensors 84, 85, and 86, the exhaust pressure of each processing unit 1 can be adjusted more appropriately.

[0208] In detail, when the opening of the first baffle 73 is adjusted solely based on the measurement value of the first pressure sensor 71, a slight delay occurs between detecting a change in the measurement value of the first pressure sensor 71 and actually adjusting the opening of the first baffle 73. This results in a period of inappropriate exhaust pressure for the processing unit 1. In contrast, according to this embodiment, the opening of the first baffle 73 can be adjusted with reference to the baffle opening adjustment information PA. Therefore, the opening of the first baffle 73 can be adjusted without relying on the measurement value of the first pressure sensor 71. As a result, the opening of the first baffle 73 can be adjusted earlier in timing than when adjusting the opening of the first baffle 73 after detecting a change in the measurement value of the first pressure sensor 71. Therefore, the exhaust pressure of each processing unit 1 can be adjusted more appropriately.

[0209] The same applies when adjusting the opening of the second baffles 87, 88, and 89 solely based on the measurements from the second pressure sensors 84, 85, and 86. A slight delay occurs between detecting a change in the measured values ​​of the second pressure sensors 84, 85, and 86 and the actual adjustment of the opening of the second baffles 87, 88, and 89. This results in a period of inappropriate pressure (exhaust pressure) within the common exhaust pipes 81-83. Since the exhaust pressure of each processing unit 1 is affected by the pressure within the common exhaust pipes 81-83, this leads to a period of inappropriate pressure within the common exhaust pipes 81-83, and consequently, a period of inappropriate exhaust pressure for each processing unit 1. In contrast, according to this embodiment, the opening of the second baffles 87, 88, and 89 can be adjusted with reference to the baffle opening adjustment information PA. Therefore, the opening of the second baffles 87, 88, and 89 can be adjusted earlier than when the opening of the second baffles 87, 88, and 89 is adjusted after detecting a change in the measured values ​​of the second pressure sensors 84, 85, and 86. Therefore, the exhaust pressure of each processing unit 1 can be adjusted more appropriately.

[0210] Next, refer to Figure 17 The substrate processing apparatus 100L will be described. Figure 17 This is a block diagram representing the substrate processing apparatus 100L. (Example) Figure 17 As shown, compared with the reference Figure 12Compared to the substrate processing apparatus 100 described herein, the substrate processing apparatus 100L further includes a plurality of first baffle opening degree sensors 107, a plurality of second baffle opening degree sensors 108, a plurality of temperature sensors 109, a plurality of first concentration sensors 110, a plurality of second concentration sensors 111, a plurality of third concentration sensors 112, and a plurality of airflow sensors 113.

[0211] Multiple first baffle opening sensors 107 measure the opening degree of multiple first baffles 73. Based on the measurement values ​​of the multiple first baffle opening sensors 107, information (pressure adjustment information during learning) representing the action of each first baffle 73 during the processing of the learning target substrate WL is generated.

[0212] Multiple second baffle opening sensors 108 measure the opening degree of second baffles 87, 88, and 89. Based on the measurement values ​​of the multiple second baffle opening sensors 108, information (pressure adjustment information during learning) representing the action of each second baffle 87, 88, and 89 during the processing of the learning target substrate WL is generated.

[0213] A temperature sensor 109 is provided for each processing unit 1. The temperature sensor 109 measures the temperature inside the processing housing 2. Based on the measurement value of the temperature sensor 109, temperature information representing the temperature inside the processing unit 1 is generated.

[0214] A first concentration sensor 110 is provided for each processing unit 1. The first concentration sensor 110 is configured in reference... Figure 7 The first processing liquid supply piping 333 is described, and the concentration of the first processing liquid is measured. Based on the measurement value of the first concentration sensor 110, concentration information representing the concentration of the first processing liquid is generated.

[0215] A second concentration sensor 111 is provided for each processing unit 1. The second concentration sensor 111 is configured in reference... Figure 7 The second processing liquid supply piping 343 is described, and the concentration of the second processing liquid is measured. Based on the measurement value of the second concentration sensor 111, concentration information representing the concentration of the second processing liquid is generated.

[0216] A third concentration sensor 112 is provided for each processing unit 1. The third concentration sensor 112 is configured in reference... Figure 7 The third processing liquid supply piping 353 is described, and the concentration of the third processing liquid is measured. Based on the measurement value of the third concentration sensor 112, concentration information representing the concentration of the third processing liquid is generated.

[0217] An airflow sensor 113 is provided for each processing unit 1. The airflow sensor 113 is configured in reference... Figure 7The gas supply pipe 39 is described, and the flow rate of the gas flowing in the gas supply pipe 39 is measured. Based on the measurement value of the air volume sensor 113, flow rate information representing the flow rate of the gas blown from the blowing unit 38 to the interior of the processing unit 1 (processing housing 2) is generated.

[0218] In addition, similar to the substrate processing apparatus 100L, the substrate processing apparatus 100 may also include a plurality of first baffle opening degree sensors 107, a plurality of second baffle opening degree sensors 108, a plurality of temperature sensors 109, a plurality of first concentration sensors 110, a plurality of second concentration sensors 111, a plurality of third concentration sensors 112 and a plurality of airflow sensors 113.

[0219] like Figure 17 As shown, in the substrate processing apparatus 100L, the storage unit 103 stores the control program PG2, multiple test process data TRP, multiple time series data TR, position information TPD, and distance information TLD.

[0220] The control program PG2 is a computer program. The control unit 102 controls the operation of each part of the board processing device 100L based on the control program PG2 and data stored in the storage unit 103.

[0221] The Test Process Data (TRP) contains information representing multiple processes. Each process specifies the processing content and order of the learning target substrate (WL). The TRPs for each process contain distinct information.

[0222] The control unit 102 obtains measurement values ​​from multiple first baffle opening degree sensors 107, multiple second baffle opening degree sensors 108, multiple temperature sensors 109, multiple first concentration sensors 110, multiple second concentration sensors 111, multiple third concentration sensors 112, and multiple airflow sensors 113 to generate multiple time series data TR.

[0223] For example, the control unit 102 uses the measured values ​​from the first concentration sensor 110 to generate data (time-series data) representing the change in the concentration of the first processing liquid over time. The time-series data of the first processing liquid concentration is an example of concentration information representing the concentration of the first processing liquid. For example, when processing multiple learning object substrates WL constituting a batch, the control unit 102 arranges the measured values ​​from the first concentration sensor 110 in a time sequence to generate time-series data of the first processing liquid concentration. Time-series data of the first processing liquid concentration is generated for each first concentration sensor 110.

[0224] The multiple time-series data TR contains multiple first-baffle time-series data representing the actions of each of the multiple first baffles 73. Specifically, each first-baffle time-series data represents the change in the opening degree of each first baffle 73 over time. The control unit 102 generates the first-baffle time-series data based on the measurements from the first-baffle opening degree sensor 107. First-baffle time-series data is generated for each first-baffle opening degree sensor 107. The multiple first-baffle time-series data are part of the pressure regulation information learned during training.

[0225] Additionally, the multiple time-series data TR contains multiple second-baffle time-series data representing the actions of each of the second baffles 87, 88, and 89. Specifically, the multiple second-baffle time-series data represent the change in the opening degree of each of the second baffles 87, 88, and 89 along a time sequence. The control unit 102 generates the second-baffle time-series data based on the measurements from the second-baffle opening degree sensor 108. Second-baffle time-series data is generated for each second-baffle opening degree sensor 108. The multiple second-baffle time-series data are part of the pressure regulation information learned during the learning process.

[0226] The position information TPD indicates the position of each of the multiple processing units 1 in the substrate processing apparatus 100L. The position information TPD can represent the coordinates of each of the multiple processing units 1 in the substrate processing apparatus 100L. The distance information TLD indicates the distance of each exhaust path in the substrate processing apparatus 100L.

[0227] Next, refer to Figure 18 Explain the Test Process Data (TRP). Figure 18 This is a graph representing the Test Process Revision (TRP) data. For example... Figure 18 As shown, the test process data TRP includes learning sequence information TPR, number of learning objects information TC, first baffle opening setting information DS1, and second baffle opening setting information DS2.

[0228] The learning sequence information (TPR) indicates the order in which the learning target substrate (WL) is processed by the processing solution, i.e., the processing order during learning. (Compared to the reference...) Figure 14 Similarly, the processing target number information MC, and the learning target number information TC, represent the unit of the number of learning target substrates WL processed. For example, the learning target number information TC represents 1 batch.

[0229] The first baffle opening setting information DS1 represents the setting value of the opening of each first baffle 73. For example, the first baffle opening setting information DS1 may also represent the setting value of the opening of each first baffle 73 that varies along the order in which the learning object substrate WL is treated with the processing liquid.

[0230] The second baffle opening setting information DS2 represents the setting value of the opening of the second baffles 87, 88, and 89. For example, the second baffle opening setting information DS2 can also represent the setting value of the opening of the second baffles 87, 88, and 89, which varies along the order in which the learning object substrate WL is treated with the processing liquid.

[0231] The first baffle opening setting information DS1 and the second baffle opening setting information DS2 represent the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 during the processing of the learning target substrate WL. When processing the learning target substrate WL along the processing sequence during learning, the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 are controlled to the openings set in the first baffle opening setting information DS1 and the second baffle opening setting information DS2.

[0232] As already explained, the learning time sequence information (TPR) contains different content. For example, between learning time sequence information TPRs that have the same learning object piece count information (TC), at least one of the first baffle opening setting information (DS1) and the second baffle opening setting information (DS2) is different.

[0233] Next, refer to Figure 19 This indicates the sequential information TPR during learning. Figure 19 This is a diagram representing an example of the sequential information transfer during learning (TPR). Figure 19 This is an example illustrating the order in which the learning object substrate WL is processed. For example... Figure 19 As shown, the test process data TRP can also represent the sequence of chemical treatment of the learning object substrate WL using the first processing liquid and the second processing liquid.

[0234] Next, refer to Figure 16 An example of the operation of the substrate processing apparatus 100L will be described. First, in step S11, the learning target substrate WL is moved into the processing unit 1. Next, in step S12, a first processing liquid is supplied to the learning target substrate WL. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 81 via the first separate exhaust pipe 41. In addition, the control unit 102 controls the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 based on the first baffle opening degree setting information DS1 and the second baffle opening degree setting information DS2.

[0235] After the processing based on the first processing liquid is completed, in step S13, rinsing liquid is supplied to the learning target substrate WL. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 81 via the first individual exhaust pipe 41. In addition, the control unit 102 controls the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 based on the first baffle opening degree setting information DS1 and the second baffle opening degree setting information DS2.

[0236] After rinsing, in step S14, a second processing liquid is supplied to the learning target substrate WL. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 82 via the second separate exhaust pipe 42. In addition, the control unit 102 controls the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 based on the first baffle opening setting information DS1 and the second baffle opening setting information DS2.

[0237] After the processing based on the second processing liquid is completed, in step S15, rinsing liquid is supplied to the learning target substrate WL. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 82 via the second separate exhaust pipe 42. In addition, the control unit 102 controls the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 based on the first baffle opening degree setting information DS1 and the second baffle opening degree setting information DS2.

[0238] After rinsing, in step S16, a third processing liquid is supplied to the learning substrate WL. As a result, the liquid film of the rinsing liquid is replaced by the liquid film of the third processing liquid. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 83 via the third separate exhaust pipe 43. In addition, the control unit 102 controls the opening degree of the first baffle 73 and the opening degree of the second baffles 87, 88, and 89 based on the first baffle opening degree setting information DS1 and the second baffle opening degree setting information DS2.

[0239] After replacing the rinsing fluid with the third processing fluid, in step S17, the control unit 102 controls the rotation drive unit 32 to dry the learning target substrate WL. At this time, the gas discharged from the processing unit 1 is discharged to the common exhaust pipe 83 via the third separate exhaust pipe 43. In addition, the control unit 102 controls the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 based on the first baffle opening setting information DS1 and the second baffle opening setting information DS2.

[0240] After the drying process of the learning object substrate WL is completed, in step S18, the learning object substrate WL is removed from the processing unit 1, and the process ends. Figure 16 The process is shown. Furthermore, it is repeated. Figure 16 The process shown continues until the number of learning object substrates WL specified by the learning object number information TC has been processed.

[0241] Next, refer to Figure 20 Explanation of the learning data generation device 400. Figure 20 This is a block diagram representing the data generation device 400 for learning. For example... Figure 20 As shown, the learning data generation apparatus 400 includes a control unit 401. The control unit 401 includes a control unit 402 and a storage unit 403.

[0242] The control unit 402 has a processor. The control unit 402 may have a CPU or MPU, for example. Alternatively, the control unit 402 may also have a general-purpose arithmetic logic unit (ALU). The storage unit 403 stores the control program PG3 and data. The control program PG3 is a computer program. The storage unit 403 has a main storage device. The storage unit 403 may also have an auxiliary storage device. The storage unit 403 may also include a removable medium. The control unit 402 performs various processes based on the control program PG3 and data stored in the storage unit 403.

[0243] Specifically, the control unit 402 stores the learning object information LTD obtained from the substrate processing apparatus 100L in the storage unit 403. Additionally, the control unit 402 stores the processing result information RS obtained from the inspection apparatus 300 in the storage unit 403. Then, the learning object information LTD and the processing result information RS are associated to generate learning data LFD. Furthermore, as referred to... Figure 1 As explained, in the control unit 402, substrate information is input along with learning object information LTD. Additionally, in the control unit 402, substrate information is input along with processing result information RS.

[0244] like Figure 20 As shown, the learning object information LTD includes learning time sequence information TPR, learning object slice number information TC, multiple time series data TR, location information TPD, and distance information TLD.

[0245] like Figure 20 As shown, the learning data generation device 400 also includes a display unit 404, an input unit 405, and a communication unit 406.

[0246] Display unit 404 displays various screens and information. Display unit 404 is, for example, a liquid crystal display or an organic EL display. Display unit 404 may also display input screens for inputting various data to control unit 402.

[0247] The input unit 405 accepts input from the operator and outputs various information to the control unit 402. The input unit 405 may include, for example, a touch panel and an indicator device. The input unit 405 may also accept instruction input for generating learning data (LFD).

[0248] The communication unit 406 is connected to a network and communicates with external devices. The communication unit 406 is a communication device, such as a network interface controller. In this embodiment, the communication unit 406 performs communication between the board processing apparatus 100L and the inspection apparatus 300.

[0249] The communication unit 406 receives learning target information LTD and substrate information from the substrate processing apparatus 100L. The control unit 402 stores the learning target information LTD and substrate information received by the communication unit 406 in the storage unit 403. Additionally, the communication unit 406 receives processing result information RS and substrate information from the inspection apparatus 300. The control unit 402 stores the processing result information RS and substrate information received by the communication unit 406 in the storage unit 403.

[0250] Next, refer to Figure 20 and Figure 21 The method for generating learning data in this embodiment will be explained. Figure 21 This is a flowchart of the learning data generation method according to this embodiment. The generation of learning data LFD is performed in the learning data generation apparatus 400.

[0251] like Figure 21 As shown, in step S111, the learning data generation apparatus 400 (control unit 402) acquires the learning object information LTD. Typically, the learning data generation apparatus 400 (control unit 402) acquires the learning object information LTD from the substrate processing apparatus 100L. The storage unit 403 stores the acquired learning object information LTD. More specifically, the learning data generation apparatus 400 (control unit 402) acquires substrate information together with the learning object information LTD.

[0252] In step S112, the learning data generation device 400 (control unit 402) acquires the processing result information RS. Typically, the learning data generation device 400 (control unit 402) acquires the processing result information RS from the inspection device 300. The storage unit 403 stores the acquired processing result information RS. More specifically, the learning data generation device 400 (control unit 402) acquires substrate information together with the processing result information RS.

[0253] In step S113, the learning data generation apparatus 400 (control unit 402) associates the learning object information LTD with the processing result information RS to generate learning data LFD. Specifically, the learning data generation apparatus 400 (control unit 402) uses substrate information to associate the learning object information LTD with the processing result information RS. Multiple learning data LFDs are stored in the storage unit 403. For example, learning data LFDs are stored for each batch.

[0254] Furthermore, the learning data generation device 400 can also be communicatively connected to multiple substrate processing devices 100L.

[0255] Next, refer to Figure 22 The learning device 500 will be described. Figure 22 This is a block diagram representing the learning device 500. For example... Figure 22 As shown, the learning device 500 includes a control device 501. The control device 501 includes a control unit 502 and a storage unit 503. The control unit 502 is an example of a learning unit.

[0256] The control unit 502 has a processor. The control unit 502 may have a CPU or MPU, for example. Alternatively, the control unit 502 may also have a general-purpose arithmetic unit. The control unit 502 may also have an NCU (Neural Network Processing Unit).

[0257] Storage unit 503 stores control program PG4 and data. Control program PG4 is a computer program. Storage unit 503 has a main storage device. Storage unit 503 may also have an auxiliary storage device. Storage unit 503 may include removable media. Control unit 502 performs various processes based on control program PG4 and data stored in storage unit 503.

[0258] The storage unit 503 also stores the learning program LPG. The control unit 502 stores multiple learning data LFDs obtained from the learning data generation device 400 in the storage unit 503.

[0259] The learning program LPG is a program used to identify fixed rules from multiple training data sets (LFDs) and execute a machine learning algorithm to generate a learned model LM that represents the discovered rules. The control unit 502 executes the learning program LPG, performs machine learning on the multiple training data sets (LFDs), and adjusts the parameters of the inference program. As a result, a learned model LM is generated.

[0260] Machine learning algorithms, as long as they are supervised learning, are not particularly limited in scope; for example, they can be decision trees, nearest neighbor methods, simple Bayesian classifiers, support vector machines, or neural networks. Therefore, a learned model (LM) can include decision trees, nearest neighbor methods, simple Bayesian classifiers, support vector machines, or neural networks. Backpropagation can also be used in the machine learning process that generates the learned model (LM).

[0261] For example, a neural network includes an input layer, one or more intermediate layers, and an output layer. Specifically, neural networks are deep neural networks (DNNs), recurrent neural networks (RNNs), or convolutional neural networks (CNNs) used for deep learning. For example, a deep neural network includes an input layer, multiple intermediate layers, and an output layer.

[0262] like Figure 22As shown, the learning device 500 also includes a display unit 504, an input unit 505, and a communication unit 506.

[0263] Display unit 504 displays various screens and information. Display unit 504 is, for example, a liquid crystal display or an organic EL display. Display unit 504 may also display input screens for inputting various data to control unit 502.

[0264] The input unit 505 accepts input from the operator and outputs various information to the control unit 502. The input unit 505 may include, for example, a touch panel and an indicator device. The input unit 505 may also accept input indicating the generation of a learned model LM.

[0265] The communication unit 506 is connected to a network and communicates with external devices. The communication unit 506 is a communication device, such as a network interface controller. In this embodiment, the communication unit 506 performs communication with the learning data generation device 400. Additionally, as shown in the reference... Figure 12 As explained, the communication unit 506 performs communication with the board processing device 100.

[0266] The communication unit 506 receives multiple learning data LFDs from the learning data generation device 400. The control unit 502 stores the multiple learning data LFDs received by the communication unit 506 in the storage unit 503. In addition, the communication unit 506 transmits the learned model LM to the board processing device 100.

[0267] Next, refer to Figure 22 and Figure 23 The learning method and the method for generating the learned model in this embodiment will be explained. Figure 23 This is a flowchart of the learning method and the method for generating the learned model in this embodiment. The learning of the learning data LFD and the generation of the learned model LM are performed in the learning device 500.

[0268] like Figure 23 As shown, in step S121, the control unit 502 obtains multiple learning data LFDs from the storage unit 503.

[0269] Next, in step S122, the control unit 502 performs machine learning on multiple learning data LFDs based on the learning program LPG.

[0270] Next, in step S123, the control unit 502 determines whether the machine learning of multiple learning data LFDs has ended. The decision to end the machine learning is based on predetermined conditions. For example, the machine learning may end when the number of learning data LFDs exceeds a predetermined number.

[0271] If the machine learning process has not ended ("No" in step S123), the process returns to step S121. In this case, the machine learning process is repeated. On the other hand, if the machine learning process has ended ("Yes" in step S123), the process proceeds to step S124.

[0272] In step S124, the control unit 502 outputs a model (one or more functions) that applies multiple learned parameters (coefficients) as a learned model LM. The storage unit 503 stores the learned model LM. As a result, the learning method ends, and the learned model LM is generated.

[0273] In addition, the learning device 500 can also be communicatively connected to multiple learning data generation devices 400.

[0274] Above, refer to Figures 17-23 The process for generating the learned model LM is described. According to this embodiment, the processing result information RS is associated with the learning object information LTD to generate learning data LFD. The learning object information LTD includes learning time sequence information TPR, learning object piece number information TC, first baffle opening setting information DS1, and second baffle opening setting information DS2. (See reference...) Figure 1 As explained, the processing result information RS indicates the number of particles generated on the substrate WLa after processing. Alternatively, the processing result information RS indicates whether the number of particles exceeds a predetermined value. Therefore, by adjusting the opening of the first baffle 73 and the opening of the second baffles 87, 88, and 89 by referring to the baffle opening adjustment information PA output from the learned model LM, the number of particles can be suppressed.

[0275] Furthermore, according to this embodiment, the exhaust pressure of the processing unit 1 can be varied depending on the type of processing liquid used to process the substrate Wp and the order in which the substrate Wp is processed. Therefore, compared to a structure that controls the exhaust pressure of the processing unit 1 to a fixed value, the number of particles can be suppressed. For example, when the first processing liquid is SPM, smoke is sometimes generated and the number of particles increases. In contrast, according to this embodiment, when the first processing liquid is SPM, the exhaust pressure of the processing unit 1 can be increased to suppress the generation of smoke.

[0276] In this embodiment, the storage unit 103 of the substrate processing apparatus 100 stores the learned model LM, but the storage unit 103 of the substrate processing apparatus 100 may not store the learned model LM. In this case, the substrate processing apparatus 100 may also send the device information MD to the learning device 500 and obtain the baffle opening adjustment information PA from the learning device 500.

[0277] In addition, in this embodiment, the learning data generation device 400 generates learning data LFD, but the substrate processing device 100L can also generate learning data LFD, and the learning device 500 can also generate learning data LFD.

[0278] In addition, in this embodiment, the learning device 500 generates a learned model LM, but the substrate processing device 100L can also generate a learned model LM, and the learning data generation device 400 can also generate a learned model LM.

[0279] In addition, in this embodiment, the learning object information LTD includes time series data (concentration information) representing the concentration of each of the first to third processing liquids, time series data (temperature information) representing the internal temperature of the processing unit 1, time series data (flow rate information) representing the gas flow rate from the blowing unit 38 to the interior of the processing unit 1 (processing housing 2), position information TPD, and distance information TLD. However, the learning object information LTD may include a portion of the concentration information, temperature information, flow rate information, position information TPD, and distance information TLD, or it may not include these information.

[0280] The above is with reference to the attached diagram ( Figures 1 to 23 The embodiments of the present invention have been described. However, the present invention is not limited to the embodiments described above, and can be implemented in various ways without departing from its spirit. In addition, the various constituent elements disclosed in the above embodiments can be appropriately changed. For example, a constituent element from all the constituent elements shown in one embodiment can be added to the constituent elements of other embodiments, or several constituent elements from all the constituent elements shown in one embodiment can be deleted from the embodiment.

[0281] To facilitate understanding of the invention, the accompanying drawings are schematically represented by their respective constituent elements. The thickness, length, number, and spacing of each constituent element in the drawings may sometimes differ from the actual dimensions for ease of fabrication. Furthermore, the structures of the constituent elements shown in the above-described embodiments are merely examples and are not particularly limited. Of course, various modifications can be made without substantially departing from the effects of the invention.

[0282] For example, in reference Figures 1 to 23 In the described embodiment, multiple processing units 1A, 1B, 1C, and 1D are provided. However, one, two, or three of these processing units may also be provided. Alternatively, in addition to the multiple processing units 1A, 1B, 1C, and 1D, other multiple processing units 1 stacked vertically may also be provided.

[0283] In addition, in reference Figures 1 to 23 In the described embodiment, multiple processing units 1A, multiple processing units 1B, multiple processing units 1C and multiple processing units 1D are provided. However, it is also possible to provide one processing unit 1A, one processing unit 1B, one processing unit 1C and one processing unit 1D, or to provide two or three of the following processing units 1: one processing unit 1A, one processing unit 1B, one processing unit 1C and one processing unit 1D.

[0284] In addition, in reference Figures 1 to 23 In the described embodiment, the processing unit 1 is stacked in 6 layers, but the processing unit 1 can also be stacked in 2, 3, 4 or 5 layers. Alternatively, the processing unit 1 can also be stacked in 7 or more layers.

[0285] In addition, in reference Figures 1 to 23 In the described embodiment, the processing unit 1 is a structure that can use three kinds of processing liquids to process the substrate W, but the processing unit 1 can also be a structure that uses one kind of processing liquid or two kinds of processing liquids to process the substrate W, or it can be a structure that uses four or more kinds of processing liquids to process the substrate W.

[0286] In addition, in reference Figures 1 to 23 In the described embodiment, one processing unit 1 is provided with three separate exhaust pipes (first separate exhaust pipe 41, second separate exhaust pipe 42, and third separate exhaust pipe 43), but for example, when the substrate W is processed using a single processing liquid, one processing unit 1 is provided with one separate exhaust pipe.

[0287] In addition, in reference Figures 1 to 23 In the described embodiment, a baffle (first baffle 73) is used as the first pressure regulating mechanism. However, the first pressure regulating mechanism is not limited to a baffle as long as the pressure of the gas flowing from the processing unit 1 into the first separate exhaust pipe 41 to the third separate exhaust pipe 43 can be regulated. For example, a fan can also be used as the first pressure regulating mechanism. In this case, the operation of the first pressure regulating mechanism is specified by the processing pressure regulating information, which specifies the fan operation. Specifically, the fan speed is specified. In addition, the operation of the first pressure regulating mechanism is specified by the learning pressure regulating information, which represents time-series data of the fan speed during the processing of the learning target substrate WL. In addition, both a fan and a baffle can be used as the first pressure regulating mechanism.

[0288] In addition, in reference Figures 1 to 23In the described embodiment, baffles (second baffles 87, 88, 89) are used as the second pressure regulating mechanism. However, the second pressure regulating mechanism is not limited to baffles as long as the pressure of the gas inside the common exhaust pipe (common exhaust pipes 81-83) can be regulated. For example, a fan can also be used as the second pressure regulating mechanism. In this case, the operation of the second pressure regulating mechanism is specified by the pressure regulating information during processing. Specifically, the fan speed is specified. In addition, the operation of the second pressure regulating mechanism is specified by the pressure regulating information during learning, representing time-series data of the fan speed during the processing of the learning target substrate WL. Furthermore, both a fan and baffles can be used as the second pressure regulating mechanism.

[0289] In addition, in reference Figures 1 to 23 In the described embodiment, a first baffle 73 is provided in the switching mechanism 51, but for example, Figure 24 As shown, the first baffle 73 can also be installed on the exhaust pipe (exhaust pipe 40). See below for reference. Figure 24 Another embodiment of the substrate processing apparatus 100 of the present invention will be described.

[0290] Figure 24 This is a system diagram of the exhaust path of a substrate processing apparatus 100 according to another embodiment of the present invention. More specifically, Figure 24 This indicates a portion of the exhaust path. Figure 24 The substrate processing apparatus 100 shown includes a plurality of exhaust pipes 40, a plurality of switching mechanisms 50 and a plurality of first baffles 73.

[0291] like Figure 24 As shown, each processing unit 1 is provided with an exhaust pipe 40, a switching mechanism 50, and a first baffle 73. Therefore, one exhaust pipe 40 is connected to the processing unit 1. The first baffle 73 is disposed on the exhaust pipe 40. The exhaust pipe 40 allows gas discharged from the processing unit 1 to flow to the switching mechanism 50. The switching mechanism 50 allows gas flowing in from the exhaust pipe 40 to flow into any one of the first separate exhaust pipe 41, the second separate exhaust pipe 42, and the third separate exhaust pipe 43.

[0292] Industrial availability

[0293] This invention is suitable for substrate processing apparatus, substrate processing method, method for generating learning data, learning method, learning apparatus, method for generating learned model, and learned model.

[0294] Symbol Explanation

[0295] 1: Processing Unit

[0296] 41: First separate exhaust pipe

[0297] 42: Second separate exhaust pipe

[0298] 43: Third separate exhaust pipe

[0299] 51: Switching mechanism

[0300] 73: First baffle

[0301] 81: Shared exhaust pipe

[0302] 82: Shared exhaust pipe

[0303] 83: Shared exhaust pipe

[0304] 87: Second baffle

[0305] 88: Second baffle

[0306] 89: Second baffle

[0307] 100: Substrate processing apparatus

[0308] 102: Control Department

[0309] 103: Storage Department

[0310] 500: Learning Device

[0311] 502: Control Department

[0312] 503: Storage Department

[0313] LFD: Learning Data

[0314] LM: Learned Model

[0315] LTD: Learning Object Information

[0316] MPR: Processing sequence information

[0317] PA: Baffle opening adjustment information

[0318] RS: Processing result information

[0319] TPR: Learning Sequence Information

[0320] TR: Time Series Data

[0321] W: substrate.

Claims

1. A substrate processing apparatus, characterized in that, have: A processing unit supplies processing liquid to a substrate to be processed according to processing sequence information, wherein the processing sequence information indicates the order in which the substrate to be processed is processed using the processing liquid. An exhaust pipe that allows gas discharged from the processing unit to flow in; A pressure regulating mechanism that regulates the pressure of the gas flowing from the processing unit into the exhaust pipe; as well as The control unit, when inputting the processing sequence information, inputs the processing sequence information into a learned model that outputs processing pressure regulation information specifying the operation of the pressure regulation mechanism, obtains the processing pressure regulation information from the learned model, and controls the pressure regulation mechanism based on the obtained processing pressure regulation information. The learned model is constructed by performing machine learning on learning data that associates processing result information with learning object information. The processing result information represents the result of processing the learning object substrate using the processing liquid. The learning object information includes learning pressure adjustment information and learning sequence information. The learning pressure adjustment information indicates the operation of the pressure adjustment mechanism when processing the learning object substrate, and the learning sequence information indicates the order in which the learning object substrate is processed using the processing liquid.

2. The substrate processing apparatus according to claim 1, characterized in that, The substrate processing apparatus also includes a storage unit for storing the learned model.

3. The substrate processing apparatus according to claim 1 or 2, characterized in that, The learning object information also includes at least one of the following: concentration information, which represents the concentration of the treatment liquid; temperature information, which represents the internal temperature of the treatment unit; position information, which represents the position of the treatment unit; and distance information, which represents the distance from when the gas discharged from the treatment unit is discharged from the downstream end of the exhaust pipe.

4. A substrate processing method, comprising processing a substrate to be processed in a substrate processing apparatus. The substrate processing apparatus includes: A processing unit supplies processing liquid to a substrate to be processed based on processing sequence information, wherein the processing sequence information indicates the order in which the substrate to be processed is processed using the processing liquid. An exhaust pipe that allows gas discharged from the processing unit to flow in; A pressure regulating mechanism that regulates the pressure of the gas flowing from the processing unit into the exhaust pipe; and The control unit controls the pressure regulating mechanism. Its features are, The substrate processing method includes the following steps: When the processing sequence information is input, the processing sequence information is input into a learned model that outputs processing pressure regulation information specifying the action of the pressure regulation mechanism, and the processing pressure regulation information is obtained from the learned model; and The pressure regulating mechanism is controlled based on the obtained pressure regulation information during processing. The learned model is constructed by performing machine learning on learning data that associates processing result information with learning object information. The processing result information represents the result of processing the learning object substrate using the processing liquid. The learning object information includes learning pressure adjustment information and learning sequence information. The learning pressure adjustment information indicates the operation of the pressure adjustment mechanism when processing the learning object substrate, and the learning sequence information indicates the order in which the learning object substrate is processed using the processing liquid.

5. A method for generating learning data, characterized in that, The generation method includes the following steps: Obtain processing result information that indicates the result of the substrate processing device processing the learning substrate; Obtain information about the learning target; and The processing result information is associated with the learning object information and stored in the storage unit as learning data. The substrate processing apparatus includes: A processing unit supplies processing liquid to the learning target substrate based on learning time sequence information, wherein the learning time sequence information indicates the order in which the learning target substrate is processed using the processing liquid. An exhaust pipe that allows gas discharged from the processing unit to flow in; A pressure regulating mechanism that regulates the pressure of the gas flowing from the processing unit into the exhaust pipe; and The control unit controls the pressure regulating mechanism. The learning object information includes learning pressure adjustment information and learning sequence information. The learning pressure adjustment information indicates the action of the pressure adjustment mechanism when processing the learning object substrate.

6. A learning method, characterized in that, Includes the following steps: Obtain the learning data generated by the method for generating learning data according to claim 5; and The learning data is input into a learning program to perform machine learning on the learning data.

7. A learning device, characterized in that, have: Storage unit, which stores learning data generated according to the method for generating learning data as described in claim 5; and The learning department inputs the learning data into the learning program to perform machine learning on the learning data.

8. A method for generating a learned model, characterized in that, The generation method includes the following steps: Obtain the learning data generated by the method for generating learning data according to claim 5; and A learned model is generated by performing machine learning on the learning data.

9. A learned model, characterized in that, The learned model is constructed by performing machine learning on the learning data generated according to the method for generating learning data as described in claim 5. The control unit of the pressure regulating mechanism functions in the following manner: when processing sequence information indicating the order in which the substrate to be processed is processed using the processing liquid is input, processing time pressure regulation information specifying the operation of the pressure regulating mechanism is output, wherein the pressure regulating mechanism regulates the pressure of the gas flowing from the processing unit that processes the substrate to the exhaust pipe.

Images