Substrate processing device and substrate processing method

Abstract

To provide a technique capable of testing monitoring processing on a user side.SOLUTION: A substrate processing device includes a chamber 10, a discharge part 3, a camera 5, a control part 90, and a storge part. A substrate W is carried into the chamber 10. The discharge part 3 includes a discharge port 31a for discharging a fluid in the chamber 10 and causes the fluid to act on the substrate W. The camera 5 images a monitoring object in the chamber 10 and generates image data IM1. The control part 90 controls the discharge part 3 and allows the camera 5 to image the monitoring object, while processing the substrate W carried in the chamber 10. The control part 90 monitors the monitoring object by first monitoring processing which is based on the image data IM1, monitors the monitoring object by second monitoring processing which is based on the image data IM1 and is different from the first monitoring processing, determines whether or not abnormality processing is performed by using the first monitoring processing without using the second monitoring processing, and stores log information including an arithmetic result of the second monitoring processing in the storage part.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to a substrate processing apparatus and a substrate processing method.

Background Art

[0002] Conventionally, a substrate processing apparatus that executes processing on a substrate while monitoring the inside of a chamber with a camera has been proposed (Patent Document 1). In Patent Document 1, the substrate processing apparatus includes a substrate holding unit, a nozzle, a camera, and a control unit. The substrate holding unit holds the substrate in a horizontal posture in the chamber and rotates the substrate around a rotation axis passing through the center of the substrate. The nozzle discharges a processing liquid toward the main surface of the substrate in the chamber. The processing liquid adhering to the main surface of the substrate flows radially outward as the substrate rotates and scatters outward from the periphery of the substrate. At this time, when the processing liquid acts on the main surface of the substrate, processing corresponding to the type of the processing liquid is performed on the substrate.

[0003] The camera images an imaging region including the tip of the nozzle and generates image data. The control unit determines the discharge state of the processing liquid discharged from the nozzle by an algorithm (for example, a learned model) based on the image data. For example, when the control unit determines that liquid droplet-like processing liquid is falling from the nozzle at the time of stopping the discharge of the processing liquid, it determines that a dripping abnormality has occurred.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In a substrate processing apparatus installed in a user's factory, the monitoring result of the monitoring process based on the image data may be incorrect. In this case, the manufacturer's worker tunes the parameters used in the monitoring process algorithm or changes the monitoring process algorithm itself to create a new monitoring process. This new monitoring process can be implemented and tested, for example, on the manufacturer's substrate processing apparatus.

[0006] However, the manufacturer's test environment is not necessarily the same as the user's substrate processing apparatus. For example, the type of substrate used in the user's substrate processing apparatus may be different from the type of substrate used in the manufacturer's test. Or, the processing conditions such as the type of processing fluid supplied to the substrate may also be different between the user's substrate processing apparatus and the manufacturer's substrate processing apparatus. Since the image data used in the monitoring process includes not only the object to be monitored but also the environment in the surrounding chamber, this environment in the chamber can also affect the monitoring result of the monitoring process. Therefore, even if the substrate processing apparatus installed on the user side monitors the object to be monitored with the new monitoring process, there is still a possibility that an error will occur in the monitoring result.

[0007] Therefore, an object of the present disclosure is to provide a substrate processing apparatus and a substrate processing method capable of executing a test of a monitoring process on the user side.

Means for Solving the Problem

[0008] A first aspect is a substrate processing apparatus, comprising: a chamber into which a substrate is loaded; a discharge unit having a discharge port for discharging a fluid in the chamber and causing the fluid to act on the substrate; a camera for imaging a monitoring target in the chamber to generate image data; a control unit for controlling the discharge unit to process the substrate loaded into the chamber and causing the camera to image the monitoring target; and a storage unit. The control unit monitors the monitoring target by a first monitoring process based on the image data, monitors the monitoring target by a second monitoring process different from the first monitoring process based on the image data, determines whether to execute an abnormality process using the first monitoring process without using the second monitoring process, and stores log information including the calculation result of the second monitoring process in the storage unit.

[0009] A second aspect is the substrate processing apparatus according to the first aspect, wherein the log information includes the image data.

[0010] A third aspect is the substrate processing apparatus according to the first or second aspect, further comprising: a display for displaying the log information; and an input device for receiving an input of a switching instruction for switching the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process. The control unit switches the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process in response to the input of the switching instruction.

[0011] A fourth aspect is the substrate processing apparatus according to the third aspect, wherein the storage unit stores the log information including the calculation result of the first monitoring process, and the display also displays the calculation result of the first monitoring process.

[0012] A fifth aspect is the substrate processing apparatus according to the first or second aspect, wherein the control unit determines whether to switch the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process based on the log information.

[0013] The sixth aspect is the substrate processing apparatus according to the fifth aspect, wherein the monitoring accuracy of the second monitoring process when the environment in the chamber is in the first environmental state is higher than the monitoring accuracy of the first monitoring process when the environment in the chamber is in the first environmental state, and the control unit is based on a comparison between the calculation result of the first monitoring process and the calculation result of the second monitoring process in an environmental state other than the first environmental state among the log information, and determines whether to switch the monitoring process used for determining the execution of the abnormal process from the first monitoring process to the second monitoring process.

[0014] The seventh aspect is the substrate processing apparatus according to any one of the first to sixth aspects, wherein the algorithms of the first monitoring process and the second monitoring process are different from each other.

[0015] The eighth aspect is the substrate processing apparatus according to any one of the first to sixth aspects, wherein the algorithms of the first monitoring process and the second monitoring process are the same as each other, and the parameters used in the algorithm are different between the first monitoring process and the second monitoring process.

[0016] The ninth aspect is the substrate processing apparatus according to the eighth aspect, wherein the storage unit stores the parameter of the first monitoring process and the parameter of the second monitoring process in the storage unit, and the control unit copies the value of the parameter of the second monitoring process to the value of the parameter of the first monitoring process.

[0017] The tenth aspect is the substrate processing apparatus according to any one of the first to ninth aspects, wherein either one of the first monitoring process and the second monitoring process is a process using a learned model.

[0018] The 11th aspect is a substrate processing method, comprising: a step of loading a substrate into a chamber; a step of causing a fluid to act on the substrate in the chamber; a step of a camera imaging a monitoring target in the chamber to generate image data; monitoring the monitoring target by a first monitoring process based on the image data, monitoring the monitoring target by a second monitoring process different from the first monitoring process based on the image data; determining whether to execute an abnormality process using the first monitoring process without using the second monitoring process; and a step of storing log information including the calculation result of the second monitoring process in a storage unit.

Effect of the Invention

[0019] According to the 1st and 11th aspects, since the storage unit stores log information including the calculation result of the second monitoring process, the monitoring accuracy of the second monitoring process can be confirmed. That is, while performing abnormality determination by the first monitoring process, the second monitoring process can be tested in the installation environment of the substrate processing apparatus on the user side.

[0020] According to the 2nd aspect, image data is also stored as log information. Therefore, the log information can be enriched.

[0021] According to the 3rd aspect, by a worker checking on a display, the log information of the calculation result of the second monitoring process in the installation environment where the substrate processing apparatus is installed can be confirmed. And when the worker confirms based on the log information that the monitoring accuracy of the second monitoring process is sufficient, a switching instruction can be input to an input device. Thereby, the monitoring process used for determining the execution of the abnormality process can be switched from the first monitoring process to the second monitoring process. That is, the monitoring process can be switched after confirming the test result of the second monitoring process in the installation environment.

[0022] According to the 4th aspect, since the calculation result of the first monitoring process is also displayed, the worker can compare the calculation result of the first monitoring process with the calculation result of the second monitoring process. Therefore, it is easy for the worker to determine whether the monitoring accuracy of the second monitoring process is sufficient.

[0023] According to the fifth aspect, since the control unit automatically determines whether or not to switch the monitoring process, the burden on the operator can be reduced.

[0024] According to the sixth aspect, since the control unit compares the calculation results in an environment where the monitoring accuracy has not been confirmed, it is possible to appropriately determine the first monitoring process and the second monitoring process.

[0025] According to the seventh aspect, the monitoring accuracy of the second monitoring process can be further improved.

[0026] According to the eighth aspect, since the monitoring process can be switched by changing the parameter, the switching of the monitoring process is easy.

[0027] According to the ninth aspect, the control unit can easily switch the first monitoring process to the second monitoring process.

[0028] According to the tenth aspect, the control unit can monitor the object to be monitored with high monitoring accuracy.

Brief Description of the Drawings

[0029]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Embodiments for Carrying Out the Invention

[0030] Hereinafter, embodiments will be described in detail with reference to the drawings. In the drawings, for the purpose of easy understanding, the dimensions and numbers of each part are exaggerated or simplified as necessary. Also, parts having the same configuration and function are denoted by the same reference numerals, and redundant descriptions are omitted in the following description.

[0031] Also, in the descriptions shown below, the same reference numerals are used to illustrate the same components, and their names and functions are also assumed to be the same. Therefore, detailed descriptions thereof may be omitted to avoid duplication.

[0032] Also, in the descriptions described below, even when ordinal numbers such as "first" or "second" are used, these terms are used for convenience in order to facilitate understanding of the content of the embodiments, and are not limited to the order that may be caused by these ordinal numbers.

[0033] When expressions indicating relative or absolute positional relationships (such as "in one direction", "along one direction", "parallel", "orthogonal", "center", "concentric", "coaxial", etc.) are used, unless otherwise specified, such expressions shall not only precisely represent the positional relationship, but also represent a state where the relative angle or distance is displaced within the range where tolerances or equivalent functions can be obtained. When expressions indicating an equal state (such as "identical", "equal", "homogeneous", etc.) are used, unless otherwise specified, such expressions shall not only precisely represent a quantitatively equal state, but also represent a state where there are tolerances or differences that can obtain equivalent functions. When expressions indicating a shape (such as "quadrilateral shape" or "cylindrical shape", etc.) are used, unless otherwise specified, such expressions shall not only geometrically and precisely represent the shape, but also represent shapes having, for example, unevenness or chamfers within the range where equivalent effects can be obtained. When expressions such as "comprising", "having", "including", or "possessing" are used for one component, such expressions are not exclusive expressions excluding the existence of other components. When the expression "at least any one of A, B, and C" is used, such expression includes only A, only B, only C, any two of A, B, and C, and all of A, B, and C.

[0034] <Overall Configuration of Substrate Processing Apparatus> FIG. 1 is a plan view schematically showing an example of the configuration of a substrate processing apparatus 100. The substrate processing apparatus 100 is, for example, a single-wafer processing apparatus that processes substrates W one by one.

[0035] The substrate W is, for example, a semiconductor wafer, a substrate for a liquid crystal display, a substrate for an organic EL (Electroluminescence), a substrate for an FPD (Flat Panel Display), a substrate for an optical display, a substrate for a magnetic disk, a substrate for an optical disk, a substrate for a magneto-optical disk, a substrate for a photomask, or a substrate for a solar cell. The substrate W has a thin flat plate shape. Hereinafter, it is assumed that the substrate W is a semiconductor wafer. The substrate W has, for example, a disk shape. The diameter of the substrate W is, for example, about 300 mm, and the film thickness of the substrate W is, for example, about 0.5 mm or more and about 3 mm or less.

[0036] In the example of FIG. 1, the substrate processing apparatus 100 includes an index block 110, a processing block 120, a control unit 90, and a user interface 95. The processing block 120 is mainly a part for processing the substrate W, and the index block 110 is mainly a part for transporting the substrate W between the outside of the substrate processing apparatus 100 and the processing block 120.

[0037] The index block 110 includes a load port 111 and a first transfer unit 112. A substrate container (hereinafter referred to as a carrier) C carried in from the outside is placed on the load port 111. A plurality of substrates W are accommodated in the carrier C in a state of being arranged at intervals in the vertical direction, for example. In the example of FIG. 1, a plurality of load ports 111 are arranged.

[0038] The first transfer unit 112 is a transfer robot and can take out an unprocessed substrate W from the carrier C placed on each load port 111. The first transfer unit 112 can also be called an index robot. The first transfer unit 112 transports the unprocessed substrate W taken out from the carrier C to the processing block 120. The processing block 120 can perform processing on the unprocessed substrate W. Further, the first transfer unit 112 can receive the processed substrate W from the processing block 120 and transport the processed substrate W to the carrier C of the load port 111.

[0039] In the example of FIG. 1, the processing block 120 includes a plurality of processing units 1 and a second transfer unit 122. The second transfer unit 122 is a transfer robot and can transfer the substrate W between the first transfer unit 112 and the plurality of processing units 1. In the example of FIG. 1, the processing block 120 also includes a placement unit 123. The placement unit 123 is, for example, a shelf on which a plurality of substrates W can be placed side by side in the vertical direction. The first transfer unit 112 places the unprocessed substrate W on the placement unit 123. The second transfer unit 122 takes out the unprocessed substrate W from the placement unit 123 and transfers the substrate W to the processing unit 1. The processing unit 1 processes the substrate W. The configuration of the processing unit 1 will be described later. The second transfer unit 122 takes out the processed substrate W from the processing unit 1 and transfers the substrate W to the placement unit 123. The first transfer unit 112 takes out the substrate W from the placement unit 123 and transfers the substrate W to the carrier C of the load port 111.

[0040] In the example of FIG. 1, a plurality (for example, four) of processing units 1 are provided so as to surround the second transfer unit 122 in a plan view. This second transfer unit 122 may also be called a center robot. At each position in the plan view, a plurality of processing units 1 may be stacked in the vertical direction. That is, a plurality (four in the figure) of towers TW each composed of a plurality of processing units 1 stacked in the vertical direction may be provided so as to surround the second transfer unit 122.

[0041] The control unit 90 comprehensively controls the substrate processing apparatus 100. Specifically, the control unit 90 controls the first transfer unit 112, the second transfer unit 122, and the processing unit 1. FIG. 2 is a block diagram schematically showing an example of the configuration of the control unit 90. The control unit 90 is an electronic circuit and has, for example, a data processing unit 91 and a storage unit 92. In the specific example of FIG. 2, the data processing unit 91 and the storage unit 92 are interconnected via a bus 93. The data processing unit 91 may be an arithmetic processing device such as a CPU (Central Processor Unit). The storage unit 92 may have a non-temporary storage unit (for example, ROM (Read Only Memory)) 921 and a temporary storage unit (for example, RAM (Random Access Memory)) 922. A program that defines the processes executed by the control unit 90 may be stored in the non-temporary storage unit 921. By the data processing unit 91 executing this program, the control unit 90 can execute the processes defined in the program. Of course, part or all of the processes executed by the control unit 90 may be executed by hardware such as a dedicated logic circuit.

[0042] As shown in FIG. 2, the control unit 90 may be electrically connected to the storage unit 94. The storage unit 94 is a non-temporary storage unit and may be, for example, a memory or a hard disk. In FIG. 2, log information D1 is stored in the storage unit 94. The log information D1 will be described later.

[0043] The user interface 95 is an interface between the substrate processing apparatus 100 and the user. The user interface 95 includes a notification unit 96 and an input device 97 (see also FIG. 4). The notification unit 96 is controlled by the control unit 90 to notify the user of information. The notification unit 96 includes at least one of, for example, a display 961 and a sound output unit 962. The display 961 is a display such as a liquid crystal display, for example, and displays various information. The sound output unit 962 includes, for example, either a speaker or a buzzer. The speaker notifies the user of information by outputting the information as sound. The buzzer notifies the user of information by outputting, for example, a warning sound. The input device 97 receives input of various instructions from the user. Specific examples of the various instructions will be described later.

[0044] <Outline of the processing unit> FIG. 3 is a diagram schematically showing an example of the configuration of the processing unit 1. Note that not all of the processing units 1 belonging to the substrate processing apparatus 100 need to have the configuration illustrated in FIG. 3. It is sufficient that at least one processing unit 1 has the configuration illustrated in FIG. 3. The processing unit 1 includes a chamber 10, a discharge unit 3, and a camera 5.

[0045] First, the outline of the configuration of the processing unit 1 and the outline of the operation of the control unit 90 will be described below, and then these will be described in detail.

[0046] The chamber 10 has an internal space. The internal space corresponds to a processing space for processing the substrate W. The chamber 10 is provided with an openable and closable carry-in / out port (not shown). The second transfer unit 122 transfers the unprocessed substrate W into the chamber 10 through the carry-in / out port, and also transfers the processed substrate W out of the chamber 10 through the carry-in / out port.

[0047] The ejection unit 3 has an ejection port 31a in the chamber 10 (see, for example, the nozzle 31A in FIG. 3). The ejection unit 3 ejects a fluid (hereinafter referred to as a processing fluid) from the ejection port 31a toward the main surface of the substrate W, and causes the processing fluid to act on the main surface of the substrate W. For example, the ejection unit 3 ejects a liquid (hereinafter also referred to as a processing liquid) as the processing fluid from the ejection port 31a. The processing liquid ejected from the ejection port 31a adheres to the main surface of the substrate W in the chamber 10 and acts on the main surface of the substrate W. Thereby, processing corresponding to the processing liquid can be performed on the main surface of the substrate W. The configuration of the ejection unit 3 and a specific example of the processing fluid will be described later.

[0048] The camera 5 images a monitoring target in the chamber 10. The monitoring target may be, for example, various members that operate during the processing of the substrate W, or may be a processing liquid. Specific examples of the monitoring target will also be described later. If the monitoring target is normal, the processing unit 1 can appropriately process the substrate W. If an abnormality occurs in the monitoring target, there is a possibility that a problem may occur in the processing of the substrate W.

[0049] The camera 5 images the monitoring target and generates image data IM1. The camera 5 sequentially images the monitoring target, for example, in parallel with the processing of the substrate W, and generates a plurality of pieces of image data IM1. The image data IM1 may be still image data or a frame of moving image data. The image data IM1 may include not only the monitoring target but also the environment around the monitoring target. The camera 5 outputs the image data IM1 to the control unit 90.

[0050] The control unit 90 controls the ejection unit 3 to process the substrate W carried into the chamber 10 while causing the camera 5 to image the monitoring target. Further, the control unit 90 monitors the monitoring target based on the image data IM1. More specifically, the control unit 90 performs a first monitoring process and a second monitoring process for the monitoring target based on the image data IM1. The first monitoring process and the second monitoring process are different from each other but are processes for monitoring the same monitoring target.

[0051] The first monitoring process is, for example, the monitoring process before version upgrade. The version upgrade can also be called an update. The monitoring accuracy of this first monitoring process is relatively low. Specifically, the monitoring accuracy of the first monitoring process when the environment in the chamber 10 is in the first environmental state is relatively low. The first environmental state may be various environmental states in the chamber 10. For example, it may be a state in which a specific type of substrate W is carried into the chamber 10. For example, when the processing unit 1 is performing processing on a specific type of substrate W, the camera 5 images the object to be monitored in the chamber 10. The image data IM1 at this time may also include the object to be monitored and the surrounding environment in the chamber 10 (for example, a specific type of substrate W). When the control unit 90 executes the first monitoring process based on this image data IM1, for example, due to the influence of the pattern on the main surface of the substrate W included in the image data IM1, the first monitoring process may make a false determination. The specific example of this first monitoring process and the content of the false determination will also be described in detail later.

[0052] The second monitoring process is the monitoring process scheduled for version upgrade. Although the specific example of the second monitoring process will also be described in detail later, the monitoring accuracy of the second monitoring process is higher than that of the first monitoring process in the first environmental state. This second monitoring process is developed, for example, as follows.

[0053] That is, the operator of the manufacturer of the substrate processing apparatus 100 can develop a second monitoring process with high monitoring accuracy in the first environmental state by reproducing and testing the first environmental state at the manufacturer. For example, by carrying a specific type of substrate W passed from the user into the chamber 10 of the substrate processing apparatus 100 of the manufacturer, the environment in the chamber 10 can be made into the first environmental state. The operator can develop a second monitoring process with high monitoring accuracy in the first environmental state by testing the second monitoring process in this first environmental state. In this case, it can be said that the first environmental state is an environmental state that can be easily reproduced at the manufacturer among various environmental states in the chamber 10 on the user side.

[0054] Alternatively, there may be a case where the image data IM1 when the monitoring result is incorrect is passed from the user to the manufacturer. For example, the image data IM1 includes a specific type of substrate W. Based on the image data IM1, the operator of the manufacturer can develop a second monitoring process with high monitoring accuracy in the first environmental state. In this case, it can be said that the first environmental state is the environmental state shown in the image data IM1.

[0055] However, it is unknown whether the monitoring accuracy of this second monitoring process is higher than that of the first monitoring process in a second environmental state other than the first environmental state. The second environmental state can also be said to be, for example, an environmental state that is difficult to reproduce by the manufacturer among various environmental states in the chamber 10 of the substrate processing apparatus 100 on the user side. As a specific example, the second environmental state may be a state in which a type of substrate W that is not passed from the user to the manufacturer is carried into the chamber 10. Alternatively, it can be said that the second environmental state is a state other than the environmental state included in the image data IM1 from the user.

[0056] As described above, while the second monitoring process can monitor the object to be monitored with higher monitoring accuracy than the first monitoring process in the first environmental state, it is not confirmed whether the object to be monitored can be monitored with high monitoring accuracy in a second environmental state other than the first environmental state.

[0057] Therefore, as described above, the control unit 90 executes both the first monitoring process and the second monitoring process based on the image data IM1. And initially, the control unit 90 executes the abnormality processing based on the monitoring result of the first monitoring process before the version update. Specifically, when the control unit 90 determines that an abnormality has occurred in the object to be monitored as the monitoring result of the first monitoring process, it performs the abnormality processing.

[0058] On the one hand, initially, the control unit 90 stores the monitoring result of the second monitoring process, which is scheduled for version update, in the storage unit 94 as log information D1 without using it for the determination of abnormal process execution. As a specific example, the control unit 90 associates the monitoring result of the second monitoring process with the image data IM1 and stores it in the storage unit 94 as log information D1. That is, the log information D1 includes data indicating a set of the image data IM1 and the monitoring result of the second monitoring process using the image data IM1.

[0059] As described above, the control unit 90 monitors the object to be monitored by the first monitoring process before version update and performs abnormal processing according to the monitoring result, while being able to execute the test of the second monitoring process scheduled for version update in the installation environment on the user side. The test result is stored in the storage unit 94 as log information D1. Then, for example, by the operator checking the log information D1, the monitoring accuracy (that is, the test result) of the second monitoring process in the installation environment on the user side can be confirmed. And if the monitoring accuracy of the second monitoring process is sufficient, the operator inputs, for example, a monitoring process switching instruction (which can also be called an update instruction) to the input device 97. In response to the input, the control unit 90 switches the monitoring process used for the determination of abnormal processing from the first monitoring process to the second monitoring process. Thereby, after sufficiently performing the test of the second monitoring process in the installation environment on the user side, the monitoring process can be switched from the first monitoring process to the second monitoring process. Therefore, the control unit 90 can monitor the object to be monitored with appropriate monitoring accuracy using the switched monitoring process (the second monitoring process).

[0060] Hereinafter, a specific example of the configuration of the processing unit 1 and a specific example of the operation of the control unit 90 will be described in detail.

[0061] <Specific Example of Processing Unit> In the example of Figure 3, a substrate holder 2 is provided within the chamber 10. The substrate holder 2 holds the substrate W in a horizontal position and rotates the substrate W around a rotation axis Q1. The horizontal position here means that the thickness direction of the substrate W is aligned with the vertical direction. The rotation axis Q1 is an axis that passes through the center of the substrate W and is aligned with the vertical direction. Such a substrate holder 2 may also be called a spin chuck.

[0062] In the example of FIG. 3, the substrate holding unit 2 includes a spin base 21, chuck pins 22, and a rotation drive unit 23. The spin base 21 has a plate-like shape (e.g., a disk shape) and is disposed with its thickness direction aligned vertically. A plurality of chuck pins 22 are provided on the upper surface of the spin base 21. The plurality of chuck pins 22 are disposed at equal intervals along the circumferential direction about the rotation axis Q1. The plurality of chuck pins 22 are disposed so as to be displaceable between a holding position and a release position, which will be described below. The holding position is a position where the chuck pins 22 abut against the periphery of the substrate W. The plurality of chuck pins 22 hold the substrate W by stopping at their respective holding positions. FIG. 3 shows the chuck pins 22 stopped at the holding position. The release position is a position where each chuck pin 22 is separated from the substrate W. The plurality of chuck pins 22 stop at their respective release positions, thereby releasing the substrate W from the holding of the substrate W by the plurality of chuck pins 22. The substrate holder 2 also includes a pin drive unit (not shown) that displaces the chuck pins 22. The pin drive unit includes a drive source such as a motor or an air cylinder, and is controlled by the control unit 90.

[0063] The rotation drive unit 23 includes a shaft 231 and a motor 232. The upper end of the shaft 231 is connected to the lower surface of the spin base 21, and the shaft 231 extends from the lower surface of the spin base 21 along a rotation axis Q1. The motor 232 is controlled by the control unit 90 to rotate the shaft 231 about the rotation axis Q1. This causes the spin base 21, chuck pins 22, and substrate W to rotate integrally about the rotation axis Q1.

[0064] Note that the substrate holding part 2 does not necessarily have to have the chuck pins 22. For example, the substrate holding part 2 may hold the substrate W by a chuck method such as a vacuum chuck, an electrostatic chuck, and a Bernoulli chuck.

[0065] In the example of FIG. 3, the discharge part 3 includes a nozzle 31, a supply pipe 32, a supply valve 33, and a flow rate adjustment valve 34. The nozzle 31 is provided in the chamber 10 and has a discharge port 31a. In the example of FIG. 3, the nozzle 31 is positioned above the substrate W held by the substrate holding part 2 and discharges the processing fluid toward the upper surface of the substrate W. In the example of FIG. 3, the discharge port 31a is formed on the lower end surface of the nozzle 31.

[0066] The processing fluid may be a processing liquid. Liquids such as chemical liquids, rinse liquids, charge removal liquids, and coating liquids can be applied to the processing liquid. The chemical liquid may be, for example, a liquid such as hydrofluoric acid that chemically acts on the main surface of the substrate W. The chemical liquid may be, for example, a liquid that removes impurities on the main surface of the substrate W, or a liquid that etches the film on the main surface of the substrate W. The rinse liquid is, for example, a liquid that flushes away the chemical liquid on the main surface of the substrate W and may be, for example, pure water or an organic solvent such as isopropyl alcohol. The charge removal liquid is a liquid that reduces the charge amount of the substrate W and may be, for example, a liquid having conductivity such as carbonated water. Carbonated water can also be used as a rinse liquid. The coating liquid is a liquid having the components of the film to be formed on the main surface of the substrate W and may be, for example, a resist liquid.

[0067] Alternatively, the processing fluid may be a gas. For example, an inert gas is applied to the processing fluid. The inert gas is, for example, nitrogen gas or a noble gas.

[0068] The nozzle 31 is connected to a fluid supply source through a supply pipe 32. The fluid supply source includes a storage section (e.g., a tank) for storing a processing fluid. The supply pipe 32 is provided with a supply valve 33 and a flow rate adjustment valve 34. The supply valve 33 switches the opening and closing of the flow path of the supply pipe 32. The flow rate adjustment valve 34 adjusts the flow rate of the processing fluid flowing through the supply pipe 32. The supply valve 33 and the flow rate adjustment valve 34 are controlled by a control unit 90.

[0069] The nozzle 31 is movably provided by a nozzle movement driving unit 37. The nozzle movement driving unit 37 moves the nozzle 31 between a processing position and a standby position to be described below. The processing position is a position where the nozzle 31 discharges a processing fluid toward the main surface of a substrate W held by a substrate holding unit 2, for example, a position facing the central portion of the substrate W in the vertical direction. In the example of FIG. 3, the nozzle 31 (refer to nozzle 31A) located at the processing position is shown. The standby position is a position where the nozzle 31 does not discharge the processing fluid toward the main surface of the substrate W, for example, a position radially outside the substrate W. The nozzle movement driving unit 37 includes, for example, a driving source such as a motor and a power transmission unit such as a ball screw mechanism and an arm turning mechanism. The nozzle movement driving unit 37 is controlled by a control unit 90.

[0070] In a state where the nozzle movement driving unit 37 positions the nozzle 31 at the processing position, while the substrate holding unit 2 rotates the substrate W, the nozzle 31 discharges a processing liquid onto the substrate W. The processing liquid adhering to the main surface of the substrate W flows radially outward along with the rotation of the substrate W and scatters outward from the periphery of the substrate W. By the processing liquid acting on the main surface of the substrate W, processing corresponding to the type of the processing liquid is performed on the substrate W.

[0071] As shown in FIG. 3, a plurality of nozzles 31 may be provided. The plurality of nozzles 31 may be connected to different fluid supply sources. In the example of FIG. 3, nozzle 31A, nozzle 31B, and nozzle 31C are shown as the plurality of nozzles 31. In the example of FIG. 3, nozzle 31A is connected to a chemical solution supply source, nozzle 31B is connected to a rinse solution supply source, and nozzle 31C is connected to an organic solvent supply source. Nozzle 31A discharges a chemical solution, nozzle 31B discharges a rinse solution such as pure water, and nozzle 31C discharges an organic solvent such as isopropyl alcohol.

[0072] In the example of FIG. 3, nozzle 31C is provided so that not only an organic solvent but also an inert gas can be discharged. For example, inside nozzle 31C, a first flow path for an organic solvent and a second flow path through which an inert gas flows are formed. The first flow path has, for example, a cylindrical shape extending along the vertical direction, and the second flow path has, for example, an annular shape extending coaxially around the first flow path. The first flow path and the second flow path are connected to an organic solvent supply source and an inert gas supply source, respectively.

[0073] By sequentially supplying various processing fluids from each nozzle 31 to the main surface of the substrate W, the substrate W can be processed. Examples of the processing of the substrate W will be described in detail later.

[0074] In the example of FIG. 3, a guard 7 and a guard lifting and lowering drive unit 71 are provided in the processing unit 1. The guard 7 has a cylindrical shape with the rotation axis Q1 as the central axis and surrounds the substrate holding unit 2. The guard 7 can receive the processing liquid scattered from the periphery of the substrate W. The guard lifting and lowering drive unit 71 raises and lowers the guard 7 between the upper position and the lower position described below. The upper position is a position where the upper end of the guard 7 is vertically above the substrate W held by the substrate holding unit 2. The guard 7 can receive the processing liquid scattered from the periphery of the substrate W in the state of being in the upper position. The lower position is a position lower than the upper position. For example, it is a position where the upper end of the guard 7 is vertically below the upper surface of the spin base 21. The guard lifting and lowering drive unit 71 includes, for example, a drive source such as a motor and a power transmission unit such as a ball screw mechanism and a cam mechanism. The guard lifting and lowering drive unit 71 is controlled by the control unit 90.

[0075] A plurality of guards 7 may be provided. The plurality of guards 7 are arranged concentrically. These plurality of guards 7 may be selectively used according to the type of the processing liquid. In the example of FIG. 3, a cup 75 corresponding to the guard 7 is provided. The cup 75 has an annular (for example, circular ring-shaped) recess (groove) surrounding the rotation axis Q1. The cup 75 receives the processing liquid flowing down the inner peripheral surface of the corresponding guard 7. For example, the upstream end of a discharge pipe 76 is connected to the bottom of the cup 75. The processing liquid received by the cup 75 is discharged to the outside of the processing unit 1 through the discharge pipe 76.

[0076] In the example of FIG. 3, a heater 8 is provided. The heater 8 heats the substrate W held by the substrate holding unit 2. In the example of FIG. 3, the heater 8 is provided between the spin base 21 and the substrate W. The heater 8 may be an electric resistance type heater having a heating wire, or may be an optical type heater that emits heating light (for example, infrared rays). The heater 8 is controlled by the control unit 90.

[0077] In the example of FIG. 3, the camera 5 is provided inside the chamber 10. The camera 5 is fixed to the chamber 10 by a fixing member (not shown), for example. The camera 5 includes a solid-state imaging device such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor), and an optical system such as a lens. The camera 5 is controlled by the control unit 90 to image a monitoring object inside the chamber 10, generate image data IM1, and output the image data IM1 to the control unit 90.

[0078] The camera 5 is provided at a position where its imaging area includes the monitoring object inside the chamber 10. Although the monitoring object is not particularly limited, as a specific example, at least one of the substrate holding unit 2 (for example, the chuck pin 22), the nozzle 31, the processing liquid, and the guard 7 can be applied. In the example of FIG. 3, the camera 5 is provided at a position vertically above and radially outside the substrate W held by the substrate holding unit 2. As shown in FIG. 3, the camera 5 may be provided radially outside the guard 7. In the example of FIG. 3, the camera 5 images the imaging area from an obliquely upper direction. In other words, the camera 5 images the imaging area along an obliquely lower direction.

[0079] The control unit 90 also functions as an image processing unit that processes the image data IM1. The control unit 90 monitors the object to be monitored in the chamber 10 based on the image data IM1. FIG. 4 is a functional block diagram schematically showing an example of the internal configuration of the control unit 90. The control unit 90 includes a first monitoring processing unit 901, a second monitoring processing unit 902, a switching unit 903, and an abnormality processing unit 904. The first monitoring processing unit 901 executes first monitoring processing to monitor the object to be monitored based on the image data IM1 from the camera 5, and outputs the calculation result thereof. The calculation result referred to here may be, for example, a monitoring result that distinguishes between normal and abnormal of the object to be monitored, or may be a value of an index (described later) used for determining normal and abnormal of the object to be monitored. A specific example of the first monitoring processing will be described later. Initially, the control unit 90 determines whether or not to execute abnormality processing based on the calculation result of the first monitoring processing. The abnormality processing unit 904 may, for example, cause the notification unit 96 to notify an abnormality as an abnormality process, or may interrupt the operation of the processing unit 1.

[0080] The second monitoring processing unit 902 executes second monitoring processing to monitor the object to be monitored based on the image data IM1 from the camera 5, and outputs the calculation result thereof. The second monitoring processing is different from the first monitoring processing, and a specific example thereof will be described later. Initially, the control unit 90 stores the calculation result of the second monitoring processing in the storage unit 94 as log information D1 without using it for determining the execution of abnormality processing.

[0081] In response to an input of a display instruction of the log information to the input device 97, the control unit 90 may read the log information D1 from the storage unit 94 and display the log information D1 on the display 961. By checking the log information D1 on the display 961, the operator can check the image data IM1 and the monitoring result of the second monitoring processing. Therefore, the operator can determine whether or not the second monitoring processing is appropriate based on the log information D1.

[0082] The switching unit 903 switches the monitoring process used for the determination of the abnormal process. Initially, the first monitoring process before the version upgrade is used for the determination of the abnormal process. The switching unit 903 switches the monitoring process used for the determination of the abnormal process from the first monitoring process to the second monitoring process in response to, for example, the input of a switching instruction to the input device 97. That is to say, it can also be said that the switching unit 903 is an update unit that upgrades the monitoring process used for the determination of the execution of the abnormal process.

[0083] <An example of substrate processing> Next, an example of the substrate processing by the processing unit 1 will be described. FIG. 5 is a flowchart showing an example of the operation of the processing unit 1. Steps S1 to S6 show an example of the substrate processing for the substrate W. The control unit 90 causes the processing unit 1 to perform the operations from step S1 to step S6 based on a processing recipe indicating the processing procedure. The processing recipe is stored, for example, in the storage unit 94. Step S10 shows a monitoring operation for monitoring the object to be monitored. The monitoring operation will be described later.

[0084] First, the second transfer unit 122 transfers the substrate W into the processing unit 1 (step S1: transfer step). This substrate W is, for example, a substrate for a product for manufacturing a semiconductor device. Then, the substrate holding unit 2 holds the substrate W received from the second transfer unit 122. As a specific example, the substrate holding unit 2 displaces a plurality of chuck pins 22 from their respective release positions to the holding positions. Thereby, the plurality of chuck pins 22 hold the substrate W. The substrate holding unit 2 continues to hold the substrate W until the processing of the substrate W is completed.

[0085] Next, in each of steps S2 to S4, the discharge unit 3 discharges various processing liquids from the discharge port 31a toward the main surface of the substrate W, and causes the processing liquid to act on the main surface of the substrate W. More specifically, as described below, the processing unit 1 first performs a chemical liquid treatment for supplying a chemical liquid to the main surface of the substrate W (step S2: chemical liquid process), then performs a rinse treatment for supplying a rinse liquid such as pure water to the main surface of the substrate W (step S3: rinse process), and then performs an organic solvent treatment for supplying an organic solvent such as isopropyl alcohol to the main surface of the substrate W (step S4: organic solvent process).

[0086] In step S2, the guard lifting drive unit 71 raises the guard 7 to the upper position, the substrate holding unit 2 rotates the substrate W around the rotation axis Q1, and the nozzle movement drive unit 37 moves the nozzle 31A to the processing position. Then, the processing unit 1 discharges a chemical liquid from the nozzle 31A. The chemical liquid adhering to the main surface of the substrate W receives centrifugal force accompanying the rotation of the substrate W and flows radially outward, and scatters from the periphery of the substrate W. Thereby, the chemical liquid treatment is performed on the main surface of the substrate W. Then, when a predetermined chemical liquid time has elapsed, the control unit 90 stops the discharge of the chemical liquid from the nozzle 31A.

[0087] In step S3, the nozzle movement drive unit 37 moves the nozzle 31B to the processing position. This movement may be performed during the chemical liquid treatment. Then, the processing unit 1 discharges pure water as a rinse liquid from the nozzle 31B toward the main surface of the substrate W while rotating the substrate W. The pure water adhering to the main surface of the substrate W receives centrifugal force accompanying the rotation of the substrate W and flows radially outward, and scatters from the periphery of the substrate W. Thereby, the chemical liquid on the main surface of the substrate W can be washed away with pure water. For this reason, the chemical liquid on the main surface of the substrate W is replaced with pure water. Then, when a predetermined rinse time has elapsed, the control unit 90 stops the discharge of pure water from the nozzle 31B.

[0088] In step S4, the nozzle movement drive unit 37 moves the nozzle 31C to the processing position. This movement may be performed during the rinsing process. Then, while the processing unit 1 rotates the substrate W, the processing unit 1 discharges the organic solvent from the nozzle 31C toward the main surface of the substrate W. The organic solvent that has adhered to the main surface of the substrate W receives the centrifugal force associated with the rotation of the substrate W and flows radially outward, scattering from the periphery of the substrate W. As a result, the pure water on the main surface of the substrate W can be washed away by the organic solvent. For this reason, the pure water on the main surface of the substrate W is replaced by the organic solvent. The volatility of the organic solvent is higher than that of pure water. Then, when a predetermined organic solvent time has elapsed, the control unit 90 stops the rotation of the substrate W by the substrate holding unit 2 and stops the discharge of the organic solvent from the nozzle 31C. As a result, a liquid film of the organic solvent is formed on the main surface of the substrate W.

[0089] Next, the processing unit 1 performs a drying process on the substrate W (step S5: drying step). FIG. 6 is a diagram schematically showing an example of the state of the drying process. For example, first, the heater 8 heats the substrate W. As a result, the portion of the liquid film of the organic solvent on the substrate W that is in contact with the substrate W evaporates, and a vapor layer is formed between the liquid film and the substrate W (see FIG. 6(a)). Subsequently, the processing unit 1 discharges an inert gas from the nozzle 31C. The nozzle 31C discharges the inert gas along the vertical direction. As a result, the inert gas first collides with the liquid film near the center of the substrate W, and the liquid film flows radially outward due to the collision. As a result, an opening AP1 is formed near the center of the liquid film (see FIG. 6(b)), and the main surface of the substrate W is exposed at the opening AP1. That is, the exposed portion of the main surface of the substrate W dries. The opening AP1 ideally has a circular shape in plan view. Then, as the liquid film is pressed radially outward by the inert gas, the liquid film continues to move radially outward, and the opening AP1 (that is, the drying region) expands (see FIG. 6(c)), and finally the entire main surface of the substrate W dries.

[0090] When the entire main surface of the substrate W dries, the processing unit 1 stops discharging the inert gas from the nozzle 31C and the heating operation of the heater 8. Next, the nozzle movement drive unit 37 moves the nozzle 31 to the standby position, the guard lifting drive unit 71 lowers the guard 7 to the guard standby position, and the substrate holding unit 2 releases the holding of the substrate W.

[0091] Next, the second transfer unit 122 carries out the processed substrate W from the processing unit 1 (step S6: carry-out process).

[0092] As described above, by the appropriate operation of each part of the processing unit 1, the processing of the substrate W can be appropriately carried out. However, due to various abnormal occurrences, there may be cases where the processing of the substrate W cannot be appropriately carried out. For example, if the chuck pin 22 cannot operate properly, the substrate holding unit 2 cannot hold the substrate W appropriately, and if the guard lifting drive unit 71 cannot raise the guard 7 appropriately, the processing liquid cannot be received by the guard 7. If discharge abnormalities such as liquid breakage or dripping of the processing liquid from the discharge ports 31a of various nozzles 31 occur, the processing of the main surface of the substrate W may become non-uniform. Also, if an abnormality occurs in the shape of the opening AP1 of the liquid film on the main surface of the substrate W in the drying process in a plan view, it may cause poor drying of the substrate W. Therefore, the processing unit 1 performs a monitoring operation (step S10) in parallel with the substrate processing.

[0093] <Monitoring operation> FIG. 7 is a flowchart showing a first example of the monitoring operation. In the example of FIG. 7, the monitoring process includes steps S11 to S17. The series of processes from step S11 to step S17 are repeatedly executed until the monitoring operation ends.

[0094] As the object to be monitored, at least one of the substrate holding unit 2 (for example, chuck pin 22), nozzle 31, processing liquid, and guard 7 can be applied. Here, as an example, the case where the processing liquid is applied will be described. As a specific example, the control unit 90 monitors the shape of the opening AP1 of the liquid film of the organic solvent formed on the main surface of the substrate W in the drying process. First, the camera 5 images the object to be monitored in the chamber 10 (here, the liquid film of the organic solvent on the substrate W) and generates image data IM1 (step S11). This image data IM1 includes almost the entire main surface of the substrate W. FIG. 8 is a diagram schematically showing an example of the image data IM1. In the example of FIG. 8, the entire upper inner periphery (that is, the upper opening of the guard 7) of the guard 7 located at the upper position is included in the image data IM1. The upper inner periphery of the guard 7 has a circular shape in plan view, and since the imaging direction of the camera 5 is obliquely downward, in the image data IM1, the upper inner periphery of the guard 7 has an elliptical shape. The main surface of the substrate W is included in the upper opening of the guard 7 in the image data IM1. Also, in the example of FIG. 8, the opening AP1 of the liquid film on the main surface of the substrate W is also included. In the opening AP1, the main surface of the substrate W is exposed.

[0095] In FIG. 8, although not shown, the exposed portion of the main surface of the substrate W (that is, the portion within the opening AP1) in the image data IM1 includes a pattern. The pattern can include various patterns such as, for example, a wiring pattern, an insulating pattern, and a semiconductor pattern. In the image data IM1, the distribution of the pixel values (for example, luminance distribution) of these patterns is non-uniform.

[0096] Next, the first monitoring processing unit 901 of the control unit 90 performs the first monitoring process based on the image data IM1 (step S12). The first monitoring process is, for example, a process of determining whether the object to be monitored is normal based on the image data IM1. Here, the first monitoring processing unit 901 determines whether the opening AP1 is normal. For example, the first monitoring processing unit 901 first executes a spatial filter process on the image data IM1. Hereinafter, the image data IM1 after the spatial filter process is also referred to as the first filtered image data. An averaging filter process or a Gaussian filter process is used for the spatial filter process. By this filter process, the variation in the distribution of the pixel values of the image data IM1 is reduced. Therefore, the pattern on the main surface of the substrate W within the opening AP1 in the image data IM1 can be averaged. The degree of this averaging can be adjusted by the parameters of the spatial filter process. For example, the degree of averaging can be adjusted by the coefficients of the elements of the kernel matrix in the Gaussian filter. These parameters can be set in advance.

[0097] The control unit 90 determines whether the shape of the opening AP1 is normal based on the first filtered image data. A learned model can be used for this determination algorithm. For example, the control unit 90 uses a learned model such as deep learning to classify the first filtered image data into one of a normal category and an abnormal category. This learned model is generated by training based on a plurality of teacher image data including normal openings AP1 and a plurality of teacher image data including abnormal openings AP1. Each parameter of the generated learned model is stored in the storage unit 94. The parameters include, for example, the weighting coefficients in deep learning. The learned model can monitor the object to be monitored with relatively high monitoring accuracy.

[0098] When the first monitoring processing unit 901 uses a learned model, the learned model calculates, for example, a score value indicating whether the opening AP1 is normal. This score value is a value for classifying the first filtered image data into a normal category and an abnormal category. The first monitoring processing unit 901 determines that the opening AP1 is normal when the score value is within the range of the normal category, and determines that the opening AP1 is abnormal when the score value is within the range of the abnormal category. Note that the score value is also an example of the calculation result of the first monitoring process, and the classification result indicating the distinction between the normal category and the abnormal category is also an example of the calculation result of the first monitoring process.

[0099] Alternatively, as a determination algorithm, a rule-based algorithm may be adopted. For example, the first monitoring processing unit 901 may first generate a difference image between the first filtered image data generated in the immediately previous step S11 and the first filtered image data generated in the current step S11. When generating this difference image, pixels at the same position in both pieces of first filtered image data are subtracted from each other. Therefore, in the difference image, the pixel value becomes almost zero in the region having a common pixel value, and the pixel value takes a relatively large value only in the region having different pixel values. For this reason, the difference image mainly includes the opening AP1. Then, the first monitoring processing unit 901 binarizes the difference image to generate a binarized image. As a result, only the opening AP1 is included in the binarized image. Then, the first monitoring processing unit 901 performs edge extraction processing such as the Canny method on the binarized image to specify the contour shape of the opening AP1.

[0100] Next, the first monitoring processing unit 901 obtains an evaluation value that is an index of the size or shape of the opening AP1. Specifically, the first monitoring processing unit 901 may calculate the number of pixels (i.e., area) indicating the opening AP1 as the evaluation value of the size of the opening AP1, or may obtain the roundness as the evaluation value of the shape of the contour of the opening AP1. The first monitoring processing unit 901 may determine whether the evaluation value is within a predetermined normal range. The first monitoring processing unit 901 may determine that the opening AP1 is normal when the evaluation value is within the normal range, and may determine that the opening AP1 is abnormal when the evaluation value is outside the normal range. Note that the evaluation value is also an example of the calculation result of the first monitoring process, and the determination result indicating the distinction between normal and abnormal is also an example of the calculation result of the first monitoring process.

[0101] Here, when the environment in the chamber 10 is in the first environmental state, the monitoring accuracy of the first monitoring process is relatively low. For example, due to the presence of the pattern on the main surface of the substrate W, the monitoring accuracy of the first monitoring process may decrease. For example, in the drying process, the organic solvent evaporates, and the solvent vapor moves in the space above the substrate W. Therefore, this solvent vapor varies with the passage of time, and a part of the solvent vapor overlaps with the pattern in the opening AP1 in the image data IM1. Therefore, in a plurality of first filtered image data that are continuous in time series, the pattern in the opening AP1 sways with the passage of time due to the variation of the solvent vapor. As a result, the determination accuracy (i.e., monitoring accuracy) regarding the shape of the opening AP1 may decrease.

[0102] Now, as shown in FIG. 7, the control unit 90 performs not only the first monitoring process but also the second monitoring process (step S15). Specifically, the second monitoring processing unit 902 performs the second monitoring process based on the image data IM1. The second monitoring process is a process of monitoring the same monitoring object as the first monitoring process. Here, the second monitoring process is a process of monitoring the liquid film of the organic solvent on the substrate W based on the image data IM1.

[0103] The second monitoring processing unit 902 may first perform spatial filtering on the image data IM1 to generate second filtered image data. However, the parameters of this spatial filtering are different from those of the first monitoring processing. For example, the parameters of the spatial filtering in the second monitoring processing are set such that the degree of averaging in the spatial filtering of the second monitoring processing is higher than the degree of averaging in the first monitoring processing. Therefore, the pattern of the substrate W within the aperture AP1 of the second filtered image data is more averaged than the pattern of the substrate W within the aperture AP1 of the first filtered image data.

[0104] Based on the second filtered image data, the second monitoring processing unit 902 determines whether the shape of the aperture AP1 is normal. An example of this determination algorithm is the same as that of the first monitoring processing. When the second monitoring processing includes a learned model, the score value is also an example of the calculation result of the second monitoring processing, and the classification result indicating the distinction between the normal category and the abnormal category is also an example of the calculation result of the second monitoring processing. When the second monitoring processing includes a rule-based determination algorithm, the evaluation value is also an example of the calculation result of the second monitoring processing, and the determination result indicating the distinction between normal and abnormal is also an example of the calculation result of the second monitoring processing.

[0105] In the second monitoring processing, the pattern of the substrate W in the second filtered image data is more averaged. Therefore, the presence of the pattern becomes more ambiguous in the second filtered image data. Thus, a decrease in monitoring accuracy due to the presence of the pattern can be suppressed or avoided.

[0106] This second monitoring processing can be tested in the substrate processing apparatus 100 of the manufacturer. In this case, it is confirmed in advance that the monitoring accuracy of the second monitoring processing is higher than that of the first monitoring processing in the environment of the manufacturer. Alternatively, the second monitoring processing can be developed based on the image data IM1 passed from the user. In this case, it is confirmed in advance that the monitoring accuracy of the second monitoring processing is higher than that of the first monitoring processing in the environment indicated by the image data IM1.

[0107] However, in the substrate processing apparatus 100 installed in the user's factory, the environment inside the chamber 10 takes various environmental states. For example, when only some types of the substrates W used on the user side are passed to the manufacturer, it is difficult to test in the environmental state where the other types of substrates W are conveyed into the chamber 10. Also, for example, as various processing conditions on the user side such as the type and flow rate of the processing liquid and the heating temperature of the substrate W, it is also difficult to reproduce all of these conditions in the chamber 10. For this reason, the monitoring accuracy of the second monitoring process in all environments on the user side has not been confirmed. Similarly, since it is difficult for the manufacturer to obtain all the image data IM1 on the user side, the monitoring accuracy of the second monitoring process in all environments has not been confirmed.

[0108] In contrast, in the present embodiment, as described above, in the substrate processing apparatus 100 on the user side, the control unit 90 executes not only the first monitoring process before the version update but also the second monitoring process scheduled for version update.

[0109] Then, based on the calculation result of the first monitoring process before the version update, while determining whether abnormal processing needs to be executed (step S13: determination step), the control unit 90 stores the calculation result of the second monitoring process in the storage unit 94 without using it for determining whether abnormal processing needs to be executed (step S16: storage step).

[0110] When the monitoring result of the first monitoring process is normal, the control unit 90 continues the process without performing the abnormal processing described below (step S14). On the other hand, when the monitoring result of the first monitoring process is abnormal, the abnormal processing unit 904 performs abnormal processing (step S14: abnormal step). As abnormal processing, the abnormal processing unit 904 may, for example, cause the notification unit 96 to notify abnormal information indicating that an abnormality has occurred in the monitoring target. As a specific example, the abnormal processing unit 904 may display the abnormal information on the display 961 or output the abnormal information to the sound output unit 962. Alternatively, the abnormal processing unit 904 may interrupt the operation of the processing unit 1 as abnormal processing.

[0111] Next, the control unit 90 determines whether to end the monitoring operation (step S17). For example, the control unit 90 may determine to end the monitoring operation when the substrate W is carried out. When the monitoring operation has not been ended yet, the control unit 90 executes steps S11 to S16 again. When it is determined to end the monitoring operation, the control unit 90 ends the monitoring operation.

[0112] As described above, in the present embodiment, the control unit 90 of the substrate processing apparatus 100 on the user side performs both the first monitoring process before version update based on the image data IM1 and the second monitoring process scheduled for version update based on the image data IM1. Thereby, while the control unit 90 can perform abnormal processing using the first monitoring process, it can also perform a test of the second monitoring process in the actual installation environment. The test result is stored in the storage unit 94 as log information D1. That is, in the storage unit 94, a set of time-series image data IM1 and calculation results is stored as log information D1. Further, since the processing unit 1 sequentially processes the substrates W, the log information D1 stores the image data IM1 corresponding to a plurality of substrates W and the calculation results of the second monitoring process. The log information D1 includes the image data IM1 and the calculation results of the second monitoring process corresponding to various environmental states in the chamber 10 on the user side. The various environmental states include the second environmental state that is difficult to reproduce or confirm by the manufacturer.

[0113] <An example of the switching operation (update operation)> Next, an example of a switching operation (in other words, an update operation) for switching the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process will be described. FIG. 9 is a flowchart showing an example of the switching operation. In the example of FIG. 9, first, an operator inputs a display instruction for log information to the input device 97 (step S21). In response to the input of the display instruction, the control unit 90 reads the log information D1 from the storage unit 94 and causes the display 961 to display the log information D1 (step S22). For example, the display 961 displays a plurality of image data IM1 and a plurality of calculation results in association with each other. Note that the display 961 may display the image data IM1 and the calculation results in a list, or may display the image data IM1 and the calculation results one by one in time series in response to an input to the input device 97. The operator can visually recognize the image data IM1 and the calculation result at that time on the display 961 and determine whether there is a problem with the monitoring accuracy of the second monitoring process.

[0114] If the operator determines that there is no problem with the second monitoring process, the operator inputs a switching instruction (update instruction) to the input device 97 (step S23). In response to the switching instruction, the switching unit 903 of the control unit 90 switches the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process (step S24). For example, the switching unit 903 overwrites (copies) the parameters of the spatial filter process stored in the storage unit 94 from the value of the parameters in the first monitoring process to the value of the parameters in the second monitoring process. Thereby, the switching unit 903 can switch the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process. Thereafter, the control unit 90 determines whether it is necessary to execute the abnormality process using the second monitoring process, for example, in parallel with the substrate process.

[0115] As described above, according to the present embodiment, the control unit 90 switches the monitoring process from the first monitoring process to the second monitoring process based on the test result of the second monitoring process not only in the first environmental state but also in the second environmental state. Therefore, it is possible to more reliably improve the monitoring accuracy of the monitoring process after the switching.

[0116] This substrate processing apparatus 100 is particularly beneficial in the field of semiconductor device manufacturing equipment. This is because in this field, the transmission of information from the user to the outside can be severely restricted. For this reason, the manufacturer may not be able to obtain a sufficient variety of substrates W, and may also not be able to obtain sufficient information such as processing conditions. In addition, the manufacturer may not be able to obtain a sufficient amount of image data IM1. That is, the manufacturer is in a situation where it cannot perform a sufficient test of the second monitoring process. The substrate processing apparatus 100 according to the present embodiment can test the second monitoring process in various environments on the user side while minimizing the information transmitted from the user to the outside.

[0117] Also, in the above example, the log information D1 includes the image data IM1 as well. Therefore, the log information D1 of the second monitoring process can be enriched. Then, the display 961 displays both the image data IM1 and the calculation result of the second monitoring process as the log information D1. For this reason, the operator can determine the monitoring accuracy of the second monitoring process using the image data IM1 as well.

[0118] On the other hand, the log information D1 does not necessarily have to include the image data IM1. For example, the display 961 may display the calculation result of the second monitoring process as the log information D1 without displaying the image data IM1. Specifically, the display 961 displays a score value or an evaluation value as the calculation result of the second monitoring process. Now, if these values of the log information D1 are concentrated near the threshold values for determining (classifying) normal and abnormal, the determination of the second monitoring process becomes unstable, so the accuracy of the second monitoring process cannot be said to be sufficient. Here, the threshold value mentioned here is, for example, a threshold value for distinguishing normal and abnormal to be compared with a score value or an evaluation value. On the other hand, if these values are sufficiently distributed away from the threshold value by a predetermined margin or more, the second monitoring process can determine normal and abnormal with high accuracy. Conversely, the operator can judge the validity of the second monitoring process based on the calculation result (score value or evaluation value) of the second monitoring process of the log information D1 displayed on the display 961 and the threshold value. Specifically, the operator can judge that the monitoring accuracy of the second monitoring process is sufficient when the calculation result group of the log information D1 is distributed away from the threshold value by a predetermined margin or more. From this perspective, the display 961 may also display the threshold value together with the log information D1.

[0119] <Log information of the first monitoring process> FIG. 10 is a flowchart showing a second example of the monitoring operation. In the example of FIG. 10, the monitoring operation includes not only steps S11 to S17 but also step S18. Step S18 is executed, for example, between step S12 and step S13. In step S18, the control unit 90 also stores the calculation result of the first monitoring process in the storage unit 94. Thereby, the control unit 90 can associate the image data IM1, the calculation result of the first monitoring process using the image data IM1, and the calculation result of the second monitoring process with each other and store them in the storage unit 94 as the log information D1. That is, the log information D1 includes the corresponding image data IM1, the calculation result of the first monitoring process, and the calculation result of the second monitoring process.

[0120] Although the switching operation is the same as that in FIG. 9, in step S22, the control unit 90 causes the display 961 to display the image data IM1, the calculation result of the first monitoring process, and the calculation result of the second monitoring process. As a result, the operator can confirm not only the calculation result of the second monitoring process corresponding to the image data IM1 but also the calculation result of the first monitoring process corresponding to the image data IM1. Therefore, the operator can determine the validity of the calculation result of the second monitoring process with reference to the calculation result of the first monitoring process as well.

[0121] The control unit 90 may cause the display 961 to display only the data in the log information D1 in which the calculation result of the first monitoring process and the calculation result of the second monitoring process are different. Thereby, the work of the operator to confirm the validity of the second monitoring process can be made easier.

[0122] <Automatic determination> The control unit 90 may determine whether to switch the monitoring process used for the determination of the execution of the abnormal process from the first monitoring process to the second monitoring process based on the log information D1. In other words, the control unit 90 may automatically determine the monitoring accuracy of the second monitoring process. As a specific example, the control unit 90 may compare the calculation result of the first monitoring process and the calculation result of the second monitoring process in the log information D1.

[0123] For example, in the monitoring operation, the control unit 90 may also include the environmental state of the chamber 10 in the log information D1 and store it in the storage unit 94. The environmental state of this chamber 10 depends on the processing content at the time of imaging by the camera 5. The processing content includes various information such as, for example, the type of the substrate W to be processed, the type and flow rate of the processing liquid, the temperature of the heater 8, and the rotation speed of the substrate W. Conversely, the processing content can be utilized as information indicating the environmental state. Therefore, in steps S16 and S18 of FIG. 10, the control unit 90 may store, for example, the processing content in the storage unit 94 as the environmental state. Thereby, the environmental state at the time of imaging of the image data IM1 and the calculation results of the first monitoring process and the second monitoring process using the image data IM1 are stored in the storage unit 94 as log information.

[0124] Now, when the chamber 10 is in the first environmental state, the monitoring accuracy of the second monitoring process is equal to or higher than that of the first monitoring process. Therefore, when the chamber 10 is in the second environmental state other than the first environmental state, if the calculation result of the second monitoring process is comparable to that of the first monitoring process, it can be said that the monitoring accuracy of the second monitoring process is sufficient.

[0125] FIG. 11 is a flowchart showing a second example of the switching operation. The control unit 90 reads out the calculation results of the first monitoring process and the second monitoring process in the second environmental state other than the first environmental state from the log information D1 (step S31). For example, the first environmental state (e.g., processing content) may be stored in the storage unit 94 as prior information. The control unit 90 grasps the information of the first environmental state from the prior information and reads out the data other than this first environmental state from the log information D1.

[0126] Next, the control unit 90 compares the calculation results of the first monitoring process and the second monitoring process corresponding to each other in the second environmental state (step S32). For example, when the calculation result is indicated by a normal / abnormal determination result, it is compared whether the calculation results of the corresponding first monitoring process and second monitoring process are the same as each other.

[0127] Based on the comparison result, the control unit 90 determines whether the monitoring accuracy of the second monitoring process is sufficient (step S33). For example, in all pairs of the first monitoring process and the second monitoring process in the second environmental state in the log information D1, when the calculation results (e.g., normal and abnormal) of the first monitoring process and the second monitoring process are the same as each other, the control unit 90 determines that the monitoring accuracy of the second monitoring process is sufficient, and switches the monitoring process from the first monitoring process to the second monitoring process (step S34).

[0128] On the other hand, when the control unit 90 determines that the second monitoring accuracy is not sufficient, the control unit 90 causes the notification unit 96 to notify that the monitoring accuracy of the second monitoring process is insufficient (step S35). For example, the control unit 90 may display the calculation result of the first monitoring process, the calculation result of the second monitoring process, and the image data IM1 when the monitoring results are different on the display 961.

[0129] As described above, the control unit 90 compares the calculation results of the first monitoring process and the second monitoring process in the second environmental state (step S32). That is, for the first environmental state that has been confirmed, the control unit 90 does not compare the calculation results of the first monitoring process and the second monitoring process. In the unconfirmed second environmental state, the control unit 90 compares the calculation results of the first monitoring process and the second monitoring process. Therefore, the control unit 90 can appropriately determine the validity of the monitoring accuracy of the second monitoring process. When the second monitoring accuracy is appropriate, the control unit 90 automatically switches the first monitoring process to the second monitoring process. Thus, the burden on the operator can be reduced.

[0130] On the other hand, when the second monitoring accuracy is not appropriate, the control unit 90 displays the calculation results and the image data IM1 of the first monitoring process and the second monitoring process, which are different from each other, on the display 961. Therefore, the operator can grasp the environment in the chamber 10 where the monitoring accuracy of the second monitoring process is insufficient based on the image data IM1. Thus, it is easy for the operator to improve the second monitoring process.

[0131] Note that in the above example, the control unit 90 compares the calculation results of the first monitoring process and the second monitoring process, but it is not necessarily limited to this. For example, the control unit 90 may determine whether to switch the monitoring process used for determining the execution of the abnormal process from the first monitoring process to the second monitoring process based on the score value or the evaluation value, which is the calculation result of the second monitoring process in the log information D1. For example, the control unit 90 determines whether the calculation result group (score value or evaluation value) of the log information D1 is distributed away from the threshold by a predetermined margin or more, and when the calculation result group is distributed away from the threshold by a predetermined margin or more, the control unit 90 may switch the monitoring process from the first monitoring process to the second monitoring process.

[0132] <Monitoring Process Algorithm> In the above specific example, the algorithms of the first monitoring process and the second monitoring process are the same as each other, and the parameters used in the algorithms of the first monitoring process and the second monitoring process are different from each other. Also, in the above specific example, the parameter was a parameter of the spatial filtering process. However, it is not necessarily limited to this. For example, the parameter may be a threshold value of the binarization process, a parameter of other processes, or a value within a normal range to be compared with an evaluation value.

[0133] Also, when the first monitoring process and the second monitoring process are learned models, the parameter may be a parameter of the learned model. For example, the parameter may be various parameters (such as weighting coefficients) in deep learning. The parameter of the first monitoring process may be a parameter generated by training with the first teacher image data group, and the parameter of the second monitoring process may be a parameter generated by training with the second teacher image data group. At least a part of the second teacher image data group is different from the first teacher image data group. Specifically, the first teacher image data group does not include image data of the first environmental state, and the second teacher image data group includes image data of the first environmental state. For this reason, in the first environmental state, the monitoring accuracy of the second monitoring process is higher than that of the first monitoring process.

[0134] These parameters may be stored in the storage unit 94. That is, the parameter of the first monitoring process and the parameter of the second monitoring process may be stored in the storage unit 94. The control unit 90 may switch the first monitoring process to the second monitoring process by overwriting (or copying) the parameter of the first monitoring process with the parameter of the second monitoring process. According to this, the control unit 90 can easily switch from the first monitoring process to the second monitoring process.

[0135] The algorithms of the first monitoring process and the second monitoring process may be different from each other. For example, the algorithm of the first monitoring process may be a rule-based algorithm, and the algorithm of the second monitoring process may be an algorithm of a learned model. Since the algorithms are different, the monitoring accuracy of the second monitoring process can be improved more than that of the first monitoring process.

[0136] These algorithms may be stored in the storage unit 94 as a program, for example. The control unit 90 may switch the first monitoring process to the second monitoring process by overwriting (or copying) the program of the first monitoring process with the program of the second monitoring process.

[0137] As described above, the substrate processing apparatus 100 and the substrate processing method have been described in detail. However, the above descriptions are illustrative in all aspects, and this disclosure is not limited thereto. Also, the various modifications described above can be applied in combination as long as they do not contradict each other. And a number of modifications not illustrated can be assumed without departing from the scope of this disclosure.

[0138] For example, in the above specific example, although the substrate processing apparatus 100 performs wet processing using a processing liquid on the substrate W, dry processing may be performed. Also, although the substrate processing apparatus 100 is a single-wafer processing apparatus, it may be a batch processing apparatus.

Description of Reference Numerals

[0139] 10 Chamber 100 Substrate Processing Apparatus 3 Discharge Unit 5 Camera 90 Control Unit 94 Storage Unit 961 Display 97 Input Device D1 Log Information IM1 Image Data W Substrate

Claims

1. A chamber into which a substrate is loaded, a discharge unit having a discharge port for discharging a fluid in the chamber and causing the fluid to act on the substrate, a camera that images a monitoring target in the chamber and generates image data, a control unit that controls the discharge unit to process the substrate loaded into the chamber and causes the camera to image the monitoring target, a storage unit and wherein the control unit monitors the monitoring target by a first monitoring process based on the image data, monitors the monitoring target by a second monitoring process different from the first monitoring process based on the image data, determines whether to execute an abnormality process using the first monitoring process without using the second monitoring process, and stores log information including the calculation result of the second monitoring process in the storage unit, a substrate processing apparatus.

2. The substrate processing apparatus according to claim 1, wherein the log information includes the image data, a substrate processing apparatus.

3. The substrate processing apparatus according to claim 1 or claim 2, further comprising a display for displaying the log information, and an input device that receives an input of a switching instruction for switching the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process and wherein the control unit switches the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process in response to the input of the switching instruction, a substrate processing apparatus.

4. The substrate processing apparatus according to claim 3, wherein the storage unit stores the log information including the calculation result of the first monitoring process, and the display also displays the calculation result of the first monitoring process, a substrate processing apparatus.

5. The substrate processing apparatus according to claim 1 or claim 2, wherein the control unit determines whether to switch the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process based on the log information, a substrate processing apparatus.

6. The substrate processing apparatus according to claim 5, wherein the monitoring accuracy of the second monitoring process when the environment in the chamber is in a first environmental state is higher than the monitoring accuracy of the first monitoring process when the environment in the chamber is in the first environmental state, The control unit determines whether to switch the monitoring process used for determining the execution of the abnormality process from the first monitoring process to the second monitoring process based on a comparison between the calculation result of the first monitoring process in an environmental state other than the first environmental state among the log information and the calculation result of the second monitoring process, a substrate processing apparatus.

7. The substrate processing apparatus according to claim 1 or claim 2, wherein algorithms of the first monitoring process and the second monitoring process are different from each other, a substrate processing apparatus.

8. The substrate processing apparatus according to claim 1 or claim 2, wherein algorithms of the first monitoring process and the second monitoring process are the same as each other, and parameters used in the algorithms are different between the first monitoring process and the second monitoring process, a substrate processing apparatus.

9. The substrate processing apparatus according to claim 8, wherein the storage unit stores the parameters of the first monitoring process and the parameters of the second monitoring process in the storage unit, and the control unit copies the value of the parameters of the second monitoring process to the value of the parameters of the first monitoring process, a substrate processing apparatus.

10. The substrate processing apparatus according to claim 1 or claim 2, wherein either one of the first monitoring process and the second monitoring process is a process using a learned model, a substrate processing apparatus.

11. A step of loading a substrate into a chamber; A step of causing a fluid to act on the substrate in the chamber; A step of a camera imaging a monitoring target in the chamber to generate image data; Monitoring the monitoring target by a first monitoring process based on the image data, monitoring the monitoring target by a second monitoring process different from the first monitoring process based on the image data, determining whether to execute an abnormality process using the first monitoring process without using the second monitoring process, and storing log information including the calculation result of the second monitoring process in a storage unit A substrate processing method comprising the steps.

Images