TWI737335B - Detecting method of gas-liquid interface in nozzle and substrate processing apparatus - Google Patents

Abstract

The present invention provides a technology capable of accurately detecting the gas-liquid interface between the processing liquid and the gas in the internal flow path of the nozzle. The detection method of the present invention detects the gas-liquid interface between the processing liquid and the gas in the internal flow path of the nozzle, and is equipped with a photographing process and a detection process. In the imaging process, the camera captures the imaging area including the transparent nozzle, which ejects the treatment liquid from the ejection port, which is the port in front of the internal flow path. In the detection process, the gas-liquid interface in the internal flow path is detected based on the difference between the two images in which the presence and distribution of the processing liquid and the gas in the internal flow path are different.

Description

Method for detecting gas-liquid interface inside nozzle and substrate processing device

This case is about a method for detecting the gas-liquid interface inside the nozzle and a substrate processing device.

Patent Document 1 describes a substrate processing apparatus that applies a resist to a substrate. The substrate processing apparatus includes a substrate holding portion for holding a substrate, a spray nozzle for spraying a processing liquid (resist) to the substrate, and a camera that photographs the vicinity of the spray outlet of the spray nozzle as a resist detection area. In Patent Document 1, the resist detection area is photographed by a camera, and the obtained image is used to detect the time when the resist ejection starts and when the resist ejection ends. Specifically, based on the image acquired by the camera, it is determined whether or not the ejection nozzle ejects the resist, and based on the determination result, the start time of the resist ejection and the end time of the resist ejection are detected. In Patent Document 1, the resist ejection period is acquired based on the time when the resist ejection starts and when the resist ejection ends. When the resist ejection period falls outside a specific allowable range, it is determined that a resist ejection failure has occurred. , To issue a warning to the operator of the substrate processing device. Therefore, the operator can prevent the substrates that are likely to be processed poorly from flowing out to the subsequent process. [Prior Technical Literature] [Patent Literature]

[Patent Document 1] Japanese Patent Laid-Open No. 2003-273003

[The problem to be solved by the invention]

In Patent Document 1, it is determined whether the nozzle ejects resist, but there is no description about the state of the processing liquid flowing in the flow path inside the nozzle.

For example, in a normal process liquid ejection, the internal flow path of the nozzle becomes a liquid-tight state filled with the process liquid. However, if air bubbles are mixed in the internal flow path of the nozzle, the air bubbles will be released from the ejection port together with the processing liquid. At this time, there is a case where the treatment liquid splashes from the nozzle outlet.

Moreover, when the nozzle stops spraying, there is a case where the suction process is performed, that is, the internal flow path of the nozzle becomes a negative pressure and a certain amount of the treatment liquid is sucked from the front end face of the nozzle. If the position of the front end surface (the suck-back position) is outside the specific reference range after the suck-back process, there is a possibility that a discharge failure may occur when the treatment liquid is discharged next time.

In addition, there may also be cases where air bubbles are mixed in the internal flow path of the nozzle due to the failure to properly perform the suction treatment. In this case, the ejection failure may occur when the treatment liquid is ejected next time.

Therefore, it is expected that the state of the internal flow path of the nozzle can be monitored. More specifically, it is expected to be able to detect at least one of the suction position of the treatment liquid and the air bubbles. In other words, it is expected that the interface between the processing liquid and the gas in the internal flow path of the nozzle can be detected.

Therefore, the purpose of this case is to provide a technology capable of accurately detecting the gas-liquid interface between the processing liquid and the gas in the internal flow path of the nozzle. [Technical means to solve the problem]

The first aspect of the detection method of the gas-liquid interface inside the nozzle is the detection method of the treatment liquid and the gas-liquid interface of the gas in the internal flow path of the nozzle, which includes: a photographing process, which is photographed by a camera including the transparent nozzle The image is captured in the imaging area, the nozzle ejects the processing liquid from the outlet orifice before the internal flow path; and the detection process is based on the difference in the presence and distribution of the processing liquid and the gas in the internal flow path 2 The difference between the above-mentioned images is used to detect the above-mentioned gas-liquid interface in the above-mentioned internal flow path.

The second aspect of the detection method for the gas-liquid interface inside the nozzle is the detection method for the gas-liquid interface inside the nozzle as in the first aspect. Perform edge extraction processing on the image to obtain a first edge image and a second edge image, respectively; an edge expansion step, which performs an edge expansion process to expand the edge of the second edge image to obtain an edge expansion image; and A step of acquiring a difference image representing the difference between the first edge image and the edge expansion image, and detecting the gas-liquid interface based on the edges of the difference image excluding the edge from the edge expansion image .

The third aspect of the detection method of the gas-liquid interface inside the nozzle is the same as the second aspect of the detection method of the gas-liquid interface inside the nozzle. For example, the processing period is from a state in which the processing liquid is stationary in the internal flow path of the nozzle to the processing liquid being stopped again after the discharge of the processing liquid is completed. In the detection step, based on at least the processing period A part of the two above-mentioned images taken to detect the position of the above-mentioned gas-liquid interface.

The fourth aspect of the detection method of the gas-liquid interface inside the nozzle is the detection method of the gas-liquid interface inside the nozzle as in the third aspect. A first image is acquired when stationary, and a second image is acquired when the processing liquid flows toward the ejection port. In the edge extraction step, the first edge image is acquired based on the first image, based on the second image. Like to obtain the above-mentioned second edge image.

The fifth aspect of the detection method of the gas-liquid interface inside the nozzle is like the fourth aspect of the detection method of the gas-liquid interface inside the nozzle, which further includes the following steps: in the above-mentioned detection process, when based on the above-mentioned first image And when the second image detects a plurality of positions of the gas-liquid interface, it is determined that an abnormality has occurred.

The sixth aspect of the detection method of the gas-liquid interface inside the nozzle is the detection method of the gas-liquid interface inside the nozzle of any one of the third to the fifth aspects. During the ejection period during which the processing liquid is ejected from the nozzle of the nozzle, the images are sequentially acquired, and in the edge extraction step, the first edge image is acquired based on the current image acquired by the camera during the ejection period And acquiring the second edge image based on the image acquired by the camera most recently before the current image during the ejection period.

The seventh aspect of the detection method of the gas-liquid interface inside the nozzle is like the sixth aspect of the detection method of the gas-liquid interface inside the nozzle, which further has the following steps: in the above-mentioned detection process, when based on the above-mentioned ejection period When the gas-liquid interface is detected in the two captured images, it is determined that an abnormality has occurred.

The aspect of the substrate processing apparatus includes: a substrate holding portion that holds the substrate; a transparent nozzle that ejects the processing liquid from the port in front of the internal flow path, that is, to the substrate held by the substrate holding portion; Capturing an image in the imaging area; and an image processing unit that detects the processing in the internal flow path based on the difference between the two images in which the processing liquid and the gas in the internal flow path are different in their distribution states The gas-liquid interface between liquid and the above gas. [Effects of Invention]

According to the first aspect of the detection method of the gas-liquid interface inside the nozzle and the aspect of the substrate processing apparatus, the gas-liquid interface inside the nozzle is detected based on the image taken by the camera, so the gas-liquid interface can be detected with high accuracy.

According to the second aspect of the detection method of the gas-liquid interface inside the nozzle, by the difference between the first edge image and the edge expansion image, the edges of the first edge image except those representing the gas-liquid interface Be eliminated. That is, in the first edge image, only the edge representing the gas-liquid interface remains in the difference image. Conversely, even if there is a positional displacement of the nozzles between the images, in the differential image, it is possible to eliminate the edges of the first edge image other than the edges representing the gas-liquid interface.

On the other hand, the edge of the expanded edge image remains in the differential image, but the position of the gas-liquid interface is detected based on the edges other than the edge from the expanded edge image.

As described above, by excluding the influence of the positional displacement of the nozzle between the images and the influence of the edge of the edge expansion image, it is possible to detect the gas-liquid interface in the first edge image. As a result, the gas-liquid interface can be detected with high accuracy.

According to the third aspect of the detection method of the gas-liquid interface inside the nozzle, since the two images acquired by the camera during the processing period are used, even if the aging factors such as damage to the nozzle are generated, the difference image will be Eliminate the influence of the aging factor. As a result, the gas-liquid interface can be detected with higher accuracy.

According to the fourth aspect of the detection method of the gas-liquid interface inside the nozzle, the position of the gas-liquid interface in the first image is detected. The position of the front end surface of the processing liquid at rest includes the suck-back position, so the suck-back position can be detected.

According to the fifth aspect of the detection method of the gas-liquid interface inside the nozzle, the abnormality of the flow path inside the nozzle before ejection can be detected.

According to the sixth aspect of the detection method of the gas-liquid interface inside the nozzle, the difference between the edge expansion image based on the latest image and the first edge image based on the current image is used to detect the difference in the current image Air-liquid interface. Furthermore, when air bubbles are generated in the internal flow path of the nozzle, and the air bubbles move to the shooting area of the camera during the process of spraying the processing liquid, the current image contains air bubbles in the initial stage, and the latest image does not Contains bubbles. According to the sixth aspect, the position of the air-liquid interface based on the bubble is detected at the time when the current image contains the bubble, so that the bubble can be detected quickly.

According to the seventh aspect of the detection method of the gas-liquid interface inside the nozzle, it is possible to detect the abnormality of the flow path inside the nozzle during the ejection process.

Hereinafter, the embodiments will be described with reference to the accompanying drawings. In addition, the drawings are schematic diagrams, and for convenience of description, the configuration is omitted or simplified as appropriate. In addition, the relationship between the size and position of the structure shown in the drawings may not be accurately recorded, and may be changed as appropriate.

In addition, in the description shown below, the same constituent elements are labeled with the same symbols and illustrated, and the names and functions thereof are the same. Therefore, in order to avoid repetition, detailed descriptions of the same constituent elements are sometimes omitted.

<Schematic structure of substrate processing equipment> Fig. 1 is a plan view schematically showing an example of the structure of a substrate processing apparatus. The substrate processing apparatus of FIG. 1 is an apparatus that forms a resist film or the like on a substrate (for example, a semiconductor wafer) W, and develops the exposed substrate W.

In the example of FIG. 1, the substrate processing apparatus includes a transfer unit 110, a processing unit 120, an interface surface 130, and a control unit 140. The control unit 140 controls various configurations of the substrate processing apparatus.

The control unit 140 is an electronic circuit, for example, it may also have a data processing device and a storage medium. The data processing device may also be an arithmetic processing device such as a CPU (Central Processor Unit, central processing unit), for example. The storage medium may also have a non-transitory storage medium (such as ROM (Read Only Memory) or hard disk) and a temporary storage medium (such as RAM (Random Access Memory)). The non-transitory storage medium may store, for example, a program that specifies the processing to be executed by the control unit 140. When the program is executed by the processing device, the control unit 140 can execute the processing specified by the program. Of course, part or all of the processing executed by the control unit 140 may also be executed by hardware.

On both sides of the processing portion 120, the transfer and transfer portion 110 and the interface surface 130 are adjacently arranged. An exposure machine EXP is further arranged adjacent to the interface surface 130, and the exposure machine EXP is an external device separate from the main device.

The transfer conveying unit 110 includes a plurality of (four in the figure) cassette placement table 111 and an ID (Identification) transport mechanism TID. A plurality of cassette placement tables 111 are arranged in a row, and one cassette C is placed on each cassette placement table 111.

The transport mechanism TID for ID can be horizontally moved along the arrangement direction of the cassettes C on the side of the cassette mounting table 111, and can be stopped at a position opposite to each cassette C. The ID transfer mechanism TID includes a holding arm, and transfers the substrate W to each cassette C and each processing unit 120. The ID transport mechanism TID takes out the substrate W from the cassette C and transports it to the processing unit 120, and stores the substrate W received from the processing unit 120 in the cassette C.

The processing unit 120 processes the substrate W. In the example of FIG. 1, the processing unit 120 is divided into cells 121 and 122. The partition 121 includes the main transport mechanism T1, and the partition 122 includes the main transport mechanism T2. A plurality of processing units are respectively provided in the partitions 121 and 122. In the example of FIG. 1, only the partitions 121 and 122 are shown, but in the processing unit 120, a plurality of partitions 121 and 122 may be provided along the vertical direction. That is, it is possible to stack the same subregion with the subregion 121 on the subregion 121, and also stack the subregion with the same layer as the subregion 122 on the subregion 122. In short, the processing unit 120 may also have multiple levels of structure. In the partition 121 (and the partition of the upper step), a resist film or the like is formed on the substrate W, and the substrate W is developed in the partition 122 (and the partition of the upper step).

The partitions 121 and 122 are arranged horizontally and connected to each other to form a substrate processing line that connects the transfer portion 110 and the interface portion 130. It is the same in all levels. The substrate processing lines are arranged substantially in parallel in the vertical direction. In other words, the processing unit 120 is composed of a substrate processing line with a hierarchical structure.

The interface surface 130 is arranged between the processing unit 120 and the exposure machine EXP, and the substrate W is relayed between them.

Hereinafter, in order to simplify the description, the description of the upper-level partitions will be omitted, and the partitions 121 and 122 will be described. In the partition 121, a transfer space A1 for transferring the substrate W is formed. The transport space A1 passes through the center of the partition 121 and is formed in a strip shape parallel to the arrangement direction of the partitions 121 and 122. The processing unit of the partition 121 includes a coating processing unit 123 for coating the substrate W with a processing liquid, and a heat treatment unit 124 for thermally treating the substrate W. The coating processing unit 123 is arranged on one side with respect to the conveying space A1, and the heat treatment unit 124 is arranged on the other side.

A plurality of coating processing units 123 are arranged side by side so as to face the conveying space A1, respectively. In this embodiment, a plurality of coating processing units 123 are also arranged side by side along the vertical direction. For example, a total of 4 coating processing units 123 are arranged in 2 rows and 2 stages. The coating processing unit 123 includes a coating processing unit for an anti-reflection film that performs processing of forming an anti-reflection film on the substrate W, and a coating processing unit for a resist film that performs processing of forming a resist film on the substrate W. For example, the two coating processing units 123 in the lower stage are formed on the substrate W to form an anti-reflection film, and the two coating processing units 123 in the upper stage are formed on the substrate W to form a resist film.

A plurality of heat treatment units 124 are arranged side by side so as to face the conveying space A1, respectively. In this embodiment, a plurality of heat treatment units 124 are also arranged side by side along the vertical direction. For example, three heat treatment units 124 can be arranged in the horizontal direction, and five heat treatment units 124 can be stacked in the vertical direction. The heat treatment units 124 each include a flat plate 125 on which the substrate W is placed, and the like. The heat treatment unit 124 includes a cooling unit that cools the substrate W, a heating and cooling unit that continuously performs heating and cooling treatments, and is performed in a vapor atmosphere of hexamethylsilazane (HMDS) in order to improve the adhesion between the substrate W and the coating. Adhesion treatment unit for heat treatment. Furthermore, the heating and cooling unit has two flat plates 125, and includes a partial transport mechanism (not shown) that moves the substrate W between the two flat plates 125. There are a plurality of various heat treatment units, and they are arranged in appropriate positions.

A placement part PASS1 is provided at the boundary between the transfer portion 110 and the partition 121, and a placement part PASS2 is provided at the boundary between the partitions 121 and 122. The placing part PASS1 relays the substrate W between the transfer conveying part 110 and the partition 121, and the placing part PASS2 relays the substrate W between the partitions 121 and 122. The placement parts PASS1 and PASS2 include a plurality of support pins that support the substrate W in a horizontal posture. The horizontal posture mentioned here refers to the posture in which the thickness direction of the substrate W is along the vertical direction. The mounting part PASS1 can mount two substrates W, for example. The placing portion PASS1 has, for example, a two-stage structure, and one substrate W is placed in each stage. In one stage, the substrate W that is transferred from the transfer and transfer unit 110 to the partition 121 is placed, and the other stage is the substrate W that is transferred from the partition 121 to the transfer and transfer part 110. The placement part PASS2 also has a two-stage structure similarly.

A main conveyance mechanism T1 is provided in the approximate center of the conveyance space A1. The main transport mechanism T1 and the processing unit of the partition 121, the placement part PASS1, and the placement part PASS2 perform the transfer of the substrate W, respectively. In the example of FIG. 1, the main transport mechanism T1 includes two holding arms H1 and H2. Thus, the main transport mechanism T1 can use one of the holding arms H1 to take out the substrate W from the target portion (for example, the processing unit of the partition 121), and use the other holding arm H2 to deliver the other substrate W to the target portion.

A transfer space A2 for transferring the substrate W is formed in the partition 122. The conveying space A2 is formed to be on the extension line of the conveying space A1.

The processing unit of the partition 122 includes a coating processing unit 127 for applying a processing liquid to a substrate, a thermal processing unit 126 for thermally treating the substrate W, and an edge exposure unit (not shown) for exposing the peripheral portion of the substrate W. The coating processing unit 127 is arranged on one side with respect to the conveying space A2, and the heat treatment unit 126 and the edge exposure unit are arranged on the other side. Here, the coating processing unit 127 is preferably arranged on the same side as the coating processing unit 123. In addition, the heat treatment unit 126 and the edge exposure unit are preferably in the same row as the heat treatment unit 124.

In this embodiment, a plurality of coating processing units 127 are also arranged side by side along the vertical direction. For example, a total of 6 coating processing units 127 are arranged in 3 rows and 2 stages. The coating processing unit 127 includes a development processing unit for developing the substrate W, and a resist coating processing unit for forming a resist coating film on the substrate W. For example, the three coating processing units 127 in the lower stage form a resist coating film on the substrate W, and the three coating processing units 127 in the upper stage develop the substrate W.

A plurality of heat treatment units 126 are arranged side by side along the lateral direction of the conveying space A2, and a plurality of heat treatment units 126 are stacked in the vertical direction. The heat treatment unit 126 includes a heating unit that heats the substrate W and a cooling unit that cools the substrate W.

The edge exposure unit is a separate body and is set at a specific position. The edge exposure unit includes a rotation holding part (not shown) that rotatably holds the substrate W, and a light irradiation part (not shown) that exposes the periphery of the substrate W held by the rotation holding part.

A placing and buffering portion P-BF is provided at the boundary between the partition 122 and the interface surface 130. On the placing and buffering part P-BF, the substrate W transported from the partition 122 to the interface surface 130 is placed.

The main conveyance mechanism T2 is installed in the approximate center of the conveyance space A2 in a plan view. The main transport mechanism T2 has the same configuration as the main transport mechanism T1. In addition, the main transport mechanism T2 transfers the substrate W to each of the placing section PASS2, the coating processing unit 127, the heat treatment unit 126, the edge exposure unit, and the placing and buffering section P-BF.

The inner surface 130 includes a washing processing block 131 and a carry-in and carry-out block 132. A placement part PASS3 is provided at the boundary between the cleaning processing block 131 and the carry-in and carry-out block 132. An example of the configuration of the placement section PASS3 is the same as that of the placement sections PASS1 and PASS2. A placement and cooling unit (not shown) is installed on the upper or lower side of the placement part PASS3. The mounting and cooling unit cools the substrate W to a temperature suitable for exposure.

The transport mechanism TIF for IF is installed in the carry-in and carry-out block 132. The IF transport mechanism TIF transports the substrate W from the placement and cooling unit to the loading section LPa of the exposure machine EXP, and transports the substrate W from the transport section LPb of the exposure machine EXP to the placement section PASS3.

The washing processing block 131 includes two washing processing units 133a, 133b, and two conveying mechanisms T3a, T3b. The two washing processing units 133a, 133b are arranged so as to sandwich a set of conveying mechanisms T3a, T3b. The cleaning processing unit 133a cleans and dries the substrate W before exposure. A plurality of washing processing units 133a may also be stacked in multiple stages. The transport mechanism T3a transports the substrate W from the placement and buffering section P-BF to the cleaning processing unit 133a, and transports the cleaned substrate W from the cleaning processing unit 133a to the placement and cooling unit.

The cleaning processing unit 133b cleans and dries the exposed substrate W. A plurality of washing processing units 133b may also be stacked in multiple stages. The transport mechanism T3b transports the substrate W from the placement section PASS3 to the cleaning and drying processing unit 133b, and transports the cleaned substrate W from the cleaning and drying processing unit 133b to the placement and buffering section P-BF.

In this substrate processing system, the substrate W is processed according to the following method. That is, the substrate W taken out from the cassette C is cooled by the cooling unit of the partition 121. The cooled substrate W is subjected to the coating treatment of the coating treatment unit for the anti-reflection film of the partition 121. Thereby, an anti-reflection film is formed on the surface of the substrate W. The substrate W on which the anti-reflection film is formed is heated by the heating and cooling unit and then cooled. The cooled substrate W is subjected to coating processing by a resist film coating processing unit. Thereby, a resist film is formed on the surface of the substrate W. The substrate W on which the resist film is formed is heated again by the heating and cooling unit and then cooled. The substrate W on which the resist film is formed is subjected to the coating process of the coating processing unit for the resist cover film of the partition 122. Thereby, a resist coating film is formed on the surface of the substrate W. The substrate W on which the resist coating film is formed is heated by the heating and cooling unit of the zone 122 and then cooled.

The peripheral portion of the substrate W after cooling is exposed by the edge exposure unit of the partition 122. The substrate W after the peripheral portion is exposed is subjected to washing and drying processing in the washing processing unit 133a. The cleaned substrate W is cooled by the placing and cooling unit. The cooled substrate W is exposed by an external exposure machine EXP. The exposed substrate W is subjected to washing and drying processing by the washing and drying processing unit 133b. The cleaned substrate W is subjected to post-exposure baking treatment by the heating and cooling unit of the zone 122. The baked substrate W is cooled by the cooling unit of the zone 122. The cooled substrate W is subjected to development processing by the development processing unit. The substrate W subjected to the development process is heated and then cooled by the heating and cooling unit. The cooled substrate W is transported to the cassette C of the transfer conveying unit 110. In the above manner, the substrate processing apparatus processes the substrate W.

＜Coating processing unit＞ FIG. 2 schematically shows an example of a partial configuration of the processing unit 1 as an example of the coating processing unit 123 or the coating processing unit 127. As illustrated in FIG. 2, the processing unit 1 includes a substrate holding unit 10, an ejection nozzle 12, a camera 35, an image processing unit 30, and a control unit 140.

The substrate holding portion 10 holds the substrate W in a substantially horizontal posture. The horizontal posture mentioned here refers to the state where the thickness direction of the substrate W is along the vertical direction. In addition, the center lower surface of the substrate holding portion 10 is connected to an electric motor (not shown) via a rotating shaft 11. The control part 140 rotates the substrate holding part 10 and the substrate W held by the substrate holding part 10 in a horizontal plane by rotating the electric motor.

The ejection nozzle 12 is a hollow cylindrical body, and is connected to a processing liquid supply source 15 via a supply pipe 18. The ejection nozzle 12 is transparent, and its internal flow path 14 is visible. In other words, the ejection nozzle 12 has transparency to such an extent that the internal flow path 14 can be seen from the outside. On the lower surface of the ejection nozzle 12, an ejection port 13 which is a port in front of the internal flow path 14 is formed. The base port of the internal flow path 14 is connected to the supply pipe 18 in communication.

In the middle of the supply pipe 18, a supply valve 16 and a suction valve 17 are provided. By opening the supply valve 16, the processing liquid from the processing liquid supply source 15 is supplied to the ejection nozzle 12 via the supply pipe 18. The ejection nozzle 12 supplies the supplied processing liquid onto the substrate W. The treatment liquid includes, for example, a resist liquid, a coating liquid for film formation, a developing liquid, or a cleaning liquid (also referred to as a rinse liquid). While the substrate holding portion 10 is rotated by the electric motor, the supply valve 16 is opened to supply the processing liquid from the ejection port 13 of the ejection nozzle 12 to the substrate W, thereby performing the coating process.

When the suction valve 17 stops spraying the treatment liquid, it sucks the treatment liquid in the internal flow path 14 of the spray nozzle 12 toward the opposite side of the discharge port 13 and keeps the front end of the treatment liquid away from the discharge port 13. Thereby, it is possible to prevent the processing liquid from falling from the ejection port 13 of the ejection nozzle 12 due to gravity (so-called dripping).

The ejection nozzle 12 is installed in such a way that it can be moved between the processing position and the standby position by a nozzle moving mechanism not shown. The processing position is the position when the ejection nozzle 12 ejects the processing liquid to the substrate W. For example, the processing position is a position facing the vertical upper side with respect to the center portion of the substrate W held by the substrate holding portion 10. The standby position is, for example, a position that does not face the substrate W in the vertical direction. The nozzle moving mechanism has, for example, a ball screw structure.

The camera 35 is a so-called two-dimensional CCD camera including a CCD (charge coupled device). The camera 35 shoots at a specific time interval (frame rate), sequentially acquires image data, and sends the image data to the image processing unit 30. In addition, the camera 35 is provided at a specific location in the processing unit 1 facing the main surface of the substrate W and the ejection port 13 via the supporting part 37 so as to be able to image the imaging area including the ejection port 13 of the ejection nozzle 12.

In the example of FIG. 2, a lighting 36 is provided in the processing unit 1. The illumination 36 is, for example, a light source composed of a light-emitting diode, and is provided at a specific location in the processing unit 1 via the support portion 38, such as near the camera 35, in a manner of illuminating the shooting area.

The control unit 140 controls various configurations of the processing unit 1. Specifically, the control unit 140 controls the supply valve 16, the suction valve 17, the electric motor, and the nozzle moving mechanism. In addition, the control unit 140 can also control the camera 35. For example, the control unit 140 outputs an imaging instruction to the camera 35, and the camera 35 performs imaging in accordance with the imaging instruction.

The control unit 140 is an electronic circuit, and may include a data processing device 141 and a storage medium 142, for example. The data processing device 141 may also be an arithmetic processing device such as a CPU (Central Processor Unit), for example. The storage medium 142 may also include a non-transitory storage medium (such as ROM (Read Only Memory) or a hard disk) and a temporary storage medium (such as RAM (Random Access Memory)). The non-transitory storage medium may store, for example, a program that specifies the processing to be executed by the control unit 140. When the program is executed by the data processing device 141, the control unit 140 can execute the processing specified by the program. Of course, part or all of the processing executed by the control unit 140 may also be executed by hardware.

The image processing unit 30 receives the image data quantized by the camera 35 from the camera 35. The image processing unit 30 performs image processing on the image data. In the example of FIG. 2, the image processing unit 30 includes a processor 31, a first processing memory 32 and a second processing memory 33. The first processing memory 32 and the second processing memory 33 store quantized image data in the camera 35 or the processor 31.

The processor 31 is an electronic circuit. The processor 31 performs image processing on the image data stored in the first processing memory 32 and the second processing memory 33 to detect the gas-liquid interface between the processing liquid and the gas in the internal flow path 14 of the ejection nozzle 12. This point will be described in detail below.

Furthermore, in the example of FIG. 2, the control unit 140 and the image processing unit 30 are provided separately. However, the image processing function of the image processing unit 30 may also be installed in the control unit 140. In this case, the control unit 140 functions as the image processing unit 30.

＜Actions of the processing unit＞ FIG. 3 is a flowchart showing an example of the operation of the processing unit 1. First, the main transport mechanism T1 or the main transport mechanism T2 transports the substrate W into the processing unit 1, and the substrate holder 10 holds the transported substrate W (step S1).

Then, the processing unit 1 executes coating processing (step S2) and monitoring processing (step S3) in parallel. The coating treatment refers to the treatment of coating the surface of the substrate W with a treatment liquid. Monitoring processing refers to processing for monitoring the state of the processing liquid in the internal flow path 14 of the ejection nozzle 12.

＜Coating treatment＞ In the coating process, first, a nozzle moving mechanism (not shown) moves the ejection nozzle 12 from the standby position to the processing position. Thereby, the ejection nozzle 12 enters into the imaging area of the camera 35. Then, an electric motor (not shown) rotates the substrate holding portion 10 and the substrate W, and opens the supply valve 16. Thereby, the processing liquid is ejected from the ejection nozzle 12 toward the surface of the rotating substrate W. The processing liquid ejected from the ejection nozzle 12 to the surface of the substrate W spreads on the surface of the substrate W as the substrate W rotates, and splashes from the periphery of the substrate W.

When a certain period of time elapses after the opening of the supply valve 16, the supply valve 16 is closed. Thereby, the ejection of the processing liquid from the ejection nozzle 12 is ended. Then, the electric motor ends the rotation of the substrate W. Thereby, a specific film (for example, resist) is formed on the upper surface of the substrate W. In the above manner, the processing unit 1 performs coating processing on the substrate W.

＜Monitoring processing＞ In the monitoring process, the camera 35 photographs the imaging area and sequentially acquires image data, and the image processing unit 30 monitors the state of the processing liquid in the internal flow path 14 of the ejection nozzle 12 based on the image data. More specifically, the processor 31 of the image processing unit 30 detects the gas-liquid interface between the processing liquid and the gas in the internal flow path 14 of the ejection nozzle 12 based on the image data. The following first summarizes the thinking methods for detecting the gas-liquid interface.

FIG. 4 is a diagram schematically showing an example of the image data IM1 acquired by the camera 35. In FIG. 4, three image materials IM1[i-1] to IM1[i+1] sequentially acquired by the camera 35 are shown. The image data IM1[i-1] is the image data IM1 acquired at the earliest time point, and the image data IM1[i+1] is the image data IM1 acquired at the latest time point. In the example of FIG. 4, the image data IM1 before and after the processing liquid starts to flow are shown. The ejection nozzle 12 is transparent, and therefore, the processing liquid in the internal flow path 14 is also shown in each image data IM1.

In the example of FIG. 4, the aging element 19 is generated on the surface of the ejection nozzle 12. The so-called aging element 19 refers to an element that represents the year-on-year change in the appearance of the ejection nozzle 12, such as damage generated on the surface of the ejection nozzle 12 or stains attached to the surface. The age element 19 is commonly included in any image data of the image data IM1[i-1] to IM1[i+1].

The image data IM1[i-1] is the image data IM1 acquired in the closed state where the supply valve 16 is closed. In this closed state, the treatment liquid is stationary in the internal flow path 14 of the ejection nozzle 12, and the front end surface of the treatment liquid is located on the upper side of the ejection outlet 13. The position of the front end surface when the treatment liquid is at rest is also called the suck-back position. As long as the suck-back position is within an appropriate range, the processing liquid can be appropriately ejected from the ejection nozzle 12. Furthermore, the front end surface of the treatment liquid is an example of the gas-liquid interface between the treatment liquid and the gas.

The image data IM1[i] is also the image data IM1 acquired in the closed state where the supply valve 16 is closed. When the processing liquid is stationary, the position of the front end face does not change substantially. Therefore, the existence and distribution state of the processing liquid and gas in the internal flow path 14 in the image data IM1[i] is the same as that in the image data IM1[i-1] The existence and distribution status are the same. Ideally, the image data IM1[i-1] and IM1[i] are consistent with each other.

The image data IM1[i+1] is the image data IM1 acquired immediately after the supply valve 16 is opened. When the supply valve 16 starts to open, the treatment liquid starts to move toward the ejection port 13 in the internal flow path 14 of the ejection nozzle 12. In the image data IM1[i+1], the processing liquid is ejected from the ejection port 13 of the ejection nozzle 12. As a result, the internal flow path 14 of the ejection nozzle 12 is filled with the treatment liquid, and the front end surface of the treatment liquid no longer exists in the internal flow path 14. That is, the volume ratio of the gas in the internal flow path 14 is zero. The existence distribution state in the internal flow path 14 of the image data IM1[i+1] is different from the existence distribution state in the image data IM1[i-1] and IM1[i].

As described above, in the image data IM1[i+1] illustrated in FIG. 4, the processing liquid is ejected from the ejection port 13 of the ejection nozzle 12. Therefore, the existence distribution state in the image data IM1 acquired by the camera 35 later is the same as the image data IM1[i+1] as long as the processing liquid can be properly ejected.

Secondly, the difference between IM1[i] and IM1[i+1] with different distribution states is studied. Ideally, the image data IM1[i] and the image data IM1[i+1] are different only in the region R1 described below, and are consistent with each other in other regions. That is, the area R1 is a rectangular area between the front end of the processing liquid in the image data IM1[i] and the lower end of the processing liquid in the image data IM1[i+1] (the lower end of the image data IM1 in the illustration). In the image data IM1[i], the gas occupies the area R1, while in the image data IM1[i+1], the processing liquid occupies the area R1. Furthermore, in the example of FIG. 4, in order to avoid the diagram from becoming complicated, the region R1 is shown slightly smaller.

The image data representing the region R1 can be obtained by the difference between the image data IM1[i] and the image data IM1[i+1]. Ideally, the regions other than the region R1 are consistent with each other in the image data IM1[i-1] and IM1[i]. Therefore, ideally, by eliminating information outside the region R1 by the difference, image data including only the region R1 can be obtained. The upper end of the region R1 represents the front end surface of the processing liquid in the image data IM1[i]. In the image data, since the information outside the region R1 is eliminated, it is possible to easily detect the front end surface of the processing liquid based on the image data.

Therefore, the processor 31 of the image processing unit 30 detects the gas-liquid interface (for example, the front end surface of the processing liquid) in the internal flow path 14 based on the difference between the two image data IM1 with different distribution states. Hereinafter, a specific example of the monitoring process will be described.

Fig. 5 is a flowchart showing a specific example of the monitoring process (step S3 in Fig. 3). This monitoring process is performed in parallel with the coating process (step S2 in FIG. 3). First, the camera 35 captures the imaging area to acquire image data IM1[n], and sends the image data IM1[n] to the image processing unit 30 (step S31). The initial value of the value n is 0, for example. As described above, step S31 is repeatedly performed, and therefore, the camera 35 sequentially takes pictures during the entire period of the coating process (processing period).

At the beginning of the processing period, the ejection nozzle 12 has not moved to the processing position. Therefore, the ejection nozzle 12 is not located in the shooting area, and the ejection nozzle 12 is not yet included in the image data IM1[n] obtained by the camera 35.

The processor 31 of the image processing unit 30 confirms the position of the ejection nozzle 12 based on the image processing (shape recognition processing) on the image data IM1[n] (step S32). Specifically, the processor 31 determines whether the ejection nozzle 12 is located at the processing position. This determination is performed by pattern matching, for example. For example, a reference image obtained by pre-photographing the ejection nozzle 12 located at the processing position is stored in a storage medium (for example, the first processing memory 32) in advance. Then, the processor 31 specifies the position of the ejection nozzle 12 by matching the pattern of the image data IM1[n] with the reference image, and determines whether the ejection nozzle 12 is located at the processing position. If the ejection nozzle 12 is not located at the processing position, the processor 31 does not perform the processing of steps S32 to S35 described later, and determines whether to end the monitoring processing (step S36). Here, since the monitoring process has not yet ended, step S31 is executed again after incrementing the value n.

When the ejection nozzle 12 moves to the processing position, the ejection nozzle 12 is included in the specific area of the image data IM1[n]. The processor 31 determines that the ejection nozzle 12 is located at the processing position based on the image data IM1[n]. When it is determined that the ejection nozzle 12 is located at the processing position, the processor 31 detects the gas-liquid interface in the internal flow path 14 of the ejection nozzle 12 based on the difference between the two image data IM1 (steps S33, S34).

In addition, when the supply valve 16 is in the process of closing, ideally, the image data IM1[n] sequentially acquired by the camera 35 are the same as each other (for example, refer to the image data IM1[i-1], IM1[ i]). Thus, the pixel values of the pixels of the image data obtained by the difference of the image data IM1 are ideally all zero, and the information on the front end surface of the processing liquid also disappears. Therefore, in the gas-liquid interface detection process (steps S33, S34) of this embodiment, the position of the gas-liquid interface is not detected when the supply valve 16 is in the process of closing.

Here, for easy understanding, first, take the image data IM1[n] acquired immediately after the supply valve 16 starts to open as an example to describe the processing. That is, the case where the image material IM1[i+1] of FIG. 4 is acquired as the image material IM1[n] will be described.

The processor 31 preprocesses the image data IM1[n] (step S33). First, the processor 31 performs edge extraction processing such as the Canny method on the image data IM1[n] to obtain the edge image IM2[n]. FIG. 6 is a diagram schematically showing an example of the edge image IM2[n]. Here, the pixel value of the edge pixel is "1", and the pixel value of the edge pixel is "0".

The processor 31 stores the edge image IM2[n] in the first processing memory 32. Furthermore, in the second processing memory 33, the edge image IM2[n-1] acquired in the most recent (that is, previous) step S33 is stored. The edge image IM2[n-1] is the image data obtained by performing edge extraction processing on the nearest image data IM1[n-1]. The image data IM1[n-1] is the last image data IM1 acquired when the supply valve 16 is in the closing process, which is equivalent to the image data IM1[i] in FIG. 4.

Then, the processor 31 performs edge dilation processing on the edge image IM2[n] to obtain the edge dilation image IM3[n]. The so-called edge expansion process refers to the process of expanding the width of the edge in the edge image IM2[n], for example, refers to the process of expanding the edge by a specific width in the up, down, left, and right directions with the pixel representing the edge as the center. Fig. 7 is a diagram schematically showing an example of an edge dilated image IM3[n]. It can be understood from the comparison between FIG. 6 and FIG. 7 that the width of the edge in the edge expansion image IM3[n] is larger than the width of the edge in the edge image IM2[n]. Therefore, hereinafter, the edge in the edge expansion image IM3[n] is also referred to as a wide edge. The processor 31 stores the edge expansion image IM3[n] in the first processing memory 32.

Then, the processor 31 uses the nearest edge image IM2[n-1] and the current edge expansion image IM3[n] to perform differential processing to detect the gas-liquid interface (for example, the front end surface of the processing liquid) (step S34). FIG. 8 is a flowchart showing an example of specific operations of the gas-liquid interface detection process (step S34). The processor 31 uses the edge image IM2[n-1] and the edge dilated image IM3[n] to perform difference processing, and obtains the difference image IM4[n] (step S341). Fig. 9 is a diagram for explaining the difference processing. The so-called difference processing refers to the processing of subtracting pixels at the same position of two images from each other. Here, the processor 31 subtracts the edge dilated image IM3[n] from the edge image IM2[n-1] to perform difference processing.

As illustrated in Fig. 9, the edge image IM2[n-1] includes an edge E1 representing the ejection nozzle 12, an edge E2 representing the age element 19, and an edge E3 representing the front end surface of the treatment liquid. On the other hand, the edge expansion image IM3[n] includes the wide edge WE1 corresponding to the ejection nozzle 12, the wide edge WE2 corresponding to the age element 19, and the outline of the processing liquid ejected from the ejection port 13 The wide edge WE4.

The edge E1 and the wide edge WE1 indicate the same object (specifically, the ejection nozzle 12), and the edge E1 is thinner than the wide edge WE1. Therefore, even if the position of the ejection nozzle 12 is slightly shifted between the edge image IM2[n-1] and the edge expansion image IM3[n], the edge image IM2[n-1] and the edge expansion image When IM3[n] overlaps, the edge E1 will also be covered by the wide edge WE1. In other words, the pixels of the edge dilated image IM3[n] at the same positions as the pixels representing the edge E1 become pixels representing the wide edge WE1.

Similarly, the edge E2 and the wide edge WE2 represent the same object (specifically, the age element 19), and the edge E2 is thinner than the wide edge WE2. Therefore, even if the position of the ejection nozzle 12 is slightly shifted, when the edge image IM2[n-1] and the edge expansion image IM3[n] are overlapped, the edge E2 will be covered by the wide edge WE2.

Therefore, through the difference processing, the edges E1 and E2 are eliminated in the difference image IM4[n]. That is, the pixel value of the pixels corresponding to the edges E1 and E2 in the difference image IM4[n] is "0" (=1-1).

On the other hand, regions of the wide edges WE1 and WE2 that do not overlap with the edges E1 and E2 (hereinafter referred to as edges SE1 and SE2) remain in the difference image IM4[n]. However, by the difference processing, the pixel value of each pixel of the edges SE1 and SE2 becomes "-1" (=0-1). In the example of FIG. 9, the pixels with the pixel value "-1" are shaded with diagonal lines.

The edge corresponding to the wide edge WE4 in the edge expansion image IM3[n] does not exist in the edge image IM2[n-1], so the wide edge WE4 remains in the difference image IM4[n]. However, through the difference processing, the pixel value of each pixel of the wide edge WE4 also becomes "-1" (=0-1).

The wide edge corresponding to the edge E3 in the edge image IM2[n-1] does not exist in the edge expansion image IM3[n], so the edge E3 remains in the difference image IM4[n]. The pixel value of each pixel of the edge E3 of the difference image IM4[n] is still "1" (=1-0).

As described above, the difference image IM4[n] includes the edge E3 representing the front end surface of the processing liquid in the edge image IM2[n-1], and the edges SE1, SE2 from the edge expansion image IM3[n] WE4. In the difference image IM4[n], the pixel value of each pixel of the edge SE1, SE2, WE4 is "-1", the pixel value of each pixel of the edge E3 is "1", and the pixel value of other pixels is " 0".

The processor 31 detects the gas-liquid interface in the image data IM1[n-1] based on the position of the edge E3 in the difference image IM4[n]. As an example of specific processing, the processor 31 cuts off the edges SE1, SE2, and WE4 from the edge dilated image IM3[n] in the difference image IM4[n] (step S342). In other words, the processor 31 cuts off pixels with negative pixel values ("-1"). For example, the processor 31 replaces the pixel value of the pixel from "-1" to "0". Only the edge E3 remains in the difference image IM4[n] (removed image) after the cut. The pixel value of each pixel of the edge E3 is "1". Therefore, the processor 31 obtains the position of the gas-liquid interface based on the position of the pixel having the positive ("1") pixel value (step S343).

The processor 31 can also detect the position of the gas-liquid interface based on the position of the ejection outlet 13 of the ejection nozzle 12, for example. The position of the ejection port 13 can be preset, for example, and stored in a storage medium (for example, the first processing memory 32). Alternatively, the processor 31 may also detect the position of the ejection port 13 based on the image data IM1[n-1]. For example, the processor 31 can detect the position of the ejection port 13 by matching the pattern of the image data IM1[n-1] with the reference image.

As described above, when the image data IM1[n] is acquired by the camera 35 immediately after the supply valve 16 is opened, the processor 31 detects the gas-liquid interface in the image data IM1[n-1] acquired most recently before it.的 position (steps S32 ~ S34). The most recent image data IM1[n-1] is the last image data IM1 acquired during the closing process of the supply valve 16, and the position of the gas-liquid interface indicates the suction position. That is, the position of the gas-liquid interface first detected in this monitoring process (step S3) represents the suck-back position.

Then, the processor 31 makes an abnormality determination based on the detected gas-liquid interface (step S35). Specifically, the processor 31 specifies the position of the gas-liquid interface first detected in the monitoring process as the suck-back position, and determines whether the suck-back position is within the reference range. The reference range is preset and stored in a storage medium (for example, the first processing memory 32).

When it is determined that the suck-back position is outside the reference range, the processor 31 determines that an abnormality has occurred, and sends the determination result to the control unit 140. The control unit 140 causes the notification unit 40 to notify the abnormality, for example. The notification unit 40 is, for example, a device such as a display or a speaker. With this notification, the operator can be notified that an abnormality has occurred and can take appropriate countermeasures.

Then, the processor 31 determines whether the monitoring process should be ended (step S36). For example, when the coating process (step S2) ends, the processor 31 determines that the monitoring process should be ended, and ends the monitoring process. When it is determined that the monitoring processing should not be ended, the processor 31 stores the edge image IM2[n] in the second processing memory 33. And, after incrementing the value n, the photographing by the camera 35 in step S1 is performed again. The above operations are repeated until the coating process is completed.

As described above, the processor 31 detects the position of the gas-liquid interface in the internal flow path 14 of the ejection nozzle 12 based on the image data IM1 acquired by the camera 35, and therefore can detect the position of the gas-liquid interface with high accuracy.

Also, in the above example, the processor 31 uses the image data IM1[n-1] when the processing liquid is stationary and the image data IM1[n] when the processing liquid is flowing to detect the image data IM1[n-1] The front end of the treatment liquid. In this way, the suck-back position can be detected.

Furthermore, in the above example, the processor 31 detects the position of the gas-liquid interface based on the difference between the edge image IM2[n-1] and the edge expansion image IM3[n]. Thereby, even if there is a position shift of the ejection nozzle 12 between the image data IM1, the edges other than the edge E3 in the edge image IM2[n-1] can be eliminated in the difference image IM4[n]. In addition, the processor 31 can leave only the edge E3 in the difference image IM4[n] by cutting off the edges SE1, SE2, and WE4 from the edge expansion image IM3[n]. The edge E3 represents the edge of the front end surface (gas-liquid interface) of the processing liquid in the edge image IM2[n-1]. Therefore, the processor 31 can eliminate the influence of the positional deviation of the ejection nozzle 12 between the image data IM1, and detect the position of the gas-liquid interface with high accuracy.

Furthermore, in the above example, the processor 31 detects the position of the gas-liquid interface based on the difference between the two image data IM1 acquired by the camera 35 during the processing (ie, during the coating processing). Therefore, even if the aging element 19 such as dirt or damage occurs in the ejection nozzle 12, the aging element 19 is commonly included in the two image data IM1. Therefore, by using the difference between the two image data IM1, the influence of the age element 19 can be eliminated to detect the position of the gas-liquid interface. Therefore, the position of the gas-liquid interface can be detected with higher accuracy. In addition, the processing period mentioned here refers to the period from the state where the processing liquid is stationary in the internal flow path 14 of the ejection nozzle 12 to the time when the processing liquid is finished spraying and the processing liquid is stationary again.

＜Bubble before spraying＞ In addition, there are cases where the suction treatment cannot be performed properly when the treatment liquid stops ejecting, and air bubbles are mixed in the internal flow path 14 of the ejection nozzle 12. FIG. 10 is a diagram schematically showing an example of the image data IM1 acquired by the camera 35. The image data IM1 is acquired in the closed state where the supply valve 16 is closed. In the example of FIG. 10, air bubbles B1 are mixed in the internal flow path 14 of the ejection nozzle 12. That is, in the previous coating process, the suction process was not properly performed, and the air bubbles B1 were mixed in the internal flow path 14. The bubble B1 is formed in the processing liquid. In the example of FIG. 10, the processing liquid is separated into upper and lower sides. In this case, there are three gas-liquid interfaces in the internal flow path 14. That is, there are a gas-liquid interface representing the upper end surface of the bubble B1, a gas-liquid interface representing the lower end surface of the bubble B1, and a gas-liquid interface representing the front end of the processing liquid.

In this case, the processor 31 detects the positions of the three gas-liquid interfaces through the above-mentioned monitoring process (step S3). When the positions of a plurality of gas-liquid interfaces are detected in this way, it can be said that the state of the processing liquid in the internal flow path 14 of the ejection nozzle 12 is abnormal. Therefore, in the abnormality determination (step S35), the processor 31 determines whether the positions of a plurality of gas-liquid interfaces are detected, and when the positions of a plurality of gas-liquid interfaces are detected, it is determined that an abnormality has occurred. At this time, the processor 31 notifies the control unit 140 of the content of the abnormality. Based on the notification, the control unit 140, for example, causes the notification unit 40 to notify the abnormality.

＜Bubble detection during ejection＞ In the above example, the bubble B1 before the discharge of the treatment liquid was described. However, there may be cases where air bubbles are mixed into the supply pipe 18 during the process of discharging the processing liquid. The air bubbles move in the supply pipe 18 together with the processing liquid and enter the internal flow path 14 of the ejection nozzle 12. Then, the air bubbles move toward the ejection port 13 in the internal flow path 14 of the ejection nozzle 12, and are released from the ejection port 13 to the outside. When the bubbles are released, the treatment liquid may splash in all directions.

FIG. 11 is a diagram schematically showing an example of the image data IM1 acquired by the camera 35. In FIG. 11, the image data IM1 acquired by the camera 35 during the ejection period (ejection period) of the processing liquid from the ejection port 13 of the ejection nozzle 12 is shown. In FIG. 11, three image data IM1[j-1]～IM1[j+1] sequentially acquired by the camera 35 are displayed in a row. The image data IM1[j-1] is the image data IM1 acquired at the earliest time point, and the image data IM1[j+1] is the image data IM1 acquired at the latest time point. In the example of FIG. 11, the image data IM1 during the process of discharging the processing liquid is shown.

In the image data IM1[j-1], the internal flow path 14 of the ejection nozzle 12 is filled with the processing liquid, and the bubble B2 is not yet included. However, the bubble B2 was generated on the upstream side of the internal flow path 14. The bubble B2 moves toward the ejection port 13 as time passes. In the image data IM1[j], the internal flow path 14 contains bubbles B2. The air bubble B2 moves toward the ejection port 13 as time passes, and is released from the ejection port 13 to the outside. In the example of FIG. 11, in the image data IM1[j+1], the bubble B2 has been released to the outside, and the internal flow path 14 is filled with the processing liquid again.

In the example of FIG. 11, the existence and distribution state of the processing liquid and gas in the internal flow path 14 of the image data IM1[j] and the existence and distribution state of the image data IM1[j-1] and IM1[j+1] are either One is different. In addition, the image data IM1[j+1] ideally matches the image data IM1[j-1].

According to the above monitoring process (step S3), when the camera 35 acquires the image data IM1[j], the processor 31 is based on the most recently acquired image data IM1[j-1] and the current image data IM1[j], The position of the gas-liquid interface in the image data IM1[j-1] is detected (steps S33, S34). However, since the bubble B2 is not included in the image data IM1[j-1], there is no gas-liquid interface. Therefore, at this point in time, the position of the gas-liquid interface was not detected.

Then, when the camera 35 acquires the image data IM1[j+1], the processor 31 similarly detects the position of the gas-liquid interface in the image data IM1[j] based on the image data IM1[j], IM1[j+1] ( Steps S33, S34). The image data IM1[j] contains the bubble B2. Therefore, the processor 31 detects the position of the contour surface of the bubble B2 (the position of the gas-liquid interface) in the image data IM1[j].

As described above, according to this monitoring process, the bubble B2 generated in the internal flow path 14 of the ejection nozzle 12 during the ejection of the processing liquid can be detected.

However, in the above specific example, the processor 31 does not detect the position of the gas-liquid interface in the current image data IM1, but detects the position of the gas-liquid interface in the above-mentioned nearest image data IM1. Therefore, when the processor 31 obtains the image data IM1[j], although the image data IM1[j] contains the bubble B2, the image data IM1[j-1] does not contain the bubble B2. Therefore, The position of the gas-liquid interface of the bubble B2 cannot be detected. The processor 31 detects the position of the air-liquid interface of the bubble B2 when the image data IM1[j+1] is acquired as described above. Thus, the detection of bubble B2 is postponed.

Therefore, the processor 31 detects the position of the gas-liquid interface in the current image data IM1[n] instead of the most recent image data IM1[n-1] after the treatment liquid is ejected. Fig. 12 is a diagram for explaining the difference processing. As shown in FIG. 12, the processor 31 uses an edge expansion image IM3[n-1] based on the latest image data IM1[n-1] and an edge image IM2 based on the current image data IM1[n] [n] Perform difference processing.

The edge image IM2[n] is an image obtained by performing edge extraction processing on the current image data IM1[n]. Here, the image data IM1[n] is the image data IM1[j]. In other words, FIG. 12 shows the difference processing when the camera 35 acquires the image data IM1[j]. The edge image IM2[n] is stored in the first processing memory 32, for example.

The edge expansion image IM3[n-1] is an image obtained by sequentially performing edge extraction processing and edge expansion processing on the nearest image data IM1[n-1]. Here, the image data IM1[n-1] is the image data IM1[j-1]. The most recent edge expansion image IM3[n-1] is stored in the second processing memory 33 in the previous step S33, for example.

In the example of FIG. 12, the edge expansion image IM3[n-1] includes the wide edge WE1 corresponding to the ejection nozzle 12, the wide edge WE2 corresponding to the age element 19, and the wide edge WE2 corresponding to the ejection port 13 The wide edge of the contour of the treatment liquid WE4. The edge image IM2[n] includes the edge E1 representing the ejection nozzle 12, the edge E2 representing the age element 19, the edge E4 representing the outline of the processing liquid ejected from the ejection port 13, and the upper end surface of the bubble B2 and The edge of the lower end face E5.

The edges E1, E2, and E4 in the edge image IM2[n] respectively represent the same objects as the wide edges WE1, WE2, and WE4 in the edge expansion image IM3[n-1]. Therefore, through the difference processing, It is eliminated in the difference image IM4[n]. That is, the pixel value of each pixel of the edges E1, E2, E4 other than the edge E5 in the edge image IM2[n] becomes "0" (=1-1) in the difference image IM4[n].

On the other hand, regions of the wide edges WE1, WE2, and WE4 that do not overlap with the edges E1, E2, and E4, respectively (hereinafter referred to as edges SE1, SE2, and SE4, respectively) remain in the difference image IM4[n]. The pixel value of each pixel of the edges SE1, SE2, SE4 is "1" (=1-0).

The wide edge corresponding to the edge E5 in the edge image IM2[n] does not exist in the edge expansion image IM3[n-1], so it remains in the difference image IM4[n]. However, the pixel value of each pixel of the edge E5 is "-1" (=0-1).

As mentioned above, the difference image IM4[n] includes the edge E5 representing the upper and lower end faces of the bubble B2 in the edge image IM2[n], and the edge expansion image IM3[n-1]. Edge SE1, SE2, SE4. In the difference image IM4[n], the pixel value of each pixel of the edge SE1, SE2, SE4 is "1", the pixel value of each pixel of the edge E5 is "-1", and the pixel value of other pixels is " 0".

The processor 31 detects the gas-liquid interface of the processing liquid in the image data IM1[n] based on the edge E5 in the difference image IM4[n]. Here, the position of the gas-liquid interface represents the contour surface of the bubble B2. As an example of specific processing, the processor 31 cuts off the edges SE1, SE2, and SE4 from the edge expansion image IM3[n-1]. In short, the processor 31 cuts off pixels with positive ("1") pixel values. For example, the processor 31 may also replace the pixel value of the pixel from "1" to "0". Only the edge E5 remains in the difference image IM4[n] (removed image) after the cut. Then, the processor 31 can also perform absolute value processing on the pixel value of each pixel of the cut-off difference image IM4[n]. In other words, the processor 31 replaces the pixel value of each pixel of the edge E5 from "-1" to "1". The processor 31 obtains the position of the gas-liquid interface (the position of the contour surface of the bubble B2) based on the position of the pixel with the positive ("1") pixel value.

As described above, the processor 31 detects the gas-liquid interface in the latest image data IM1[n-1] before the processing liquid is ejected to detect the suction position, and after the processing liquid is ejected, detects the current image data IM1 The position of the gas-liquid interface in [n] can detect bubble B2 more quickly.

Fig. 13 is a flowchart showing an example of the above-mentioned processing. Fig. 13 corresponds to a specific example of step S34 in Fig. 5. The processor 31 determines whether it is after the treatment liquid has been discharged (step S340). For example, the control unit 140 may also notify the processor 31 of information indicating that the supply valve 16 has been switched from the closed state to the open state. Based on this information, the processor 31 determines whether it is after the treatment liquid is discharged.

Alternatively, the processor 31 may determine whether to eject the processing liquid based on the image data IM1. For example, the processor 31 determines whether or not the treatment liquid is ejected by image processing on the image data IM1. For example, the processor 31 determines whether or not the processing liquid is ejected based on the pixel value of the area below the ejection port 13 of the ejection nozzle 12 (hereinafter referred to as ejection area) in the image data IM1. The pixel value of the ejection area when the processing liquid is ejected is different from the pixel value of the ejection area when the processing liquid is not ejected. Therefore, the processor 31 can determine whether there is ejection or not based on the pixel value of the ejection area. For example, when the treatment liquid is ejected, the light reflected by the treatment liquid from the illumination 36 may be received by the camera 35. In this case, the brightness value of the spray area during the process of spraying the treatment liquid is higher than the brightness value of the spray area when the treatment liquid is not sprayed. In this case, for example, the processor 31 determines that the treatment liquid has been ejected when the average of the brightness values of the ejection area is higher than the reference brightness value. The reference brightness value is preset and stored in a storage medium (for example, the first processing memory 32).

When the processing liquid has not been ejected yet (step S340: NO), the processor 31 performs the above-mentioned steps S341 to S343. Thereby, the processor 31 uses the edge image IM2[n-1] based on the latest image data IM1[n-1] and the edge expansion based on the current image data IM1[n] before the processing liquid is ejected. The image IM3[n] is subjected to difference processing (step S341). Thereby, when the camera 35 acquires the image data IM1[n] just after the supply valve 16 starts to open, the processor 31 can detect the gas-liquid interface in the most recent image data IM1[n-1], that is, suck back Location.

In addition, in the case after the treatment liquid is ejected (step S340: Yes (YES)), the processor 31 uses the edge expansion image IM3[n-1] and the edge image IM2[n] to perform differential processing to obtain the differential Image IM4[n] (step S344, also refer to FIG. 12). Then, the processor 31 cuts off the edge from the edge expansion image IM3[n-1] in the difference image IM4[n] (step S345). Then, the processor 31 performs absolute value processing on each pixel of the cut-off difference image IM4[n] (step S346). Thereby, in the difference image IM4[n], the pixel value of each pixel of the edge E5 is replaced from "-1" to "1". The processor 31 detects the position of the gas-liquid interface based on the pixels with positive pixel values in the difference image IM4[n] (step S347).

<Abnormal judgment> Before the treatment liquid is discharged, since the front end surface of the treatment liquid exists in the internal flow path 14 of the discharge nozzle 12 (refer to FIG. 4), if the bubble B1 is not generated, a gas-liquid interface is detected. On the other hand, if the bubble B1 (refer to FIG. 10) is generated, a plurality of gas-liquid interfaces will be detected. Therefore, if a plurality of gas-liquid interfaces are detected before the treatment liquid is discharged, it can be said that an abnormality has occurred. Therefore, when the processor 31 detects a plurality of gas-liquid interfaces before the treatment liquid is discharged, it is determined that an abnormality has occurred.

In contrast, in the normal process liquid ejection process, there is no process liquid front end surface in the internal flow path 14 of the ejection nozzle 12, and the internal flow path 14 is filled with the process liquid. Furthermore, if the bubble B2 (refer to FIG. 11) is generated, a gas-liquid interface is detected. Therefore, if the gas-liquid interface is detected during the process liquid ejection, it can be said that an abnormality has occurred. Therefore, when the processor 31 detects a plurality of gas-liquid interfaces during the process of discharging the treatment liquid, it is determined that an abnormality has occurred.

Fig. 14 is a flowchart showing an example of the above-mentioned operation. FIG. 14 corresponds to a specific example of abnormality determination in step S35 in FIG. 5. However, in the example of FIG. 14, the illustration of the procedure for determining the suction position is omitted.

The processor 31 determines whether the treatment liquid is ejected from the ejection port 13 of the ejection nozzle 12 (step S350). That is, the processor 31 determines whether or not the treatment liquid is currently being discharged.

When the treatment liquid is not ejected (step S350: No), the processor 31 determines whether a plurality of gas-liquid interfaces are detected (step S351). When a plurality of gas-liquid interfaces are detected, the processor 31 determines that an abnormality has occurred, and notifies the control unit 140 of the abnormality. Based on the notification, the control unit 140, for example, causes the notification unit 40 to notify the abnormality (step S352). When a plurality of gas-liquid interfaces are not detected, the processor 31 does not execute step S352.

On the other hand, when the treatment liquid has been ejected (step S350: Yes), the processor 31 determines whether a gas-liquid interface is detected (step S353). When the gas-liquid interface is detected, the processor 31 determines that an abnormality has occurred, and the control unit 140 notifies that the abnormality has occurred. Based on the notification, the control unit 140, for example, causes the notification unit 40 to notify the abnormality (step S354).

＜Stop spraying of treatment liquid＞ Next, the action when the treatment liquid stops spraying will be explained. FIG. 15 is a diagram schematically showing an example of the image data IM1 acquired by the camera 35. In FIG. 15, three image materials IM1[k-1] to IM1[k+1] sequentially acquired by the camera 35 are shown. The image data IM1[k-1] is the image data IM1 acquired at the earliest time point, and the image data IM1[k+1] is the image data IM1 acquired at the latest time point. In the example of FIG. 15, the image data IM1 before and after the processing liquid stops ejecting are shown.

The image data IM1[k-1] is the image data IM1 acquired when the supply valve 16 is opened. In the image data IM1[k-1], the processing liquid is ejected from the ejection port 13 of the ejection nozzle 12, and the internal flow path 14 is filled with the processing liquid.

The image data IM1[k] is the image data IM1 obtained immediately after the supply valve 16 is closed and the suction valve 17 performs the suction action. In the image data IM1[k], the processing liquid is no longer ejected from the ejection port 13 of the ejection nozzle 12, and only exists in the internal flow path 14 thereof. The front end surface of the treatment liquid is located above the discharge port 13 in the internal flow path 14. Thus, the existence and distribution states of the processing liquid and the gas in the internal flow path 14 of the image data IM1[k-1] and the image data IM1[k] are different from each other.

After the supply valve 16 is closed, the treatment liquid is stationary in the internal flow path 14. Therefore, as long as the supply valve 16 is closed, the position of the front end surface of the treatment liquid does not change. Therefore, the image data IM1[k+1] acquired under the image data IM1[k] is ideally consistent with the image data IM1[k].

According to the action of FIG. 13, after the processing liquid is ejected, the processor 31 detects the air-liquid interface in the current image data IM1[n] (steps S344-S347). Thus, when the camera 35 acquires the image data IM1[k], the processor 31 performs a gas-liquid interface based on the difference between the above-mentioned latest image data IM1[k-1] and the current image data IM1[k]. The detection process (steps S344 ~ S347). The image data IM1[k-1] and IM1[k] have different existence distribution states in the internal flow path 14; therefore, the processor 31 can detect the position of the gas-liquid interface of the image data IM1[k].

In a word, the processor 31 acquires the edge image IM2[k] based on the image data IM1[k] when the processing liquid is stationary, and acquires the edge expansion image IM3[k] based on the image data IM1[k-1] when the processing liquid flows. -1]. As a result, the processor 31 can detect the position of the gas-liquid interface in the image data IM1[k] when the processing liquid is stationary by performing the difference processing.

Then, when the camera 35 obtains the image data IM1[k+1], the processor 31 also detects the gas-liquid interface based on the difference between the above-mentioned latest image data IM1[k] and the current image data IM1[k+1] Processing (steps S344 to S347). However, the existence and distribution states of the internal flow paths 14 of the image data IM1[k-1] and IM1[k] are the same. Therefore, the processor 31 does not detect the position of the gas-liquid interface of the image data IM1[k+1]. Same from now on.

As mentioned above, it can be said that the position of the gas-liquid interface last detected in this monitoring process represents the suck-back position after the treatment liquid stops spraying. Therefore, the processor 31 can also specify the last detected position of the gas-liquid interface as the suck-back position. In addition, the processor 31 may also determine whether or not the suck-back position is within the reference range.

In addition, according to the abnormality determination in FIG. 14, at the point in time when the image data IM1[k] is acquired, the processing liquid is not ejected. Therefore, the processing of steps S351 and S352 is performed. As a result, when the suction process is not performed properly when the treatment liquid is stopped and the bubbles B1 are mixed in, the processor 31 detects a plurality of gas-liquid interfaces, and therefore determines that an abnormality has occurred.

The embodiments have been described above, but this substrate processing apparatus can be modified in various ways other than the above as long as it does not deviate from the gist. This embodiment mode can freely combine the various embodiments within the scope of the disclosure, or change any constituent elements of the various embodiments, or omit any constituent elements in each embodiment.

For example, the processing unit 1 is not necessarily limited to the coating processing unit 123 or the development processing unit DEV. The processing unit 1 only needs to have a transparent ejection nozzle 12 that ejects the processing liquid to the substrate W.

Furthermore, in the above example, the processor 31 detects the air-liquid interface based on the difference between the two image data IM1 acquired by the camera 35 during the processing period (during coating processing). However, the reference image captured by the camera 35 in advance may be stored in the second processing memory 33 in advance. As the reference image, for example, the image data IM1 (for example, the image data IM1[i+1] in FIG. 4) during the normal ejection of the processing liquid can be used. The processor 31 can also detect the gas-liquid interface in the image data IM1 based on the difference between the image data IM1 acquired during the processing period and the reference image.

Also, in the above example, the camera 35 takes pictures during the entire processing period. However, when the period to be monitored is only a part of the processing period and it is sufficient, the camera 35 may also take pictures during only a part of the processing period.

10: Board holding part 11: Rotation axis 12: Nozzle (spray nozzle) 13: spout 14: Internal flow path 15: Treatment liquid supply source 16: supply valve 17: suction valve 18: Supply piping 19: The elements of aging 30: Image Processing Department 31: processor 32: The first processing memory 33: The second processing memory 35: Camera 36: lighting 37: Support Department 38: Support Department 40: Notification Department 110: Transfer and Transport Department 111: cassette placement table 120: Processing Department 121: Partition 122: Partition 123: Coating processing unit 124: Heat treatment unit 125: tablet 126: Heat treatment unit 127: Coating processing unit 130: face 131: Wash processing block 132: Moving in and out of the block 133a: Washing processing unit 133b: Washing processing unit 140: Control Department 141: Data Processing Device 142: Memory Media A1: Transport space A2: Transport space B1: Bubble B2: bubbles C: Box E1: Edge E2: Edge E3: Edge E4: Edge E5: Edge EXP: Exposure machine H1: Keep arm H2: Keep arm IM1[i-1]~IM1[i+1]: Image data IM1[n]: Image data IM2[n]: Edge image IM3[n]: Inflated edge image IM4[n]: Difference image LPa: Move-in department LPb: Moving out department PASS1: Placement Department PASS2: Placement Department PASS3: Placement Department P-BF: Mounting and buffering part S31~S36, S340~S347, S350~S354: steps SE1: Edge SE2: Edge SE3: Edge SE4: Edge T1: Main transport mechanism T2: Main transport mechanism T3a: Transport mechanism T3b: Transport mechanism TID: ID transfer mechanism TIF: IF transfer mechanism W: substrate WE1: Wide edge WE2: Wide edge WE3: Wide edge WE4: Wide edge

FIG. 1 is a diagram schematically showing an example of the structure of a substrate processing apparatus. Fig. 2 is a diagram schematically showing an example of the configuration of the processing unit. Fig. 3 is a flowchart showing an example of the operation of the processing unit. Fig. 4 is a diagram schematically showing an example of image data. Fig. 5 is a flowchart showing an example of monitoring processing. Fig. 6 is a diagram schematically showing an example of an edge image. Fig. 7 is a diagram schematically showing an example of an edge dilated image. Fig. 8 is a flowchart showing an example of the detection process of the gas-liquid interface. Fig. 9 is a diagram for explaining difference processing. Fig. 10 is a diagram schematically showing another example of image data. Fig. 11 is a diagram schematically showing another example of image data. Fig. 12 is a diagram for explaining the difference processing. Fig. 13 is a flowchart showing an example of difference processing. Fig. 14 is a flowchart showing an example of abnormality determination processing. Fig. 15 is a diagram schematically showing another example of image data.

S31~S36: steps

Claims (8)

A method for detecting the gas-liquid interface inside the nozzle, which detects the gas-liquid interface between the processing liquid and the gas in the internal flow path of the nozzle, and has: A photographing process, in which a camera photographs a photographed area including the transparent nozzle to obtain an image, and the nozzle sprays the treatment liquid from the discharge port that is the port before the internal flow path; and The detecting step includes detecting the gas-liquid interface in the internal flow path based on a difference between the two images in which the presence and distribution state of the processing liquid and the gas in the internal flow path are different. Such as the detection method of the gas-liquid interface inside the nozzle of claim 1, where The above-mentioned inspection process includes: An edge extraction process, which performs edge extraction processing on the two above-mentioned images acquired by the above-mentioned camera, and respectively acquires a first edge image and a second edge image; An edge expansion step, which performs an edge expansion process to expand the edge of the second edge image to obtain an edge expansion image; and A step of acquiring a difference image representing the difference between the first edge image and the edge expansion image, and detecting the gas-liquid interface based on the edges of the difference image excluding the edge from the edge expansion image . Such as the detection method of the gas-liquid interface inside the nozzle of claim 2, where In the photographing step, the camera sequentially acquires the images during at least a part of the processing period from the state where the processing liquid is stationary in the internal flow path of the nozzle to the end of the processing liquid after the ejection of the processing liquid Until the liquid is stationary again, In the detection step, the position of the gas-liquid interface is detected based on the two images taken during at least a part of the processing period. Such as claim 3, the detection method of the gas-liquid interface inside the nozzle, where In the imaging step, the camera acquires a first image when the processing liquid is stationary in the internal flow path, and acquires a second image when the processing liquid flows toward the ejection port, In the edge extraction step, the first edge image is acquired based on the first image, and the second edge image is acquired based on the second image. For example, the detection method of the gas-liquid interface inside the nozzle of claim 4, which further has the following steps: In the detection step, when a plurality of positions of the gas-liquid interface are detected based on the first image and the second image, it is determined that an abnormality has occurred. Such as the detection method of the gas-liquid interface inside the nozzle of any one of claims 3 to 5, wherein In the photographing step, the camera sequentially acquires the images during the ejection period in which the processing liquid is ejected from the ejection port of the nozzle, and In the edge extraction step, the first edge image is acquired based on the current image acquired by the camera during the ejection period, and the first edge image is acquired based on the image acquired by the camera most recently before the current image during the ejection period. Image to obtain the second edge image described above. For example, the method for detecting the gas-liquid interface inside the nozzle of claim 6, which further has the following steps: In the detection step, when the gas-liquid interface is detected based on the two images taken during the ejection period, it is determined that an abnormality has occurred. A substrate processing device, which includes: The substrate holding portion, which holds the substrate; A transparent nozzle that ejects the processing liquid from the port in front of the internal flow path, that is, the ejection port, to the substrate held by the substrate holding portion; A camera that captures the shooting area including the above-mentioned nozzle to obtain an image; and An image processing unit that detects the gas-liquid interface between the processing liquid and the gas in the internal flow path based on the difference between the two images in which the presence and distribution of the processing liquid and the gas in the internal flow path are different .

Images