How to adjust the nozzle start condition

Abstract

To provide an adjustment method of the ejection start state of a nozzle which is easy to adjust and can suppress errors between operators and between devices.SOLUTION: A controller 51 calculates an ejection time from the detection of the ejection start to the detection of the ejection end, and extracts the maximum ejection time from the ejection times of the multiple ejections. Then, the controller 51 sets an open / close valve V1 using the valve opening speed corresponding to the maximum ejection time. As a result, the operator can easily adjust the ejection start state of a resist nozzle 3 without making a visual determination. In addition, since the operator does not visually make a determination, errors between operators can be suppressed, and errors between devices can also be suppressed. As a result, the quality of a resist film can be stabilized.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method for adjusting a discharge start state of a nozzle that discharges a treatment liquid onto a substrate. Examples of the substrate include a semiconductor substrate, a substrate for an FPD (Flat Panel Display), a glass substrate for a photomask, a substrate for an optical disk, a substrate for a magnetic disk, a ceramic substrate, a substrate for a solar cell, etc. Examples of the FPD include a liquid crystal display device, an organic EL (electroluminescence) display device, etc. [Background technology]

[0002] A conventional substrate processing apparatus includes a substrate holder, a nozzle, and a camera (see, for example, Patent Document 1). The substrate holder holds a substrate. The nozzle ejects a coating liquid onto the substrate. The camera captures an image of an area centered on the tip of the nozzle. The imaging result output from the camera is compared with reference information to determine the presence or absence of coating liquid, etc. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 5457384 Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional substrate processing apparatuses have the following problems. The state when the nozzle starts discharging the processing liquid (including the coating liquid) is called the discharge start state. The discharge start state is adjusted by the speed at which the on-off valve (or discharge valve) is opened (hereinafter referred to as the "valve opening speed"). If the valve opening speed is fast, the valve opens before the processing liquid is pushed out from the pump, so the liquid level inside the nozzle rises (i.e., the processing liquid flows back) before being discharged from the nozzle. If the valve opening speed is slow, the processing liquid is pushed out from the pump before the valve is fully opened (fully open), causing liquid puddles at the nozzle outlet and other disturbances in the liquid flow. Note that liquid puddles are a phenomenon in which liquid remains at the nozzle outlet and the liquid column swells.

[0005] Therefore, adjustments must be made to prevent backflow and disturbances in the liquid flow, such as pooling of liquid. Currently, this adjustment is done manually. Specifically, an operator inputs each parameter into a computer that controls the valve, then visually checks the nozzle and judges the results to make adjustments. However, this method is prone to errors between operators and between devices. Also, adjusting the discharge start state for a large number of nozzles is a tedious task.

[0006] The present invention has been made in consideration of the above circumstances, and has an object to provide a method for adjusting the ejection start state of a nozzle that is easy to adjust and can reduce errors between operators and between devices. [Means for solving the problem]

[0007] In order to achieve the above object, the present invention has the following configuration: That is, a method for adjusting a nozzle discharge start state according to the present invention includes a discharge step of discharging resist liquid from the nozzle multiple times while changing a valve opening speed of an opening / closing valve that starts and stops discharging resist liquid from the nozzle, a photographing step of taking a video of the multiple discharges from the nozzle with a camera, a start detection step of detecting the start of discharge for each of the multiple discharges by a controller based on the video, an end detection step of detecting the end of discharge for each of the multiple discharges by the controller based on the video, a calculation step of calculating a discharge time from detecting the start of discharge to detecting the end of discharge for each of the multiple discharges by the controller, and a calculation step of calculating a discharge time from the discharge start to the discharge end for each of the multiple discharges by the controller. As a feature The method is characterized by comprising an extraction step of extracting a maximum discharge time, and a setting step of setting the on-off valve using a valve opening speed corresponding to the maximum discharge time by the controller.

[0008] According to the method for adjusting the discharge start state of the nozzle according to the present invention, the controller calculates the discharge time from when the discharge start is detected to when the discharge end is detected, and selects the discharge time from among the discharge times of a plurality of discharges. As a feature The discharge time of the maximum value is extracted. Then, the controller sets the opening and closing valve using the valve opening speed corresponding to the discharge time of the maximum value. This makes it easy for the operator to adjust the discharge start state of the nozzle without visually judging. In addition, since the operator does not visually judge, the error between operators is suppressed, and further, the error between devices can also be suppressed. As a result, the quality of the resist film can be stabilized.

[0009] In the above-mentioned method for adjusting the nozzle discharge start state, it is preferable that the controller detects the discharge start when the resist liquid from the nozzle reaches a reference position that is a preset distance away from the tip face of the nozzle in each image data constituting the moving image, thereby making it possible to detect the start of supply of the resist liquid to the reference position as the discharge start.

[0010] In the above-mentioned method for adjusting the nozzle discharge start state, it is preferable that the nozzle discharges the resist liquid toward the substrate surface multiple times in the discharge step, and the reference position is the substrate surface. The discharge start is detected with the substrate surface as the reference position. Therefore, the discharge start state can be set in an environment close to the actual use state.

[0011] In the above-mentioned method for adjusting the nozzle discharge start state, it is preferable that the nozzle discharges the resist liquid toward the collection container multiple times in the discharge step, and the reference position is a horizontal virtual line set in common within each of the image data, thereby making it possible to set the discharge start state without using a substrate.

[0012] In the above-mentioned method for adjusting the nozzle discharge start state, it is preferable that the controller detects the end of discharge when, in each image data constituting the moving image, a gap between the reference position and the tip position of the resist liquid discharged from the nozzle reaches a preset value from a state in which the resist liquid discharged from the nozzle has reached the reference position. The end of discharge can be detected when the resist liquid is no longer supplied to the reference position.

[0013] In addition, in the above-mentioned method for adjusting the discharge start state of the nozzle, it is preferable that the controller selects a valve opening speed within a range from a reference valve opening speed, which is a valve opening speed corresponding to the maximum discharge time, to a preset valve opening speed slower than the reference valve opening speed, and sets the on-off valve using the selected valve opening speed. This makes it possible to select a valve opening speed to be set for the on-off valve within a predetermined range. Also, a range of valve opening speeds is set in a direction slower than the reference valve opening speed. If the valve opening speed becomes faster than the reference valve opening speed, there is a possibility that the liquid will split into two in the nozzle due to backflow, generating bubbles. This generation of bubbles can be prevented.

[0014] Furthermore, a method for adjusting a nozzle discharge start state according to the present invention includes a discharge step of discharging resist liquid from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve that starts and stops discharging resist liquid from the nozzle, a photographing step of taking a video of the plurality of discharges from the nozzle with a camera, a start detection step of detecting the start of discharge for each of the plurality of discharges by a controller based on the video, a calculation step of calculating a discharge delay time from when the controller issues a discharge signal to when the controller detects the start of discharge for each of the plurality of discharges, and a calculation step of calculating a discharge delay time from the discharge delay time for each of the plurality of discharges by the controller. As a feature The method is characterized by comprising an extraction step of extracting a minimum value of the discharge delay time, and a setting step of setting the on-off valve using a valve opening speed corresponding to the minimum value of the discharge delay time by the controller.

[0015] Further, a method of adjusting a nozzle ejection start state according to the present invention includes a ejection step of ejecting a solvent from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve which ejects and stops the ejection of the solvent from the nozzle, a photographing step of taking a video of the plurality of ejections from the nozzle with a camera, a start detection step of detecting the start of ejection for each of the plurality of ejections based on the video by a controller, an end detection step of detecting the end of ejection for each of the plurality of ejections based on the video by the controller, a calculation step of calculating an ejection time from detecting the start of ejection to detecting the end of ejection for each of the plurality of ejections by the controller, and a calculation step of calculating a ejection time from the ejection start to the ejection end for each of the plurality of ejections by the controller. As a feature The method is characterized by comprising an extraction step of extracting a minimum value of the discharge time, and a setting step of setting the on-off valve using a valve opening speed corresponding to the minimum value of the discharge time by the controller.

[0016] Further, a method of adjusting a nozzle ejection start state according to the present invention includes a ejection step of ejecting a solvent from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve which ejects and stops the ejection of the solvent from the nozzle, a photographing step of taking a video of the plurality of ejections from the nozzle with a camera, a start detection step of detecting the start of ejection for each of the plurality of ejections by a controller based on the video, a calculation step of calculating an ejection delay time from when the controller issues an ejection signal to when the controller detects the start of the ejection for each of the plurality of ejections, and a calculation step of calculating a delay time from the ejection delay time for each of the plurality of ejections by the controller. As a feature The method is characterized by comprising an extraction step of extracting a maximum value of the discharge delay time, and a setting step of setting the on-off valve using a valve opening speed corresponding to the maximum value of the discharge delay time by the controller.

[0017] The present specification also discloses an invention relating to the following substrate processing apparatus.

[0018] (1) A substrate processing apparatus includes a nozzle for discharging a resist liquid, an on-off valve for discharging and stopping the discharge of the resist liquid, a camera, and a controller for controlling the on-off valve and the camera. The controller causes the nozzle to discharge the resist liquid multiple times while changing a valve opening speed of the on-off valve. The camera takes a video of the multiple discharges from the nozzle. The controller detects the start and end of discharge for each of the multiple discharges based on the video. The controller calculates a discharge time from detecting the start of discharge to detecting the end of discharge for each of the multiple discharges. The controller extracts a maximum discharge time from the discharge times of the multiple discharges. The controller sets the on-off valve using a valve opening speed corresponding to the maximum discharge time. Effect of the Invention

[0019] According to the method of adjusting the discharge start state of a nozzle according to the present invention, adjustments are easy and errors between operators and between devices can be reduced. [Brief description of the drawings]

[0020] [Figure 1] 1 is a schematic configuration diagram of a substrate processing apparatus according to a first embodiment. [Diagram 2] FIG. 2 is a diagram showing a configuration of an on-off valve. [Diagram 3] FIG. 13 is a diagram for explaining a discharge start state. [Figure 4] FIG. 1A is a diagram for explaining a method for detecting the start of ejection, and FIG. 1B is a diagram for explaining a method for detecting the end of ejection. [Diagram 5] 11 is a time chart for explaining a method of calculating a discharge time, etc. [Figure 6] 11 is a diagram showing an example of the relationship between a valve opening speed and a discharge time when a resist liquid is discharged from a resist nozzle. FIG. [Figure 7] 10 is a flowchart for explaining adjustment of a discharge start state by the substrate processing apparatus according to the first embodiment. [Figure 8] 10 is a flowchart for explaining adjustment of a discharge start state by the substrate processing apparatus according to the second embodiment. [Figure 9] 11 is a diagram showing an example of the relationship between a valve opening speed and a discharge delay time when a resist liquid is discharged from a resist nozzle. [Figure 10] FIG. 4 is a diagram showing an example of the relationship between the valve opening speed and the ejection time when a solvent is ejected from a solvent nozzle. [Figure 11] 11 is a diagram showing an example of the relationship between a valve opening speed and a discharge delay time when a solvent is discharged from a solvent nozzle. FIG. [Figure 12] FIG. 13 is a diagram showing a schematic configuration of a substrate processing apparatus according to a modified example. [Figure 13] FIG. 13A is a diagram for explaining a method for detecting the start of ejection according to a modified example, and FIG. 13B is a diagram for explaining a method for detecting the end of ejection according to a modified example. Example 1

[0021] A first embodiment of the present invention will be described below with reference to the drawings. Fig. 1 is a schematic diagram of a substrate processing apparatus 1 according to the first embodiment.

[0022] <Configuration of the Substrate Processing Apparatus 1> Referring to Fig. 1, a substrate processing apparatus 1 includes a holding / rotating unit 2, a resist nozzle 3, and a solvent nozzle 4. The resist nozzle 3 or the solvent nozzle 4 corresponds to the nozzle of the present invention.

[0023] The holding and rotating unit 2 rotates the substrate W while holding it in a horizontal position. This causes the substrate W to rotate around a vertical axis AX1 at the center of the substrate W. The holding and rotating unit 2 includes a spin chuck 7 and a rotation drive unit 8. The spin chuck 7 holds the back surface of the substrate W, for example, by sucking air from a suction port provided on the support surface for the substrate W. The rotation drive unit 8 includes an electric motor, and rotates the spin chuck 7 around the vertical axis AX1 using the electric motor. Reference numeral 9 denotes a cup.

[0024] The resist nozzle 3 ejects a resist liquid (e.g., a photoresist liquid) onto the substrate W. The resist nozzle 3 is supplied with a resist liquid from a resist liquid supply source 11 through a pipe 13. The resist liquid supply source 11 and a solvent supply source 17, which will be described later, are constituted by containers such as bottles. One end of the pipe 13 is connected to the resist liquid supply source 11, and the other end of the pipe 13 is connected to the resist nozzle 3.

[0025] The piping 13 is provided with a pump 15 and an on-off valve V1 in this order from the resist liquid supply source 11. The pump 15 sends out the resist liquid to the resist nozzle 3 side. The on-off valve V1 discharges the resist liquid and stops the discharge. A suck-back valve may be provided between the resist nozzle 3 and the on-off valve V1. When the on-off valve V1 is in a closed state, the suck-back valve increases the volume of the internal flow path of the suck-back valve, thereby retracting the liquid level of the resist liquid in the resist nozzle 3. When the on-off valve V1 is in an open state, the suck-back valve reduces the volume of the internal flow path of the suck-back valve.

[0026] The solvent nozzle 4 ejects a solvent (e.g., thinner) onto the substrate W. The solvent is supplied to the solvent nozzle 4 from a solvent supply source 17 through a pipe 19. One end of the pipe 19 is connected to the solvent supply source 17, and the other end of the pipe 19 is connected to the solvent nozzle 4. A pump 21 and an on-off valve V2 are provided on the pipe 19, in this order from the solvent supply source 17. The pump 21 sends the solvent to the solvent nozzle 4 side. The on-off valve V2 ejects the solvent and stops the ejection of the solvent. A suck-back valve may be provided between the solvent nozzle 4 and the on-off valve V2.

[0027] The resist nozzle moving mechanism 23 moves the resist nozzle 3 in the horizontal and vertical directions. The solvent nozzle moving mechanism 25 moves the solvent nozzle 4 in the horizontal and vertical directions. The two moving mechanisms 23, 25 each include an electric motor and are driven by the electric motor. The inner diameter of the resist nozzle 3 and the solvent nozzle 4 is approximately 1.0 mm. The inner diameter of each of the nozzles 3, 4 may be 0.5 mm to 2.5 mm.

[0028] 2 is a diagram showing the configuration of on-off valves V1 and V2. The configuration of on-off valve V1 will be described as a representative. The on-off valve V2 is configured similarly to on-off valve V1. On-off valve V1 includes a housing 27, a diaphragm 29 (valve body), a rod 31, a valve seat 33, a flow path 35, an electric motor 37, and a conversion mechanism 39. Flow path 35 includes an upstream flow path 35A, a valve chamber 35B, and a downstream flow path 35C.

[0029] The peripheral portion of the diaphragm 29 is fixed to the inner side surface of the housing 27. The center portion of the diaphragm 29 is connected to one end of a rod 31 that is movable in the vertical direction. A flow path 35 (upstream flow path 35A, valve chamber 35B, and downstream flow path 35C) is formed inside the housing 27. A valve seat 33 for receiving the diaphragm 29 is provided in the valve chamber 35B inside the housing 27. An output shaft 37A of the electric motor is connected to the other end of the rod 31 via a conversion mechanism 39. The conversion mechanism 39 converts the rotation about the vertical axis AX2 by the electric motor 37 into linear movement in the vertical direction. The conversion mechanism 39 includes, for example, a plurality of gears and a guide.

[0030] When the electric motor 37 rotates the output shaft 37A in a first direction around the vertical axis AX2, the center of the diaphragm 29 and the rod 31 descend. As a result, the center of the diaphragm 29 is pressed against the valve seat 33, and the on-off valve V1 is in a closed state (fully closed state). In this case, the resist liquid cannot be sent from the upstream flow path 35A to the downstream flow path 35C. Then, when the electric motor 37 rotates the output shaft 37A in a second direction (the opposite direction to the first direction), the center of the diaphragm 29 and the rod 31 ascend. As a result, the center of the diaphragm 29 moves away from the valve seat 33 by a preset distance. At this time, the on-off valve V1 is in an open state (fully open state). In this case, the resist liquid can be sent from the upstream flow path 35A to the downstream flow path 35C. In FIG. 2, the diaphragm 29 in the closed state is indicated by a dashed line, and the diaphragm 29 in the open state is indicated by a solid line.

[0031] The "valve opening speed" is the time required to change from a fully closed state to a fully open state. Alternatively, the "valve opening speed" is a value obtained by dividing the distance between the fully closed state and the fully open state by the time required to change from the fully closed state to the fully open state. A controller 51, which will be described later, can selectively send commands to the electric motors 37 of the on-off valves V1 and V2. This changes the valve opening speed of the on-off valves V1 and V2. An operator can also selectively send commands to the electric motors 37 of the on-off valves V1 and V2 through an input unit 57 (see FIG. 1), which will be described later.

[0032] Fig. 3 is a diagram for explaining the initial discharge state. That is, Fig. 3 shows the change in the position of the liquid level LQ inside the resist nozzle 3 when the valve opening speed is "fast", "optimum", and "slow". In Fig. 3, reference numeral 41 denotes a flow path inside the resist nozzle 3, and reference numeral 43 denotes a discharge port.

[0033] When the valve opening speed is "fast", the valve body (diaphragm 29) opens before the resist liquid is pushed out from the pump 15. As a result, the resist liquid flows backward once. That is, the liquid level LQ of the resist liquid rises. After that, the resist liquid is discharged from the discharge port 43. When the backward flow occurs, the resist liquid is divided into two in the vertical direction in the flow path 41, and there is a risk of air bubbles being generated in the divided portion. Also, when the valve opening speed is "slow", the resist liquid is pushed out before the valve body (diaphragm 29) is fully opened (fully opened). As a result, a liquid puddle is formed at the discharge port 43 of the resist nozzle 3 due to the viscosity of the resist liquid and the discharge pressure. When a liquid puddle is formed, the flow of the resist liquid discharged from the resist nozzle 3 is disturbed. This affects the film quality.

[0034] In contrast, when the valve opening speed is "optimal," the timing at which the resist liquid is pushed out and the timing at which the valve body (diaphragm 29) opens are balanced. In other words, when the valve opening speed is optimal, the discharge start state is optimal. Conventionally, after changing the valve opening speed, an operator judges the discharge start state by direct visual inspection. In contrast, the substrate processing apparatus 1 of this embodiment automatically finds and sets the optimal discharge start state.

[0035] 2, each of the on-off valves V1, V2 is driven by an electric motor 37. In this regard, each of the on-off valves V1, V2 may be driven by an electro-pneumatic regulator. In this case, each of the on-off valves V1, V2 may be configured to be operated by gas supplied by the electro-pneumatic regulator.

[0036] Referring to Fig. 1, the substrate processing apparatus 1 includes a camera 45, an image processing unit 47, and lighting (not shown). The camera 45 includes an image sensor such as a charge-coupled device (CCD) or a complementary metal-oxide semiconductor (CMOS). The camera 45 is disposed to the side of the resist nozzle 3 and the substrate W. The camera 45 captures an image so that the tip of the resist nozzle 3 and the surface of the substrate W are reflected in the image. The lighting irradiates light onto the range captured by the camera 45.

[0037] The camera 45 captures a video of the state in which the resist nozzle 3 is discharging the resist. The camera 45 captures the video at a preset frame rate (for example, 100 fps (frames / second)) and sequentially acquires image data (still images). The image data acquired by the camera 45 is sequentially sent to the image processing unit 47.

[0038] The image processing unit 47 includes one or more processors and a memory (storage unit) for storing data. The memory stores, for example, a computer program for operating the image processing unit 47. The image processing unit 47 performs image processing on each of the multiple image data sent from the camera 45.

[0039] The substrate processing apparatus 1 includes a controller 51, a storage unit 53, a display unit 55, and an input unit 57. The controller 51 controls each component of the substrate processing apparatus 1 (for example, the holding and rotating unit 2, the two pumps 15 and 21, the two on-off valves V1 and V2, the two moving mechanisms 23 and 25, the camera 45, and the image processing unit 47). The controller 51 includes one or more processors, such as a central processing unit (CPU). The storage unit 53 includes at least one of a read-only memory (ROM), a random-access memory (RAM), and a hard disk. The display unit 55 is, for example, a liquid crystal display or an organic electro-luminescence (EL) display. The input unit 57 includes at least one of a keyboard, a mouse, and various switches.

[0040] The image processing unit 47 may be configured as a computer program (software) that runs on the controller 51 and is stored in the storage unit 53.

[0041] <Basic operation of the substrate processing apparatus 1> Here, the basic operation of the substrate processing apparatus 1 will be described. A transport robot (not shown) transports the substrate W onto the holding and rotating unit 2. The holding and rotating unit 2 holds the substrate W and rotates it. The solvent nozzle moving mechanism 25 moves the solvent nozzle 4 from the standby position to above the center of the substrate W. Thereafter, the pump 21 pumps out the solvent. The on-off valve V2 is opened. This causes the solvent to be discharged from the solvent nozzle 4 onto the substrate W. The solvent discharged onto the substrate W spreads over the substrate W due to the rotation of the substrate W. This forms a solvent film on the substrate W.

[0042] When the discharge of the solvent from the solvent nozzle 4 is to be stopped, the pump 21 stops sending out the solvent and the on-off valve V2 is closed. After that, the solvent nozzle moving mechanism 25 moves the solvent nozzle 4 to the standby position.

[0043] Instead of the solvent nozzle 4, the resist nozzle moving mechanism 23 moves the resist nozzle 3 from the standby position to above the center of the substrate W. Thereafter, the pump 15 pumps out the resist liquid. Also, the on-off valve V1 is opened. This causes the resist liquid to be discharged from the resist nozzle 3 onto the solvent film on the substrate W. The resist liquid discharged onto the solvent film on the substrate W spreads over the substrate W while replacing the solvent film due to the rotation of the substrate W. As a result, a resist film is formed on the substrate W.

[0044] When the discharge of the resist solution from the resist nozzle 3 is to be stopped, the pump 15 stops sending out the resist solution and the on-off valve V1 is closed. Thereafter, the resist nozzle moving mechanism 23 moves the resist nozzle 3 to the standby position. After the resist film is formed, the holding and rotating part 2 releases the hold of the substrate W while stopping the rotation of the substrate W. A transport robot (not shown) transports the substrate W on which the resist film has been formed from the holding and rotating part 2.

[0045] <Outline of the characteristic parts of this embodiment> Next, the characteristic features of this embodiment will be described. As described above, the substrate processing apparatus 1 automatically finds and sets an optimal discharge start state. First, an overview will be described. FIG. 4(a) is a diagram for explaining a method for detecting the start of discharge. FIG. 4(b) is a diagram for explaining a method for detecting the end of discharge. FIG. 5 is a time chart for explaining a method for calculating the discharge time. FIG. 6 is a diagram showing an example of the relationship between the valve opening speed and the discharge time.

[0046] The controller 51 discharges the resist liquid multiple times from the resist nozzle 3 while changing the valve opening speed of the on-off valve V1. The camera 45 captures a video of the multiple discharges from the resist nozzle 3. The image processor 47 detects the start of each of the multiple discharges and detects the end of each of the multiple discharges based on the video (see FIG. 4(a) and FIG. 4(b)).

[0047] The controller 51 calculates the discharge time DT for each of the multiple discharges (see FIG. 5). The discharge time DT is the time from when the start of discharge is detected to when the end of discharge is detected (time t3-time t1). This provides multiple (e.g., nine) discharge times DT corresponding to multiple (e.g., nine) discharges. If these multiple discharge times DT are graphed, it will look like FIG. 6. In FIG. 6, the horizontal axis indicates the valve opening speed. The valve opening speed is expressed as the time (seconds: sec) required to go from a fully closed state to a fully open state. The vertical axis indicates the discharge time (milliseconds: msec).

[0048] In Fig. 6, the discharge time DT is distributed in a mountain shape, and the maximum discharge time DT appears as a characteristic point. The controller 51 extracts the maximum discharge time DTmax from the discharge times DT of multiple discharges. The controller 51 then sets the on-off valve V1 using the valve opening speed corresponding to the maximum discharge time DTmax. Details will be described later.

[0049] <Method of Adjusting the Discharge Start State of the Resist Nozzle 3> Next, a method for adjusting the discharge start state of the resist nozzle 3 of the substrate processing apparatus 1 will be described. Fig. 7 is a flowchart showing the operation of the substrate processing apparatus 1 according to the first embodiment. This adjustment is performed when the substrate processing apparatus 1 is started up or when the film thickness is adjusted thereafter. However, the adjustment is not limited to this, and the adjustment of the discharge start state may be performed at any timing.

[0050] [Step S01] Instruction to adjust discharge start condition to controller 51 An operator instructs the controller 51 to adjust the discharge start state through the input unit 57. The resist nozzle 3 is moved to above the substrate W held by the holding and rotating unit 2 by the resist nozzle moving mechanism .

[0051] [Step S02] Start video recording The controller 51 starts capturing video with the camera 45. The camera 45 captures video so as to capture the tip of the resist nozzle 3 and the liquid landing point on the surface of the substrate W. The camera 45 sequentially acquires image data (still images) and sequentially transmits the acquired image data to the image processing unit 47.

[0052] [Step S03] Provisional setting of valve opening speed For example, nine conditions SP1 to SP9 that change stepwise with respect to the valve opening speed (seconds) are stored in advance in the storage unit 53. The number of conditions is not limited to nine. In the first case, the controller 51 reads out the condition SP1 of the nine conditions SP1 to SP9 from the storage unit 53 and sends the condition SP1 to the on-off valve V1. The on-off valve V1 provisionally sets the valve opening speed based on the condition SP1. In the second case, the controller 51 sends the condition SP2 to the on-off valve V1, and in the third case, sends the condition SP3 to the on-off valve V1.

[0053] [Step S04] Discharge command from controller 51 At time t0, the controller 51 sends a discharge signal "ON" to the on-off valve V1 and the pump 15. Based on this signal, the pump 15 sends out the resist liquid to the resist nozzle 3 side. Also, at the provisionally set valve opening speed, the on-off valve V1 is opened, i.e., changed from a fully closed state to a fully open state. This causes the resist liquid to be discharged from the resist nozzle 3 and land on the substrate W. This state is photographed by the camera 45. When the resist liquid is discharged, the substrate W may be rotated by the holding and rotating part 2, or may be stopped.

[0054] [Step S05] Discharge stop command from the controller 51 After that, at time t2, the controller 51 sends a discharge stop signal "OFF" to the on-off valve V1 and the pump 15. Based on this signal, the pump 15 stops sending out the resist liquid, and the on-off valve V1 is closed, that is, changes from a fully open state to a fully closed state. This stops the discharge of the resist liquid from the resist nozzle 3. This state is photographed by the camera 45.

[0055] The valve closing speed is constant during the nine discharges. The operation of the pump 15 when the on-off valve V1 is opened and closed (for example, the pressure to push out the liquid) is also constant. Even if a suck-back valve is provided, its operation is also constant. In addition, the "valve closing speed" is the time required to go from a fully open state to a fully closed state.

[0056] The controller 51 repeats the operations of steps S03 to S05 described above based on, for example, nine conditions SP1 to SP9. That is, the controller 51 discharges the resist liquid from the resist nozzle 3 toward the surface of the substrate W multiple times (for example, nine times) while changing the valve opening speed of the on-off valve V1. The camera 45 also captures a video of the nine discharges from the resist nozzle 3. The camera 45 continues to capture the video until the final, ninth discharge is completed.

[0057] [Step S06] Detection of discharge start and end The video captured by the camera 45 is sent to the image processing unit 47 every time image data is acquired. First, the image processing unit 47 judges whether or not the position of the resist nozzle 3 shown in each image data is correct. This judgment is performed by storing a reference image in advance in a memory and performing pattern matching. In this explanation, it is assumed that the resist nozzle 3 is shown in the correct position in each image data.

[0058] The image processor 47 detects the start and end of discharge for each of the nine discharges based on each image data in which the resist nozzle 3 is shown in the correct position.

[0059] Please refer to Fig. 4(a). In Fig. 4(a) and Fig. 4(b), the symbol RS denotes resist liquid. The image processing unit 47 detects the start of discharge when the resist liquid from the resist nozzle 3 reaches the surface of the substrate W at a position a preset distance GP1 away from the tip surface 3A of the resist nozzle 3. The image data when the start of discharge is detected is image data G1. When the arrival of the resist liquid is detected for two or more pieces of image data, for example, the earliest image data is selected.

[0060] In each image data, the resist liquid discharged from the resist nozzle 3 appears white. The tip position TP of the resist liquid that appears white is identified. The surface position of the substrate W in each image data is stored in memory in advance. Then, when the identified tip position TP of the resist liquid reaches the surface of the substrate W, the image processing unit 47 detects the start of discharge. In the method of detecting the start of discharge and the end of discharge described later, processes such as edge extraction, binarization, black and white inversion, and difference may be selectively adopted. For example, the difference process is performed by subtracting two pieces of image data at different times. The image processing unit 47 may detect the start of discharge by performing pattern matching using a reference image stored in memory in advance.

[0061] Refer to Fig. 4(b). The image processing unit 47 detects the end of discharge when the gap between the surface of the substrate W and the tip position TP of the resist liquid discharged from the resist nozzle 3 from the state where the resist liquid discharged from the resist nozzle 3 has reached the surface of the substrate W becomes a preset value GP2 (distance). In Fig. 4(b), for example, it is assumed that the gap between the tip surface 3A of the resist nozzle 3 and the surface of the substrate W is distance GP1. At this time, the value GP2 is set, for example, as a value of 15% (arbitrary) of the distance GP1 (GP2 = GP1 x 0.15).

[0062] The image processing unit 47 identifies the leading edge position TP of the resist liquid. Then, when the gap between the identified leading edge position TP and the surface of the substrate W reaches a value GP2, the image processing unit 47 detects the end of discharge. The image processing unit 47 may detect the end of discharge by performing pattern matching using a reference image previously stored in the memory. The image data for which the end of discharge is detected is set as image data G3.

[0063] Each image data constituting the moving image is associated with a time. For example, the image processing unit 47 transmits to the controller 51, for each of nine ejections, the time t1 (see FIG. 5) of the image data G1 at which the ejection start is detected and the time t3 of the image data G3 at which the ejection end is detected.

[0064] [Step S07] Calculate the discharge time DT The controller 51 calculates the discharge time DT for each of the nine discharges. The discharge time DT is the time from when the start of discharge is detected to when the end of discharge is detected. That is, as shown in FIG. 6, the discharge time DT is the time from time t1 of the image data G1 to time t3 of the image data G3 (time t3-time t1). This process provides nine discharge times DT corresponding to the nine discharges. The calculation in this process may be performed by the image processing unit 47, not by the controller 51.

[0065] [Step S08] Extraction of characteristic values ​​(feature points) The controller 51 extracts the maximum discharge time DTmax from the nine discharge times DT corresponding to the nine discharges. That is, in a graph (see FIG. 6) with the valve opening speed on the horizontal axis and the discharge time on the vertical axis, the nine discharge times DT tend to show a mountain-shaped distribution. Therefore, the maximum discharge time DTmax is extracted as a characteristic point.

[0066] [Step S09] Setting the valve opening speed The controller 51 sets the on-off valve V1 using a valve opening speed (for example, condition SP5) corresponding to the maximum discharge time DTmax. That is, the substrate processing apparatus 1 automatically sets the valve opening speed of the on-off valve V1 to an optimal state. This sets the discharge start state of the resist nozzle 3 to an optimal state.

[0067] <Supplementary explanation for Figure 6> In Fig. 6, the horizontal axis indicates the valve opening speed. The valve opening speed is expressed as the time (seconds: sec) required to go from a fully closed state to a fully open state. The vertical axis indicates the discharge time DT (msec). The horizontal and vertical axes are shown as relative values, with the characteristic point set to 0 (zero).

[0068] In Figure 6, when the valve opening speed is changed stepwise and the discharge time DT after the change is plotted, the distribution of the discharge time DT forms a mountain shape. In this mountain-shaped distribution, the maximum value is the characteristic point. When the valve opening speed corresponding to this maximum value of the discharge time DTmax is set to the on-off valve V1 and observed directly with the naked eye, it shows the optimal discharge start state (see Figure 3).

[0069] In addition, in FIG. 6, the following can be considered. When the valve opening speed is fast with respect to the characteristic point (left side), the resist liquid flows back slowly in the resist nozzle 3, so it takes time for the liquid to land. Therefore, the discharge time DT is shortened. When the valve opening speed is slow with respect to the characteristic point (right side), the valve body (diaphragm 29) opens slowly, so the resist liquid is difficult to come out. Therefore, the discharge time DT is shortened. In other words, it is considered that the influence of the valve opening speed being fast or slow around the characteristic point is reflected in the discharge time DT. In addition, the results in FIG. 6 are the results when the resist liquid is 2 to 4 cP (centipoise). It is considered that the same mountain-shaped distribution is shown even if the viscosity of the resist liquid is high. In other words, it is considered that the same mountain-shaped distribution is shown even if the viscosity of the resist liquid is other than 2 to 4 cP (greater than zero).

[0070] According to this embodiment, the controller 51 calculates the discharge time DT (time t3-time t1) from when the discharge start is detected to when the discharge end is detected, and extracts the maximum discharge time DTmax from the discharge times DT of multiple discharges. Then, the controller 51 sets the on-off valve V1 using the valve opening speed corresponding to the maximum discharge time DTmax. This makes it easy for the operator to adjust the discharge start state of the resist nozzle 3 without visually judging. In addition, since the operator does not visually judge, errors between operators are suppressed, and errors between devices can also be suppressed. As a result, the quality of the resist film can be stabilized.

[0071] Furthermore, the image processing unit 47 detects the start of discharge when the resist liquid from the resist nozzle 3 reaches the surface of the substrate W, which is a reference position that is a preset distance GP1 away from the tip surface 3A of the resist nozzle 3, in each piece of image data constituting the moving image. This makes it possible to detect the start of supply of the resist liquid to the reference position as the start of discharge. Furthermore, the start of discharge is detected with the surface of the substrate W as the reference position. This makes it possible to set the discharge start state in an environment that is close to the actual usage state.

[0072] Also, it is assumed that the resist liquid discharged from the resist nozzle 3 has reached the surface of the substrate W. From this state, the image processing unit 47 detects the end of discharge when the gap between the surface of the substrate W and the tip position TP of the resist liquid discharged from the resist nozzle 3 in each piece of image data constituting the moving image reaches a preset value GP2. This makes it possible to detect the end of discharge as the fact that the resist liquid is no longer being supplied to the surface of the substrate W. Example 2

[0073] Next, a second embodiment of the present invention will be described with reference to the drawings. Note that the description overlapping with the first embodiment will be omitted.

[0074] In the first embodiment, a plurality of (nine) ejection times DT corresponding to a plurality of (for example, nine) ejections are calculated. In this regard, in the second embodiment, a plurality of ejection delay times DL corresponding to a plurality of ejections are calculated.

[0075] Please refer to FIG. 5. The discharge delay time DL is the time from when the controller 51 issues a discharge signal "ON" to when it detects the start of discharge. That is, at time t0, the controller 51 issues a discharge signal "ON" to the on-off valve V1, etc. Meanwhile, each image data (still image) constituting the video is associated with a time. The image processing unit 47 detects the start of discharge based on the video (each image data). Time t1 associated with the image data G1 when the start of discharge is detected is obtained. As a result, the discharge delay time DL is calculated as time t1 - time t0.

[0076] <Method of Adjusting the Discharge Start State of the Resist Nozzle 3> Next, a method for adjusting the discharge start state of the resist nozzle 3 of the substrate processing apparatus 1 of this embodiment will be described. Fig. 8 is a flowchart showing the operation of the substrate processing apparatus 1. Note that some of the operations that are the same as those in the flowchart of Fig. 7 will not be described.

[0077] The operator instructs the controller 51 to adjust the discharge start state through the input unit 57 (step S11). The controller 51 causes the camera 45 to start taking a moving image (step S12).

[0078] [Step S13] Provisional setting of valve opening speed For example, nine conditions SP1 to SP9 that change stepwise with respect to the valve opening speed (seconds) are stored in advance in the storage unit 53. In the first case, the controller 51 reads out the condition SP1 of the nine conditions SP1 to SP9 from the storage unit 53 and sends the condition SP1 to the on-off valve V1. The on-off valve V1 provisionally sets the valve opening speed based on the condition SP1.

[0079] [Step S14] Discharge command from controller 51 At time t0, the controller 51 sends a discharge signal "ON" to the on-off valve V1, the pump 15, and the camera 45. Based on this signal, the pump 15 discharges the resist liquid to the resist nozzle 3 side. Also, the on-off valve V1 is opened, i.e., changed from a fully closed state to a fully open state, at the temporarily set valve opening speed. This causes the resist liquid to be discharged from the resist nozzle 3, and the resist liquid lands on the substrate W. This state is captured by the camera 45.

[0080] [Step S15] Discharge stop command from the controller 51 After that, at time t2, the controller 51 sends a discharge stop signal "OFF" to the on-off valve V1, the pump 15, and the camera 45. The pump 15 stops sending out the resist liquid, and the on-off valve V1 is closed, that is, changed from a fully open state to a fully closed state. This causes the discharge of the resist liquid from the resist nozzle 3 to stop. This state is captured by the camera 45.

[0081] The valve closing speed is constant during the nine discharges. The operation of the pump 15 (for example, the pressure to push out the liquid) when the on-off valve V1 is opened and closed is also constant. Even if a suck-back valve is provided, its operation is also constant.

[0082] The controller 51 repeats the operations of steps S13 to S15 described above based on, for example, nine conditions SP1 to SP9. That is, the controller 51 discharges the resist liquid from the resist nozzle 3 multiple times (for example, nine times) while changing the valve opening speed of the on-off valve V1. The camera 45 also captures a video of the nine discharges from the resist nozzle 3.

[0083] [Step S16] Detection of discharge start The image processing unit 47 detects the start of discharge for every nine discharges based on the multiple image data constituting the video. Specifically, as shown in Fig. 4(a), the image processing unit 47 detects the start of discharge when the resist liquid from the resist nozzle 3 reaches the surface of the substrate W, which is a position a preset distance GP1 away from the tip surface 3A of the resist nozzle 3. The image data when the start of discharge is detected is set as image data G1. When the arrival of the resist liquid is detected for two or more pieces of image data, the earliest image data is selected.

[0084] Each image data constituting the video is associated with a time. For example, the image processing unit 47 transmits to the controller 51, for every nine ejections, the time t1 (see FIG. 5) associated with the image data G1 at which the start of ejection is detected. Note that image data captured at the same time that the controller 51 issues an "ON" ejection signal is set as image data G0. The image data G0 may also have information that the "ON" ejection signal has been issued. In this case, the time t0 associated with the image data G0 may be transmitted to the controller 51.

[0085] [Step S17] Calculate the discharge delay time DL The controller 51 calculates the discharge delay time DL for each of the nine discharges. The discharge delay time DL is the time from when the controller 51 issues the discharge signal "ON" to when it detects the start of discharge (time t1-time t0). Time t0 may be the stored time when the controller 51 issues the discharge signal "ON". Alternatively, time t0 may be the time associated with the image data G0. This process provides nine discharge delay times DL corresponding to the nine discharges.

[0086] [Step S18] Extraction of characteristic values ​​(feature points) The controller 51 extracts the minimum discharge delay time DLmin from the nine discharge delay times DL corresponding to the nine discharges. Now, Fig. 9 will be described. Fig. 9 is a diagram showing an example of the relationship between the valve opening speed and the discharge delay time DL.

[0087] In Fig. 9, the horizontal axis indicates the valve opening speed (seconds: sec). The vertical axis indicates the discharge delay time DL (msec). A positive value of the discharge delay time DL indicates "delay." The horizontal and vertical axes are shown as relative values, with 0 (zero) as the characteristic point.

[0088] In FIG. 9, when the valve opening speed is changed stepwise and the discharge delay time DL after the change is plotted, the distribution of the discharge delay time DL becomes a "valley type". In this valley type distribution, the minimum value is the characteristic point. When the valve opening speed corresponding to this minimum value of the discharge delay time DLmin is set for the on-off valve V1 and directly observed with the naked eye, it shows the optimal discharge start state. That is, in a graph with the valve opening speed on the horizontal axis and the discharge delay time on the vertical axis, the nine discharge times DT tend to show a valley type distribution as shown in FIG. 9. Therefore, the minimum value of the discharge delay time DLmin is extracted as the characteristic point.

[0089] [Step S19] Setting the valve opening speed The controller 51 sets the on-off valve V1 using a valve opening speed (for example, condition SP5) corresponding to the minimum value of the discharge delay time DLmin. That is, the substrate processing apparatus 1 automatically sets the valve opening speed of the on-off valve V1 to an optimal state. This sets the discharge start state of the resist nozzle 3 to an optimal state.

[0090] According to this embodiment, the controller 51 calculates the discharge delay time DL (time t1-time t0) from when the controller 51 issues a discharge signal to when it detects the start of discharge, and extracts the minimum discharge delay time DLmin from the discharge delay times DL of multiple discharges. Then, the controller 51 sets the on-off valve V1 using the valve opening speed corresponding to the minimum discharge delay time DLmin. This makes it easy for the operator to adjust the discharge start state of the resist nozzle 3 without visually judging. In addition, since the operator does not visually judge, errors between operators are suppressed, and errors between devices can also be suppressed. As a result, the quality of the resist film can be stabilized. Example 3

[0091] Next, a third embodiment of the present invention will be described with reference to the drawings. Note that the description overlapping with the first and second embodiments will be omitted.

[0092] In the first embodiment, the substrate processing apparatus 1 automatically adjusted the discharge start state of the resist nozzle 3. In this regard, in the third embodiment, the substrate processing apparatus 1 automatically adjusts the discharge start state of the solvent nozzle 4 that discharges a solvent (e.g., thinner). In the present embodiment, the controller 51 calculates a discharge time DT (time t3-time t1) from when the discharge start is detected to when the discharge end is detected.

[0093] A specific method for adjusting the discharge start state of the solvent nozzle 4 is performed in the same procedure as that of the flowchart shown in Fig. 7 in the first embodiment, except for step S08. Therefore, step S08 will be described.

[0094] [Step S08] Extraction of characteristic values ​​(feature points) The controller 51 extracts the minimum discharge time DTmin from the nine discharge times DT corresponding to the nine discharges. Fig. 10 is a diagram showing an example of the relationship between the valve opening speed and the discharge time DT. As in Fig. 6, the horizontal axis indicates the valve opening speed (seconds: sec). The vertical axis indicates the discharge time DT (milliseconds: msec). The horizontal and vertical axes are shown as relative values, with the characteristic point being 0 (zero).

[0095] In Figure 10, when the valve opening speed is changed stepwise and the discharge time after the change is plotted, the distribution of the discharge time forms a valley. In this valley-shaped distribution, the minimum value is the characteristic point. When the valve opening speed corresponding to this minimum discharge time DTmin is set for the on-off valve V2 and observed directly with the naked eye, it shows the optimal discharge start state. Therefore, the minimum discharge time DTmin is extracted as the characteristic point.

[0096] In addition, the following can be considered in FIG. 10. When the valve opening speed is fast with respect to the characteristic point (left side), the response is fast and the amount of change is large due to the nature of the solvent (thinner). In addition, after backflow, the solvent is forcefully discharged by its own weight, and the discharge time DT becomes long. In addition, when the valve opening speed is slow with respect to the characteristic point (right side), the solvent is discharged before the valve body (diaphragm 29) is fully opened. This increases the discharge pressure, and the solvent is forcefully discharged. The discharge time DT becomes long. When the discharge pressure reaches a certain level, the timing of discharge becomes the same and saturation occurs. In addition, thinner is unlikely to form a liquid pool. In the case of the solvent, it is considered that the effect of a fast valve opening speed or a slow valve opening speed is reflected in the discharge time DT before and after the characteristic point. In addition, the results in FIG. 10 are the results when the thinner is less than 1 cP (centipoise) (greater than zero). It is considered that even if the viscosity of the thinner is 1 cP or more, a valley-shaped distribution will be similarly shown.

[0097] According to this embodiment, the controller 51 calculates the discharge time DT (time t3-time t1) from when the discharge start is detected to when the discharge end is detected, and extracts the minimum discharge time DTmin from the discharge times DT of multiple discharges. Then, the controller 51 sets the on-off valve V2 using the valve opening speed corresponding to the minimum discharge time DTmin. This makes it easy for the operator to adjust the discharge start state of the solvent nozzle 4 without visually judging. In addition, since the operator does not visually judge, errors between operators are suppressed, and errors between devices can also be suppressed. As a result, the quality of the solvent film can be stabilized. Example 4

[0098] Next, a fourth embodiment of the present invention will be described with reference to the drawings. Note that the description overlapping with the first to third embodiments will be omitted.

[0099] In the second embodiment, the substrate processing apparatus 1 automatically adjusted the discharge start state of the resist nozzle 3. In this regard, in the fourth embodiment, the substrate processing apparatus 1 automatically adjusts the discharge start state of the solvent nozzle 4 that discharges a solvent (e.g., thinner). In the present embodiment, the controller 51 calculates a discharge delay time DL (time t1-time t0) from when the controller 51 issues a discharge signal to when it detects the start of discharge.

[0100] A specific method for adjusting the discharge start state of the solvent nozzle 4 is performed in the same procedure as that of the flowchart shown in Fig. 8 of the second embodiment, except for step S18. Therefore, step S18 will be described.

[0101] [Step S18] Extraction of characteristic values ​​(feature points) The controller 51 extracts the maximum ejection delay time DLmax from the nine ejection delay times DL corresponding to the nine ejections. Now, Fig. 11 will be described. Fig. 11 is a diagram showing an example of the relationship between the valve opening speed and the ejection delay time DL.

[0102] In Fig. 11, the horizontal axis indicates the valve opening speed (seconds: sec). The vertical axis indicates the discharge delay time DL (msec). A positive value of the discharge delay time DL indicates "delay." The horizontal and vertical axes are shown as relative values, with 0 (zero) as the characteristic point.

[0103] In FIG. 11, when the valve opening speed is changed stepwise and the discharge delay time DL after the change is plotted, the distribution of the discharge delay time DL becomes a "mountain shape." In this mountain-shaped distribution, the maximum value is the characteristic point. When the valve opening speed corresponding to this maximum value of the discharge delay time DLmax is set for the on-off valve V2 and directly observed with the naked eye, it shows the optimal discharge start state. That is, in a graph with the valve opening speed on the horizontal axis and the discharge delay time DL on the vertical axis, the nine discharge delay times DL tend to show a mountain-shaped distribution as shown in FIG. 11. Therefore, the maximum value of the discharge delay time DLmax is extracted as the characteristic point.

[0104] According to this embodiment, the controller 51 calculates the discharge delay time DL (time t1-time t0) from when the controller 51 issues a discharge signal to when it detects the start of discharge, and extracts the maximum discharge delay time DLmax from the discharge delay times DL of multiple discharges. Then, the controller 51 sets the on-off valve V2 using the valve opening speed corresponding to the maximum discharge delay time DLmax. This makes it easy for the operator to adjust the discharge start state of the solvent nozzle 4 without having to make a visual judgment. In addition, since the operator does not make a visual judgment, the error between operators is suppressed, and further the error between devices can also be suppressed. As a result, the quality of the solvent film can be stabilized.

[0105] The present invention is not limited to the above-described embodiment, but can be modified as follows.

[0106] (1) In each of the above-described embodiments, a feature point is extracted from among a plurality of discharge times DT or a plurality of discharge delay times DL, and a valve opening speed corresponding to the extracted feature point is set for the on-off valve V1 or the on-off valve V2. In this regard, the valve opening speed set for the on-off valve V1, etc. may be selected from a predetermined range.

[0107] A specific description will be given with reference to Fig. 6. The controller 51 selects a valve opening speed within a range RG from a reference valve opening speed SS, which is a valve opening speed corresponding to the maximum discharge time TD (characteristic point), to a preset valve opening speed KS which is slower than the reference valve opening speed SS, and sets the on-off valve V1 using the selected valve opening speed.

[0108] In Fig. 6, the valve opening speed KS is, for example, +0.02 sec with respect to the reference valve opening speed SS. Therefore, the valve opening speed to be set for the on-off valve V1 is determined within a range RG of the reference valve opening speed SS +0.02 sec. For example, the average value of the range RG (reference valve opening speed SS +0.01 sec) may be determined as the valve opening speed to be set for the on-off valve V1. The same applies to Figs. 9, 10, and 11. The valve opening speed KS (range RG) may be set in advance for each processing liquid, for example, by directly visually observing the discharge start state through an experiment.

[0109] According to this modification, the valve opening speed to be set for each on-off valve V1, V2 can be selected within a predetermined range RG. Also, a range RG of the valve opening speed is set in a direction slower than the reference valve opening speed. If the valve opening speed becomes faster than the reference valve opening speed, a backflow may cause the liquid to split into two in, for example, the resist nozzle 3, generating bubbles. This generation of bubbles can be prevented.

[0110] (2) In each of the above-described embodiments and modified example (1), for example, when the resist nozzle 3 is moved above the center of the substrate W, the camera 45 photographs the tip of the resist nozzle 3 so as to capture the tip of the resist nozzle 3 and the landing point on the surface of the substrate W. In this regard, the camera 45 may photograph the tip of the resist nozzle 3 when the resist nozzle 3 is moved above the collection container 61 shown in FIG.

[0111] The substrate processing apparatus 1 includes a collection container 61 for maintenance. The collection container 61 is provided beside the holding rotation unit 2 and the cup 9. The top surface of the collection container 61 is open, and a discharge port 63 is provided on the bottom surface of the collection container 61. The collection container 61 receives the resist liquid discharged from the resist nozzle 3. The resist liquid received in the collection container 61 is discharged through the discharge port 63. The camera 45 captures an image of the tip of the resist nozzle 3 that has been moved above the collection container 61.

[0112] 7 will be described. The resist nozzle 3 is moved to above the collection container 61. The resist liquid is discharged from the resist nozzle 3 toward the collection container 61 multiple times. A video of the state in which the resist nozzle 3 is discharged multiple times is captured by the camera 45.

[0113] The image processing unit 47 detects the start of discharge when the resist liquid from the resist nozzle 3 reaches a reference position that is a preset distance GP1 away from the tip surface 3A of the resist nozzle 3 in each image data constituting the moving image. Here, the reference position shown in Figures 13(a) and 13(b) is a horizontal virtual line 65 that is set in common in each image data. The start of discharge is detected when the tip position TP of the resist liquid reaches the virtual line 65.

[0114] In addition, the image processing unit 47 detects the end of ejection when, in each image data constituting the video, the gap between the virtual line 65 and the tip position TP of the resist liquid ejected from the resist nozzle 3, from the state in which the resist liquid has reached the virtual line 65, becomes a preset value GP2.

[0115] The start and end of discharge are detected using the virtual line 65 as a reference position. Therefore, the discharge start state can be set without using the substrate W. The same applies to the flowchart of FIG. 8 and to the case where the solvent is discharged from the solvent nozzle 4.

[0116] (3) In each of the above-described embodiments and modifications, a resist liquid or a solvent (e.g., thinner) is discharged in order to automatically adjust the discharge start state of each of the nozzles 3 and 4. The processing liquid to be discharged is not limited to these. For example, the liquid discharged from the nozzle may be IPA (isopropyl alcohol). In other words, the discharge start state of the nozzle that discharges IPA may be automatically adjusted. It should be noted that IPA is considered to be included in the solvent.

[0117] (4) In the above-described embodiments, the controller 51 calculates the discharge time DT for each of the multiple discharges. The calculated discharge time DT may be plotted on a graph and displayed on the display unit 55, as shown in Fig. 6 and Fig. 10. The same is true for the calculated discharge delay time DL.

[0118] In each of the above-mentioned embodiments and modifications, for example, the controller 51 extracts the maximum discharge time DTmax from among a plurality of discharge times DT. The controller 51 may display a graph (see FIG. 6) on the display unit 55 and may also display the extracted maximum discharge time DTmax (corresponding valve opening speed) superimposed on the graph. The operator looks at the graph and changes the extracted maximum discharge time DTmax to another discharge time DT via the input unit 57.

[0119] (5) In the above-described embodiments, the controller 51 or the image processing unit 47 is responsible for steps S03 to S09. In this regard, steps S03 to S09 may be divided among a plurality of processors. For example, the first processor may be responsible for step S07, and the second processor may be responsible for step S08.

[0120] (6) In each of the above-mentioned embodiments and examples, nine conditions SP1 to SP9 were prepared in step S03 or step S13. Although one ejection was performed in each of the conditions SP1 to SP9, for example, three ejections may be performed in each of the conditions SP1 to SP9 (a total of 27 ejections). In this case, for example, three ejection times DT are calculated based on the three ejections of the condition SP1. The average value of the three calculated ejection times DT may be used as the ejection time DT of the condition SP1. [Explanation of symbols]

[0121] 1 ... Substrate processing equipment 3 ... Resist nozzle 3A … Tip surface 4 … Solvent nozzle 45 ... Camera 47 ... Image processing section 51 ... Controller 61 … Collection container 65 … Virtual line V1, V2 … Open / close valve W: Substrate DT…Discharge time DL: Discharge delay time

Claims

1. a discharge step of discharging the resist liquid from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve which starts and stops discharging the resist liquid from the nozzle; a step of taking a video of a state in which ejection is performed multiple times from the nozzle with a camera; a start detection step of detecting a start of ejection for each of the plurality of ejections by a controller based on the moving image; an end detection step of detecting an end of ejection for each of the plurality of ejections by the controller based on the moving image; a calculation step of calculating, by the controller, a discharge time from detection of the discharge start to detection of the discharge end for each of the plurality of discharges; an extraction step of extracting a maximum ejection time as a feature point from the ejection times of the plurality of ejections by the controller; a setting step of setting the on-off valve by the controller using a valve opening speed corresponding to the maximum discharge time; A method for adjusting a nozzle start state, comprising:

2. 2. The method for adjusting a nozzle discharge start state according to claim 1, The method for adjusting the nozzle ejection start state is characterized in that the controller detects the start of ejection when the resist liquid from the nozzle reaches a reference position that is a predetermined distance away from the tip surface of the nozzle in each image data that constitutes the video.

3. 3. The method for adjusting a nozzle discharge start state according to claim 2, In the discharging step, the nozzle discharges the resist liquid toward the substrate surface a plurality of times; A method for adjusting a nozzle's discharge start state, wherein the reference position is the surface of the substrate.

4. 3. The method for adjusting a nozzle discharge start state according to claim 2, In the discharging step, the nozzle discharges the resist liquid toward a collection container a plurality of times; A method for adjusting a nozzle ejection start state, wherein the reference position is a horizontal imaginary line that is set in common within each of the image data.

5. 5. The method for adjusting a nozzle discharge start state according to claim 1, further comprising the steps of: The method for adjusting the nozzle ejection start state is characterized in that the controller detects the end of ejection when, in each image data constituting the video, a gap between the reference position and the tip position of the resist liquid ejected from the nozzle reaches a preset value from a state in which the resist liquid ejected from the nozzle has reached the reference position.

6. 6. The method for adjusting a nozzle discharge start state according to claim 1, further comprising the steps of: the controller selects a valve opening speed within a range from a reference valve opening speed, which is a valve opening speed corresponding to the maximum value of the discharge time, to a preset valve opening speed which is slower than the reference valve opening speed, and sets the opening / closing valve using the selected valve opening speed.

7. a discharge step of discharging the resist liquid from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve which starts and stops discharging the resist liquid from the nozzle; a step of taking a video of a state in which ejection is performed multiple times from the nozzle with a camera; a start detection step of detecting a start of ejection for each of the plurality of ejections by a controller based on the moving image; a calculation step of calculating, by the controller, a discharge delay time from when the controller issues a discharge signal to when the controller detects the start of discharge, for each of the plurality of discharges; an extraction step of extracting a minimum ejection delay time as a feature point from the ejection delay times of the plurality of ejections by the controller; a setting step of setting the on-off valve by the controller using a valve opening speed corresponding to the minimum value of the discharge delay time; A method for adjusting a nozzle start state, comprising:

8. a discharge step of discharging the solvent from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve which starts and stops discharging the solvent from the nozzle; a step of taking a video of a state in which ejection is performed multiple times from the nozzle with a camera; a start detection step of detecting a start of ejection for each of the plurality of ejections by a controller based on the moving image; an end detection step of detecting an end of ejection for each of the plurality of ejections by the controller based on the moving image; a calculation step of calculating, by the controller, a discharge time from detection of the discharge start to detection of the discharge end for each of the plurality of discharges; an extraction step of extracting, by the controller, a minimum ejection time as a feature point from the ejection times of the plurality of ejections; a setting step of setting the on-off valve by the controller using a valve opening speed corresponding to the minimum value of the discharge time; A method for adjusting a nozzle start state, comprising:

9. a discharge step of discharging the solvent from the nozzle a plurality of times while changing a valve opening speed of an opening / closing valve which starts and stops discharging the solvent from the nozzle; a step of taking a video of a state in which ejection is performed multiple times from the nozzle with a camera; a start detection step of detecting a start of ejection for each of the plurality of ejections by a controller based on the moving image; a calculation step of calculating, by the controller, a discharge delay time from when the controller issues a discharge signal to when the controller detects the start of discharge, for each of the plurality of discharges; an extraction step of extracting a maximum ejection delay time as a feature point from the ejection delay times of the plurality of ejections by the controller; a setting step of setting the on-off valve by the controller using a valve opening speed corresponding to the maximum value of the discharge delay time; A method for adjusting a nozzle start state, comprising:

Images