JP2009218622A - Substrate processing apparatus, and substrate position deviation correction method in substrate processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a practical configuration which allows a control for arranging a substrate in a set arrangement position in an apparatus to be performed with sufficiently high precision while using a substrate conveyance system. <P>SOLUTION: When a substrate 9 is held by an end effector 23 and is conveyed from a conveyance chamber 3 to a processing chamber ID through a gate valve chamber 5, images of two marks 25 on the end effector 23 and the substrate are imaged by a detector 62 consisting of an image sensor, without stopping the end effector, and an image signal is sent to a signal processing part 8. The image signal held by a holding circuit 81 is processed by an image processing circuit 82 to calculate positions of an effector reference point and a substrate reference point, and the positions are compared with data of correct positions to obtain position deviations of the end effector 23 and the substrate 9. A control part 20 is operated so as to correct the position deviations, whereby the substrate 9 is placed in a set arrangement position with high precision. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a substrate processing apparatus that performs a predetermined process on the surface of a substrate, and particularly to a configuration for accurately transporting a substrate to a set position in the apparatus in such an apparatus. is there.

Semiconductor devices such as various memories and logic are manufactured through many surface treatments on a substrate. As a substrate processing apparatus that performs such surface treatment, a sputtering apparatus, a chemical vapor deposition (CVD) apparatus, an etching apparatus, and the like are known.
The substrate processing apparatus generally includes a processing chamber that performs predetermined processing on the substrate therein, a substrate transfer system that carries an unprocessed substrate into the processing chamber and unloads the processed substrate from the processing chamber, and a processing A load lock chamber or the like is provided between the chamber and the atmosphere side. A cluster tool type apparatus employs a layout in which a transfer chamber is provided in the center, and a processing chamber and a load lock chamber are provided for the transfer chamber. The substrate transfer system includes a transfer robot provided in the transfer chamber. The transfer robot includes an end effector that holds the substrate at the tip of the arm. A cassette that can store a predetermined number of unprocessed or processed substrates is provided in the load lock chamber.

JP-A-5-287527 JP-A 64-57104 JP-A-11-131231 Japanese Patent Laid-Open No. 9-270451 JP-A-10-526

In the substrate processing apparatus as described above, it is generally necessary to position the substrate at a predetermined position in a processing chamber in which processing is performed. The reason for this is that the processing method is often different depending on the position in the processing chamber, and in order to perform highly reproducible processing, it is necessary to always place the substrate at the same predetermined position in the processing chamber. It is.
Further, when returning from the processing chamber to the load lock chamber, it is necessary to correctly return the substrate to a predetermined position of the cassette in the load lock chamber. For example, if the substrate is shifted back to the cassette, the substrate may collide with the wall surface of the cassette and the substrate may be damaged.

There is also a need for the substrate and end effector to move through the correct transport path. This is because if the substrate does not move through the correct conveyance path, the substrate or the end effector may be damaged by colliding with a structure in the apparatus, or the substrate may be displaced due to the collision. For example, the end effector moves between the processing chamber and the transfer chamber through the opening of the gate valve chamber, but in order to prevent residual gas from the processing chamber from diffusing into the transfer chamber as much as possible, Is very small. Therefore, if the end effector is displaced from the correct conveyance path, the substrate or the end effector is likely to be displaced or damaged by hitting the edge of the opening.
In addition, the volume of the processing chamber tends to be reduced in order to improve the exhaust efficiency, and the substrate is often transported while maintaining a state that is quite close to the inner surface of the processing chamber. For this reason, even if the deviation is about several mm, for example, the substrate may collide with the inner surface of the processing chamber, and the substrate may be further displaced or the substrate may be damaged.

  The placement of the substrate at a position (hereinafter referred to as a set placement position) that is finally set as the position where the substrate is to be placed is basically performed by the control of the transfer robot. The transfer robot is a mechanism that is controlled numerically by a computer using a lot of servo motors. More specifically, the set arrangement position input to the computer is determined based on some reference point in the apparatus. Usually, a point on the axis of the transfer robot is determined as a reference point (hereinafter, this point is referred to as a robot origin).

  The substrate is transferred by designating the amount and direction of movement with respect to the robot origin. For example, in the case of transferring from a transfer chamber to a processing chamber, a specific position in the transfer chamber is designated as a substrate position when starting transfer of the substrate (hereinafter referred to as a transfer start position), A specific position is designated as the set arrangement position. Then, the positional relationship between the transfer start position and the set arrangement position is calculated with reference to the robot origin, and the amount required to move from the transfer start position to the set arrangement position is commanded as X, Y, and Z data. . The transport robot is driven based on this command.

  The same applies to the case where the substrate is transferred from the load lock chamber to the transfer chamber. A specific position in the cassette in the load lock chamber is designated with reference to the robot origin as a transfer start position, and the specific position of the transfer chamber is similarly set as a set arrangement position with reference to the robot origin. The positional relationship between the transfer start position and the set arrangement position is obtained, and X, Y, and Z drive command signals are generated, thereby driving the transfer robot.

  As described above, by inputting the data of the set arrangement position of the substrate in the processing chamber to the computer, it is possible to arrange the substrate with very high positional accuracy. It is also possible to return the substrate to a predetermined position in the load lock chamber or move the end effector along the correct transfer path by inputting appropriate data to the computer that controls the transfer robot. Is possible.

  However, such highly accurate substrate transport is premised on the fact that the substrate is originally correctly positioned at the transport start position. If the substrate is not correctly positioned at the transfer start position, it is impossible to correctly position the substrate at the set position even with a high-precision servo mechanism of the transfer robot. For example, this can happen if the substrate is not housed in the correct position in the cassette within the load lock chamber and the substrate is misaligned within the cassette. In this case, the substrate is displaced from the set arrangement position in the transfer chamber, and is displaced from the set arrangement position in the processing chamber.

  In addition, although the transfer robot has high accuracy in terms of electronic control, a decrease in accuracy over time is inevitable in the mechanical part. For example, a buffer portion such as a bearing used for the mechanism portion wears with time, and even if the data is correct, the substrate may actually be placed at a slightly shifted position.

Furthermore, the reason why the substrate is not correctly placed at the set placement position is that the substrate is displaced on the end effector of the transfer robot. The end effector is to hold the substrate on the upper surface, but recently, many end effectors are configured to be held in point contact with the back surface of the substrate at several points. The reason for this is to reduce the contact area with the back surface of the substrate and suppress the generation of particles (a general term for fine particles that contaminate the substrate) due to scratching the back surface of the substrate. Thus, if the contact area with respect to the back surface of a board | substrate reduces, a board | substrate will slip on an end effector and possibility that a position will shift | deviate becomes high.
In addition, when the processing in the processing chamber is a film formation process, and a predetermined thin film is formed on the surface of the substrate, the substrate after film formation may be slightly distorted. This is because the thin film is deposited with an internal stress, and the distorted substrate is likely to slip on the end effector and cause displacement. If the substrate is displaced on the end effector in this way, even if the end effector is correctly moved by numerical control, the substrate may not be placed in the correct position in the processing chamber or returned to the correct collection position of the cassette. There is a problem to do.

  The invention of the present application has been made to solve the above-described problems, and has the significance of providing a practical configuration in which a substrate is placed at a set placement position in the apparatus while using a substrate transport system. It is meaningful to provide a configuration that can perform control with sufficiently high accuracy.

In order to solve the above-mentioned problems, the invention according to claim 1 of the present application is directed to a processing chamber for performing predetermined processing on a substrate inside, a non-processed substrate being carried into the processing chamber, and a processed substrate being processed A substrate processing apparatus comprising a substrate transport system for unloading from a chamber,
The substrate transport system has an end effector that holds the substrate and a drive mechanism that drives the end effector, and the end effector is provided with at least two marks.
An image sensor that simultaneously detects these marks and the substrate;
And a signal processing unit that processes a signal from the image sensor to obtain a positional deviation from a correct position of the substrate.
In order to solve the above-mentioned problem, in the invention according to claim 2 of the present application, in the configuration of claim 1, the signal processing unit obtains data of the positions of the marks from the detected marks and detects the detected substrate. Thus, the position data of the edge of the substrate is obtained, and the positional deviation with respect to the correct position of the substrate is obtained according to these data.
In order to solve the above-mentioned problem, in the invention according to claim 3 of the present application, in the configuration of claim 1 or 2, the positional shift is a shift of a reference point of the substrate with respect to a reference point on the end effector. The arithmetic processing unit is configured to calculate a reference point of the substrate from the detected mark and the substrate, and obtain the position shift according to the calculation result.
In order to solve the above-mentioned problem, the invention according to claim 4 of the present application is that, in the configuration of claim 3, the reference point of the substrate is the center of the substrate, and the arithmetic processing unit is configured to start from the position of the mark. Specify two points on the edge of the board, calculate the position of the center of the board according to the data of the position of the two points on the edge and the known value of the radius of the board, and use it as the position of the reference point It has the structure of being.
In order to solve the above-described problem, in the invention according to claim 5 of the present application, in the configuration of claim 1, 2, 3 or 4, the correct position is a position of the substrate at the start of conveyance of the substrate. When the end effector moves so that the substrate is transported from a certain transport start position to a set placement position that is finally set as a position where the substrate is to be placed, this is the position that the substrate takes during the transport. It has the structure of.
In order to solve the above-described problem, the invention according to claim 6 of the present application is that, in the configuration of claim 5, the signal processing unit includes a reference memory in which the data at the correct position is stored, and an actual transport. A calculation circuit that compares the detected position data with the correct position data to obtain a position shift, and the reference memory includes the substrate or the substrate jig in the set arrangement position. The image signal obtained by imaging with the image sensor can be stored and updated in the middle of returning the substrate or the substrate jig to the transfer start position after being actually correctly arranged.
In order to solve the above-described problem, the invention according to claim 7 of the present application is the configuration according to claim 5 or 6, wherein a plurality of the transfer start positions and the set arrangement positions are set and set from the transfer start positions. A plurality of the image sensors are provided so that a displacement from the correct position is required in the middle of each conveyance to the arrangement position.
In order to solve the above-mentioned problem, the invention according to claim 8 of the present application is the configuration according to claim 1, 2, 3, 4, 5, 6 or 7, wherein the drive mechanism of the substrate transport system has the required position. The end effector is driven so as to correct the deviation.
In order to solve the above-mentioned problem, in the substrate processing apparatus of the present invention, the substrate is set from the transfer start position by holding the substrate on the end effector and moving the end effector by driving the drive mechanism. A transporting process for transporting to the arrangement position;
An imaging step of imaging the substrate by an image sensor in the middle of conveying the substrate from the conveyance start position to a set arrangement position;
A signal processing step of processing an image signal from an image sensor to obtain a positional shift with respect to a correct position of the substrate,
The signal processing step processes the image signal from the image sensor to obtain the position of the peripheral edge of the substrate, calculates the position of the reference point of the substrate based on this, and then corrects the substrate based on the calculation result. A step of obtaining a positional deviation with respect to the position,
The movement of the end effector in the transporting process includes a movement of the end effector for correcting the positional deviation obtained in the signal processing process.

As described above, according to the inventions of claims 1 to 8 of the present application, the end effector is provided with two marks, and these marks are detected by the detector together with the substrate. It is possible to detect the position deviation of the substrate easily. Further, since the position shift is detected by the image sensor, the detection is easy and does not become large.
Further, according to the invention of claim 5, in addition to the above effect, since the displacement of the substrate is detected while the substrate is being transported to the set placement position, the displacement is corrected and the substrate is correctly placed at the set placement position. Can be shortened.
According to the sixth aspect of the invention, in addition to the above effect, the positional deviation is detected as compared with the data obtained by actually correctly arranging at the set arrangement position, so that the detection accuracy becomes high.
Further, according to the invention of claim 7, in addition to the above effect, the positional deviation can be corrected at each point of the conveyance path, so that the end effector and the substrate are less displaced from the correct conveyance path. Therefore, the possibility that the end effector and the substrate collide with the structure is significantly reduced.
According to the eighth aspect of the invention, in addition to the above effect, the positional deviation is corrected by driving the end effector, so that the configuration and operation for correction are simplified.
According to the ninth aspect of the invention, since the position shift is detected by the image sensor, the detection is easy and does not become large. In addition, since the position shift is detected while the substrate is being transported to the set placement position, the time required to correct the shift and correctly place the substrate at the set placement position can be shortened. Furthermore, since the position shift is corrected by driving the end effector, the configuration and operation for correction are simplified.

It is the plane schematic which showed embodiment of the substrate processing apparatus of this invention. It is the cross-sectional schematic in XX of FIG. FIG. 3 is a schematic plan view of an end effector 23 used in the apparatus shown in FIGS. 1 and 2. It is front sectional drawing which shows schematic structure of the mark detection system 6 which detects the mark 25 shown in FIG. It is a front section schematic diagram explaining the composition of the perpendicular direction detection system 7 with which the device of this embodiment was provided. It is the plane schematic explaining the positional relationship of the mark detection system 6 and the vertical direction detection system 7. It is a figure explaining the judgment which the judgment circuit 812 with which the hold circuit 81 was provided, and is a figure explaining the hold position set in the space in the gate valve chamber 5. FIG. It is a figure explaining the judgment which the judgment circuit 812 with which the hold circuit 81 is provided, and is the figure which showed the reference | standard pixel row defined on CCD621 regarding the hold position shown in FIG. It is a figure explaining the judgment which the judgment circuit 812 with which the hold circuit 81 was provided, and is a figure explaining the conditions of a hold. It is a figure explaining the image processing which the image processing circuit 82 performs. It is a figure explaining the image processing which the image processing circuit 82 performs. It is a figure explaining arrangement | positioning to the setting arrangement position by manual operation. It is a figure explaining conveyance of the board | substrate 9 which correct | amended deviation.

Embodiments of the present invention will be described below. FIG. 1 is a schematic plan view showing an embodiment of a substrate processing apparatus of the present invention, and FIG. 2 is a schematic cross-sectional view taken along line XX in FIG.
The substrate processing apparatus shown in FIGS. 1 and 2 is a cluster tool type apparatus, and includes a transfer chamber 3 disposed in the center and a plurality of processing chambers 1A, 1B, 1C, 1D provided around the transfer chamber 3. , 1E, 1F (hereinafter collectively referred to as processing chamber 1 as required) and two load lock chambers 4. Each of the chambers 1A, 1B, 1C, 1D, 1E, 1F, 3 and 4 is a vacuum container that is exhausted by a dedicated or dual-purpose exhaust system 100. The chambers 1A, 1B, 1C, 1D, 1E, 1F, and 4 are connected to the central transfer chamber 3 in an airtight manner, and a gate valve chamber in which a gate valve 51 is provided at the connection point. 5 is interposed.

  The transfer chamber 3 separates the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F from each other in an airtight manner to prevent cross-contamination of the internal atmosphere, and each processing chamber 1A, 1B, 1C, 1D, 1E, This is a space through which the substrate is transferred to the first floor and the load lock chamber 4. In the transfer chamber 3, a transfer robot 21 constituting a substrate transfer system is provided. The transfer robot 21 takes out the substrates 9 one by one from one load lock chamber 4 and sends them to the respective processing chambers 1A, 1B, 1C, 1D, 1E, 1F to sequentially process them. And after finishing the last process, it returns to the one or the other load lock chamber 4.

  Also in this embodiment, the transfer robot 21 includes an end effector 23 provided at the tip of the arm 22 so as to hold the substrate 9 and a drive mechanism 24 that drives the end effector 23. The drive mechanism 24 rotates relative to the end effector 23 around a vertical axis passing through the center of the transfer chamber 3, an arbitrary linear linear motion of a circle around the axis, and a vertical linear motion. And is configured to be able to give. In addition, a control unit 20 that controls the entire apparatus including the substrate transport system is provided.

  The major feature of the present embodiment is that the positions of the end effector 23 and the substrate 9 during the transfer so that the substrate 9 is correctly arranged at the set arrangement position when the substrate 9 is transferred by the substrate transfer system described above. Is detected. More specifically, the positions of the end effector 23 and the substrate 9 are correct when passing through the gate valve chamber 5 provided between the transfer chamber 3 and the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F. Based on this result, the substrate transport system is controlled so that the substrate 9 is correctly placed at the set placement position.

  Another major feature of the present embodiment is that the end effector 23 is provided with two marks 25. By detecting these marks 25, the positions of the end effector 23 and the substrate 9 are detected. It is that you are. Hereinafter, the configuration of the mark detection system 6 that detects the mark 25 of the end effector 23 will be described with reference to FIGS. 3 and 4. FIG. 3 is a schematic plan view of the end effector 23 used in the apparatus shown in FIGS. 1 and 2. FIG. 4 is a front sectional view showing a schematic configuration of the mark detection system 6 for detecting the mark 25 shown in FIG.

As shown in FIGS. 1 and 2, the end effector 23 is a plate-like member fixed to the tip of the arm 22 of the transport robot 21. As shown in FIG. 3, the end effector 23 has a shape in which a U-shaped cutout is provided in a substantially rectangular plate material. The depth direction of the U-shaped notch of the end effector 23 coincides with the direction of the long side of the rectangle. When the end effector 23 performs the linear motion in the radial direction, the end effector 23 moves while maintaining a posture in which the depth direction of the U-shaped notch coincides with the direction of the linear motion. Hereinafter, a virtual line extending in the depth direction of the notch on the surface of the end effector 23 is referred to as an “effector reference line”, and is indicated by _LES in FIG.

As shown in FIG. 3, two marks 25 formed in a black circular pattern are provided on the surface of the end effector 23. The mark 25 can be provided by any method such as painting, vapor deposition, or spraying. The two marks 25, the line segment connecting their centers is provided so as to be bisected by the effector reference line L _ES with orthogonal to effector reference line L _ES. As shown in FIGS. 1 and 3, the substrate 9 is placed and held so as to substantially cover the U-shaped notch of the end effector 23. Protrusions (not shown) are provided at several places around the U-shaped notch, and the back surface of the substrate 9 is in contact with these protrusions. The position where the mark 25 is provided is a position sufficiently away from the region where the substrate 9 is placed, so that the mark 25 is not covered by the substrate 9.

  As shown in FIG. 2, the gate valve chamber 5 is a vacuum vessel hermetically connected to the adjacent chambers 1A, 1B, 1C, 1D, 1E, 1F, 3 and 4, and serves as a dual-purpose or dedicated exhaust system. 100 is exhausted. The gate valve 51 is driven in the gate valve chamber 5 to open and close the opening 54 provided on the processing chambers 1A, 1B, 1C, 1D, 1E, 1F or the load lock chamber 4 side.

Specifically, the gate valve 51 is fixed to the tip of a drive rod 53 connected to a drive source 52. When the drive source 52 operates, the drive rod 53 is driven, the gate valve 51 moves up and down, and opens and closes the opening 54 provided on the processing chamber 1A, 1B, 1C, 1D, 1E, 1F or the load lock chamber 4 side. It is like that.
As shown in FIGS. 2 and 4, a detection opening 55 is provided in the upper wall portion of the gate valve chamber 5, and the mark detection window 56 is fitted so as to close the detection opening 55. It is rare. A vacuum seal such as an O-ring is provided between the mark detection window 56 and the upper wall portion.

  The mark detection system 6 takes an illumination light source 61 that illuminates the inside of the gate valve chamber 5 through the mark detection window 56 and an image in the gate valve chamber 5 that is illuminated by the illumination light source 61 through the mark detection window 56. The detector 62 is mainly configured. Above the mark detection window 56, a half mirror 63 is provided at an angle of 45 degrees. As the illumination light source 61, a small, high-intensity halogen lamp or the like is preferably used. The illumination light source 61 includes a cup-shaped reflector 611 and emits light forward. The emitted light is reflected by the half mirror 63 and reaches the gate valve chamber 5 through the mark detection window 56 to illuminate the gate valve chamber 5.

  A diffusing plate 64 is provided between the illumination light source 61 and the half mirror 63. The diffusing plate 64 diffuses the light from the illumination light source 61 so that it is not strongly illuminated by the pattern of the light emitting part such as the filament of the illumination light source 61. A total reflection mirror 65 is provided above the half mirror 63. The total reflection mirror 65 is inclined by 45 degrees on the side opposite to the half mirror 63. The detector 62 is an area image sensor provided with a CCD 621, such as STC-400 manufactured by Sensor Technology Corporation. The light receiving surface of the CCD 621 is provided to be perpendicular to the optical axis that is bent 90 degrees by the total reflection mirror 65 and extends horizontally. The detector 62 includes an optical system 622 that projects an image on the light receiving surface of the CCD 621.

  The image in the gate valve chamber 5 illuminated by the illumination light source 61 is captured by the CCD 621 through the half mirror 63 and the total reflection mirror 65. Therefore, as shown in FIG. 4, when the end effector 23 holding the substrate 9 reaches the gate valve chamber 5, images of the substrate 9 and the end effector 23 are taken by the CCD 621. Thereby, the mark 25 provided on the end effector 23 is detected.

  The mark detection system 6 described above detects the position of the end effector 23 and the substrate 9 in the horizontal plane by detecting the mark 25. Yet another major feature of the present embodiment is that a vertical direction detection system 7 that detects the position of the end effector 23 and the substrate 9 in the vertical direction is provided. This point will be described with reference to FIG. FIG. 5 is a schematic front sectional view illustrating the configuration of the vertical direction detection system 7 provided in the apparatus of the present embodiment.

  Similar to the mark detection system 6, the vertical direction detection system 7 detects when the end effector 23 reaches the gate valve chamber 5. A vertical detection window 57 is provided on the upper wall portion of the gate valve chamber 5 in addition to the mark detection window 56. The vertical detection window 57 is also an optical window that hermetically closes the opening provided in the upper wall portion. The vertical direction detection system 7 has the same configuration as the laser displacement meter. That is, the vertical direction detection system 7 is reflected by the laser light source 71 for making the laser light incident on the end effector 23 or the substrate 9 of the gate valve chamber 5 through the detection window 57 for the vertical direction, and reflected back to the end effector 23 or the substrate 9. The light receiving unit 72 mainly receives a laser beam and a vertical direction signal processing unit 73 that processes a signal from the light receiving unit 72 and detects a vertical position shift of the end effector 23.

  As the laser light source 71, a semiconductor laser, a He—Ne laser, or the like can be used. The optical axis of the laser light incident from the laser light source 71 into the gate valve chamber 5 is slightly tilted from the vertical direction as shown in FIG. The laser light reflected by the end effector 23 or the substrate 9 is emitted at the same angle as the incident angle and enters the light receiver 72. At this time, the incident position of the laser light on the light receiver 72 changes as the end effector 23 or the substrate 9 is displaced in the vertical direction, as shown in FIG. For the light receiver 72, a photodiode array or a CCD is used.

  And the incident position of the laser beam reflected on the end effector 23 or the substrate 9 when the substrate 9 is correctly conveyed with the height of the conveyance line is set as the reference position. The vertical direction signal processing unit 73 processes the signal from the light receiver 72 to determine which position the laser beam is incident on, thereby detecting the vertical position of the end effector 23 or the substrate 9. It has become. Signals are sent from the light receiver 72 to the vertical direction signal processing unit 73 continuously in time, but in this embodiment, the displacement is detected by a signal at the moment held by a hold circuit 62 described later. It has become. Since the height of the end effector 23 and the height of the substrate 9 differ only by the amount of the protrusion, the position of the end effector 23 is directly detected and the position of the substrate 9 is indirectly detected. May be.

FIG. 6 is a schematic plan view illustrating the positional relationship between the mark detection system 6 and the vertical direction detection system 7. As shown in FIG. 6, the correct path the end effector 23 to be taken during transport (hereinafter, the transport line) L _T, is set so as to pass through the gate valve chamber 5. Movement of the end effector 23 on the conveying line L _T through the gate valve chamber 5 is a linear motion in the radial direction as described above, the end effector 23 is conveying line L while matching the effector reference line L _ES the transport line L _T Move along _T.
As seen from FIG. 6, the mark detection window 56 just above the conveying line L _T are located. Vertical detection window 57 is located obliquely above the conveying line L _T. However, the distance between the vertical direction detection window 57 and the mark detection window 56 is sufficiently shorter than ½ of the width (length in the short side direction) of the end effector 23, and the position directly below the vertical direction detection window 57. Does not come off from the end effector 23.

Next, the configuration of the signal processing unit 8 that processes signals from the mark detection system 6 and the vertical direction detection system 7 described above will be described.
The first basic function of the signal processing unit 8 is to process signals from the mark detection system 6 and the vertical direction detection system 7 to obtain the position of the end effector 23 and the position of the substrate 9 in the gate valve chamber 5. That is. The position in the gate valve chamber 5 is a relative position based on a reference position in the gate valve chamber 5 (hereinafter referred to as a gate valve chamber reference position). Further, the position of the end effector 23 is actually the position of a certain reference point (hereinafter referred to as an effector reference point) of the end effector 23, and the position of the substrate 9 is a specific position of the substrate 9. This is the position of a reference point (hereinafter referred to as a substrate reference point).

  In the present embodiment, the position is detected while the end effector 23 is moved in the gate valve chamber 5 without stopping. Therefore, the position in the gate valve chamber 5 is not the stop position. The image signal sent from the CCD 621 of the detector 62 is an image or moving image of the moving end effector 23 and the substrate 9. The signal processing unit 8 always extracts an image from the moving image under the same conditions, and performs signal processing on the image to obtain the positions of the effector reference point and the substrate reference point. More specifically, as shown in FIG. 2, the signal processing unit 8 holds the image signal sent from the mark detection system 6 when a certain condition is satisfied, and the hold circuit 81 holds the image signal. And an image processing circuit 82 for processing the image signal.

  The hold circuit 81 includes a hold memory 811, a determination circuit 812, and the like. A signal from the CCD 621 is sent to the hold memory 811. The signal from the CCD 621 is a readout signal of the accumulated charge of each pixel of the CCD 621, and is an image signal sent by one frame every predetermined readout cycle. The hold memory 811 is configured to update and store this image signal for each read cycle.

  The determination circuit 812 reads out one frame of image signal stored in the hold memory 811 and makes a determination described below. The determination performed by the determination circuit 812 will be described with reference to FIGS. 7, 8, and 9 are diagrams illustrating the determination performed by the determination circuit 812 included in the hold circuit 81. Among these, FIG. 7 is a diagram for explaining a hold position set in the space in the gate valve chamber 5, and FIG. 8 shows a reference pixel row determined on the CCD 621 in relation to the hold position shown in FIG. FIG. 9 is a diagram for explaining the hold condition.

The determination circuit 812 determines whether or not the mark 25 of the end effector 23 has reached the hold position. Hold position is set to a predetermined position on the transport line L _T of the gate valve chamber 5. Specifically, as shown in FIG. 7, position hold any of the marks 25 are in contact with the conveying line L _T horizontally perpendicular to the hold-line L _H to the end effector 23 at a predetermined point on the conveying line L _T Position.

Further, a position related to the hold position is selected on the light receiving surface of the CCD 621. Specifically, among the pixels constituting the light receiving surface of the CCD621, pixels located on the line where the hold line L _H is just projected, (shown in FIG. 8 P-P) the reference pixel string It is selected as. Each pixel of the CCD 621 is arranged along two directions (hereinafter, X direction and Y direction) perpendicular to each other in a plane. For example, the reference pixel column PP is selected for pixel columns arranged in the X direction. Each pixel column, as shown in FIG. _{8, P 11, P 12,} ... P 1n, P 21, P 22, ... P 2n, ...... P m1, P m2, ... When P _mn, P _s1, P _s2 ,... P _sm is selected as the reference pixel column _PP .

Further, CCD621, along with the reference pixel columns P-P is located at the position of the image point of the hold line _{L H,} the intersection of the conveying line _{L T} and the hold line _{L H,} the center of the reference pixel columns P-P Are attached to the gate valve chamber 5 with high precision so as to correspond to the positions of the pixels (hereinafter referred to as reference pixels). Note that the intersection of the conveying line L _T and the hold line L _H is a point to be a reference for identifying the position of the gate valve chamber 5, in the gate valve chamber reference position (Fig. 7 described above, S _G Is shown).

The determination operation of the determination circuit 812 will be described in more detail with reference to FIG. (1) to 9 (3), an image signal for each frame that is sent to the hold memory 811 are respectively, how the end effector 23 holding the substrate 9 is moved along the transport line L _T Is shown. When the end effector 23 moves, the image signal changes as shown in FIGS. Then, as shown in FIG. 9 (3), an image signal when any one mark 25 of the end effector 23 is captured by any pixel on the reference pixel column PP is held. Yes.

  More specifically, for example, when the mark 25 is black and the surface of the end effector 23 is white or the like, the accumulated charge of the pixel temporarily decreases when the mark 25 is captured. Accordingly, when the magnitude of the readout signal for any of the pixels constituting the reference pixel column PP falls below a certain level, the determination circuit 812 determines the image signal to be held as one frame at that time. to decide. When the image signal is held, the data in the hold memory 811 is not updated, and the held image signal is sent to the image processing circuit 82 as it is.

  Next, the configuration of the image processing circuit 82 will be described. The image processing circuit 82 obtains the position of the effector reference point and the substrate reference point from the image signal held by the hold circuit 81 at the moment of holding. Image processing performed by the image processing circuit 82 will be described with reference to FIGS. 10 and 11 are diagrams for describing image processing performed by the image processing circuit 82. FIG.

First, a configuration for obtaining the effector reference point will be described. In the present embodiment, the effector reference point is set to the midpoint of a line connecting the centers of the two marks 25. Therefore, the image processing circuit 82 first performs an operation for _obtaining the center of each mark 25 (indicated by M _c1 and M _c2 in FIG. 10). For example, each pixel group capturing the mark 25 is specified in the held image signal. Then, in each of these pixel groups, for example, a pixel located exactly in the middle of a pixel located on the most positive side and a pixel located on the negative side in the Y direction is specified. Similarly, in the X direction, a pixel located exactly in the middle between the most positive pixel and the most negative pixel is specified. These two pixels match in principle, but if they do not match, the pixel located at the midpoint of the line segment connecting the two is specified. The pixel specified in this way is a pixel that captures the centers M _c1 and M _c2 of the mark 25.

In this way, a pixel group capturing each mark 25 is specified, and two pixels capturing the centers M _c1 and M _c2 of each mark 25 are specified from the pixel group. The pixel located at the midpoint of the line segment connecting the centers M _c1 and M _c2 of the two marks 25 is the pixel capturing the effector reference point. The position of the pixel that captures this effector reference point is the origin of coordinates when obtaining the substrate reference point described below, and this pixel is hereinafter referred to as the image origin (I _O in FIGS. 10 and 11). Show).

In the present embodiment, the center of the substrate 9 (indicated by C in FIGS. 10 and 11) is set as the substrate reference point. In order to obtain the center C of the substrate 9 as the substrate reference point, first, as shown in FIGS. 10 and 11, the distances from the centers M _c1 and M _c2 of the mark 25 to the pixel group capturing the substrate 9 are obtained. Specifically, the first pixel that captures the substrate 9 is identified by tracing in the X direction from the centers M _c1 and M _c2 of the mark 25. The end effector 23 is white, and the substrate 9 is a semiconductor wafer, for example, and has a color (for example, silver) that has a sufficient contrast with white. The pixels specified in this way are set as edge points E ₁ and E ₂ . Then, the distances between the centers M _c1 and M _{c2 of} the mark 25 and the edge points E ₁ and E ₂ are obtained. The obtained distances are defined as D ₁ and D ₂ . The distance here is directly the number of pixels, but can also be converted into an actual distance from the magnification of the optical system 622 or the like (hereinafter the same).

Here, as shown in FIG. 10, when D ₁ = D ₂ , the center C of the substrate 9 exists on the X axis centered on the image origin I _O. Therefore, if the distance from the image origin I _O to the first pixel that captures the substrate 9 is D ₃ and the radius of the substrate 9 (converted by the number of pixels) is r, the positive side on the X axis The pixel located at the coordinates of (D ₃ + r) is capturing the center C of the substrate 9. In this case, the image processing circuit 82 calculates the obtained pixel coordinates (D ₃ + r, 0) as data of the substrate reference point.

Next, as shown in FIG. 11, when D ₁ ≠ D ₂ , the center C of the substrate 9 cannot be obtained by the simple calculation as described above, and the following is performed.
First, in FIG. 11, assuming that the middle point of a line segment connecting two edge points E ₁ and E ₂ is E _M , E _M passes through I _O and is parallel to M _C1 E ₁ and M _C2 E _2. On the (X-axis). When lines are drawn from the center C of the substrate 9 to E ₁ , E _M , and E ₂ , the line segment CE _M becomes a line segment that bisects ∠ E ₁ CE ₃ .
And if the length of I _O M _C1 (= length of I _O M _C2 ) is m, the length of the line segment E ₁ E ₂ is from Pythagorean theorem,
E ₁ E ₂ = √ {4 m ² + (D ₁ −D ₂ ) ² } Equation (1)
Similarly, the length of CE _M is
^{_{CE M = √ {r 2 -}} (E 1 E 2) 2/4}
= √ [r ² − {4 m ² + (D ₁ −D ₂ ) ² } / 4] Equation (2)
It becomes.

Here, let A be the intersection of perpendicular lines drawn from E _{2 with} respect to the line segment connecting M _C1 and E _1, and let B be the intersection of perpendicular lines drawn from the center C to the X axis. From the similarity of △ E ₁ AE ₃ and △ CBE _M ,
_{E M B = CE M × (} E 1 E 2 / 2m) ... Equation (3)
BC = CE _M × {E ₁ E ₂ / (D ₁ −D ₂ )} Expression (4)
It becomes.
Therefore, if the coordinates of the center C of the substrate 9 are (X _C , Y _C ),
X _C = (D ₁ −D ₂ ) / 2 + E _MB
Y _C = BC
These are represented by m, D ₁ , D ₂ , r by substituting the equations (1) to (4). Since m and r are constants, the coordinates of the center C of the substrate 9 can be obtained by calculating D ₁ and D ₂ as described above.

In this way, the image processing circuit 82 processes the image signal from the hold circuit 81 to obtain a pixel capturing the effector reference point and a pixel capturing the substrate reference point. The image processing circuit 82 includes a microcomputer including a RAM, a ROM, an input / output circuit, and the like, and executes a program for performing the above-described calculation.
In the above-described example, the center C of the substrate is obtained as the substrate reference point, but there may be other cases. For example, an orientation flat or notch (small notch) formed on the periphery of the substrate may be used as the substrate reference point.

  The second basic function of the signal processing unit 8 is to determine how much the positions of the end effector 23 and the substrate 9 in the gate valve chamber 5 obtained as described above are deviated from the correct positions. . That is, the signal processing unit 8 obtains the positional displacement of the end effector 23 and the positional displacement of the substrate 9 by comparing the reference memory 83 storing the data of the correct position in the gate valve chamber 5 and the data of the reference memory 83. And an arithmetic circuit 84.

  First, the output of the image processing unit 82 is selected by the switch circuit 85 to be input to the reference memory 83 or to one input of the arithmetic circuit 84. The switch circuit 85 inputs the output of the image processing circuit 82 to the reference memory 83 at the time of teaching, which will be described later, and inputs it to the arithmetic circuit 84 at the time of actually transporting the substrate 9.

  The reference memory 83 is a rewritable memory such as DRAM or SRAM. As described above, the reference memory 83 is input with data of correct positions of the effector reference point and the substrate reference point in advance. Data at the correct position is created by an operation called teaching and is stored in the reference memory 83. Hereinafter, this teaching will be described.

  For teaching, in order for the end effector 23 and the substrate 9 to be correctly arranged from the transfer start position to the set arrangement position, any position of the effector reference point or the substrate reference point when passing through the gate valve chamber 5 should be taken. This is the work of determining such data (hereinafter, teaching data). In this work, after the end effector 23 and the substrate 9 are correctly arranged manually at the set arrangement position, the transfer robot 21 is operated to return to the transfer start position by the reverse operation of transferring to the set arrangement position. To do. At that time, the positions of the end effector 23 and the substrate 9 are detected by the mark detection system 6 and the vertical direction detection system 7, and the teaching data is obtained based on the image signal held under essentially the same conditions. .

  First, the arrangement at the set arrangement position by manual operation will be described. FIG. 12 is a diagram illustrating the placement at the set placement position by manual operation. In the following description, a case where the set arrangement position is in the processing chamber 1D will be described as an example. The set arrangement position is a position above the substrate holder 11 in the processing chamber 1D. More precisely, the position is on a vertical line passing through the center of the substrate holding surface of the substrate holder 11 and a predetermined distance above the substrate holding surface.

As described above, the set arrangement position is set as coordinates based on the robot origin of the transfer robot 21 (indicated by O in FIG. 2). The robot origin O is a point on the central axis of the transfer robot 21 as shown in FIG. The set arrangement position is set with the coordinates of (X, Y, Z) with reference to the operation origin.
Teaching is performed in the following procedure. First, a jig (hereinafter referred to as a holder jig) 91 as shown in FIG. The holder jig 91 is a trapezoidal member having a height obtained by subtracting the thickness of the end effector 23 from the substrate holding surface of the substrate holder 11 to the set arrangement position. In addition, a mark 111 is formed so that the center axis of the holder jig 91 and the center axis of the substrate holder 11 can be arranged to coincide with each other. The mark 111 is used to attach the holder jig 91 to the substrate holder. 11 is arranged coaxially.

  Next, the end effector 23 is manually arranged on the holder jig 91 so that a predetermined point on the upper surface of the end effector 23 coincides with the set arrangement position. Specifically, for example, the end effector 23 is arranged so that the center point of the arc portion of the U-shaped notch in the upper surface of the end effector 23 coincides with the set arrangement position. A mark 911 is provided on the upper surface of the holder jig 91 so that the end effector 23 can be disposed at such a position.

  On the end effector 23, a jig (hereinafter referred to as a substrate jig) 92 made of a plate material having the same size and shape as the substrate 9 is disposed. At this time, the center of the substrate jig 92 (more precisely, the center of the back surface of the substrate jig 92) is positioned at the set position, that is, the central axis of the substrate jig 92 and the central axis of the substrate holder 11 Make sure it fits. For this positioning, for example, a laser can be used. That is, a small hole is formed in the center of the substrate jig 92, and a hole is also formed in the holder jig 91 through the central axis. A laser is incident along the central axis of the substrate holder 11 from above the substrate holder 11. If the center of the substrate jig 92 is aligned with the center axis of the substrate holder 11, the laser passes through the hole of the substrate jig 92 and the hole of the holder jig 91 and is reflected back to the surface of the substrate holder 11.

  In this manner, the holder jig 91, the end effector 23, and the substrate jig 92 are manually operated so that a predetermined point on the upper surface of the end effector 23 and the center of the back surface of the board jig 92 are positioned at the set position. Place at. Next, an operation signal is sent to the transfer robot 21 to move the end effector 23 to the transfer start position set in the transfer chamber 3. This movement is completely opposite to the movement when the actual substrate 9 is transported.

  When the end effector 23 holding the substrate jig 92 is moved in the reverse direction from the set arrangement position to the transfer start position, the mark detection system 6 and the vertical direction detection system 7 are operated, and the inside of the gate valve chamber 5 is operated. The positions of the mark 25 of the end effector 23 and the substrate jig 92 are detected. That is, the image is held under essentially the same conditions as described above, and the position of the effector reference point and the position of the substrate reference point are specified from the held image. Since the movement direction of the end effector 23 is opposite to that described above, the hold timing is the timing at which the mark 25 passes the reference pixel row PP. That is, the accumulated charge temporarily increased by catching the mark 25 is held at the timing when it is reduced to the original amount.

As described above, the position of the effector reference point and the position of the substrate reference point are directly specified by the pixel position of the light receiving surface of the CCD 621. However, the CCD 621 is mounted with high precision so that the reference pixel and the gate valve room reference position S _G correspond to each other (so that the position of the reference pixel becomes the image point of the gate valve room reference position S _G ). The positions of the effector reference point and the substrate reference point are specified based on the gate valve indoor reference position. That is, in the hold timing, effector reference point and the substrate reference point is specified that it is in any position with respect to the gate valve chamber reference position S _G.

The vertical position of the effector reference point is detected by the vertical direction detection system 7 as described above. Although the vertical direction detection system 7 does not detect the vertical position of the effector reference point itself, the end effector 23 moves while maintaining a horizontal posture with high accuracy, and there is no inclination of the upper surface of the end effector 23. Good as a thing. Therefore, the position in the vertical direction at a point other than the effector reference point is set as the vertical position of the effector reference point. The vertical position of the substrate reference point may be a value obtained by adding the height of the protrusion of the end effector 23 to the vertical position of the effector reference point.
Data on the positions of the effector reference point and the substrate reference point specified in this way are sent to and stored in the reference memory 83. This completes a series of teaching operations.

  Next, detection of the positional displacement of the end effector 23 and the positional displacement of the substrate 9 in the actual conveyance of the substrate 9 and the conveyance of the substrate 9 in which the displacement is corrected will be described. The following description is also a description of an embodiment of the invention of the substrate displacement correction method. In the following description, the case where the substrate 9 is transferred from the transfer chamber 3 to the processing chamber 1D will be described as an example. FIG. 13 is a diagram illustrating the conveyance of the substrate 9 with the deviation corrected.

The transfer robot 21, as shown in FIG. 13, from the conveyance start position P _S which is set in the transfer chamber 3, to move the end effector 23 to convey the set position P _E in the processing chamber 1D. In the coordinate system based on the robot origin O, a drive command signal is sent to the transfer robot 21 according to the positional relationship between the transfer start position P _S and the set arrangement position P _E. That is, in the coordinate system in which the robot origin O is (0, 0, 0), the transfer start position P _S is (X ₀ , Y ₀ , Z ₀ ), and the set arrangement position P _E is (X ₁ , Y ₁ , If Z ₁ ), the control unit 20 sends a drive command signal to the transport robot 21 so that the substrate reference point moves from (X ₀ , Y ₀ , Z ₀ ) to (X ₁ , Y ₁ , Z ₁ ). That is, the drive command signal is a signal for moving the end effector 23 by (X ₁ −X ₀ , Y ₁ −Y ₀ , Z ₁ −Z ₀ ).

13, a chain line is the locus of the substrate reference point when the substrate 9 is conveyed correctly corresponds to the conveying line L _T. Moreover, two-dot chain line in FIG. 13 are not located in the original substrate 9 is conveyed start position P _S, which indicates a locus of the substrate reference point when not properly conveyed due.
The transfer robot 21 operates based on the drive command signal, and as described above, the end effector 23 passes through the gate valve chamber 5 and images of the end effector 23 and the substrate 9 are taken by the CCD 621. Then, an image is held at the timing described above, from the hold images, relatively more positional relationship between the gate valve chamber reference position S _G, the position of the effector reference point and the substrate reference points are determined.

Then, the actual position data at the time of carrying the substrate (hereinafter, actual data) obtained as described above is compared with the teaching data stored in the reference memory 83. That is, teaching and actual data are compared for the position of the substrate reference point, and the difference is obtained. In other words, the teaching data _{P T} of the substrate reference point _{(X TS, Y TS, Z} TS) and the actual data _{P R (X RS, Y RS} , Z RS) When, the arithmetic circuit 84, (X _TS - X _RS, Y _TS -Y _RS, obtains the Z _TS -Z _RS) becomes the correction vector V _C. If each component of the vector V _C is equal to or smaller than a predetermined magnitude, it is determined that the board reference point is in the correct position. If any of the components exceeds a predetermined size, a correction signal is issued assuming that the substrate 9 is not correctly placed at the set placement position. This correction signal is a vector having the same magnitude and the opposite sign as the correction vector V _C (X _TS −X _RS , Y _TS −Y _RS , Z _TS −Z _RS ) obtained above, that is, (X _RS -X _TS, Y _RS -Y _TS, a Z _RS -Z _TS).

The correction signal is sent to the control unit 20. The control unit 20 controls the transfer robot 21 according to the correction signal so that the substrate 9 is correctly positioned at the set arrangement position. That is, after the movement of the end effector 23 by the drive signal is completed, the transfer robot 21 further moves the end effector 23 by the amount of the correction signal. Thereby, the board | substrate 9 will be correctly located in a setting arrangement position.
It should be noted that the frequency misalignment in the vertical direction occurs less frequently and in comparison with the position misalignment in the horizontal direction. Therefore, the detection and correction of the vertical position deviation are not normally performed, and may be performed once in a predetermined operation time (for example, 24 hours) of the apparatus.

  Also, the teaching data for the effector reference point is compared with the actual data. And it is confirmed that those differences are below a predetermined value. When the difference between the teaching of the effector reference point and the actual data is large beyond a predetermined value, it indicates that the error due to the cause of the mechanical system such as wear is large. Therefore, in this case, a maintenance required signal indicating that maintenance work such as inspection or maintenance is necessary is issued. The maintenance required signal is displayed on a display unit (not shown) provided in the apparatus. Note that the difference between the teaching data and the actual data for the substrate reference point (that is, the positional deviation of the substrate 9) may be caused by the substrate 9 being displaced on the end effector 23, etc. As described above, the end effector 23 itself may be misaligned.

  In the above description, the positional deviation is detected and corrected when transporting from the transport chamber 3 to the processing chamber 1D. However, the positional deviation is detected and corrected when transporting from the load lock chamber 4 to the transport chamber 3. You can do the same. That is, a transfer start position is set in the load lock chamber 4, and a set arrangement position is set in the transfer chamber 3. Then, when the end effector 23 that has received the substrate 9 from the cassette 41 moves from the transfer start position to the set arrangement position, position detection and position shift detection are performed in the gate valve chamber 5 in the same manner. Then, a misalignment correction operation is performed as necessary.

  Further, the mark detection system 6 and the vertical direction detection system 7 are provided in all the gate valve chambers 5 between the processing chambers 1A, 1B, 1C, 1D, 1E, 1F and the transfer chamber 3 shown in FIG. Yes. Accordingly, the position of the end effector 23 and the position of the substrate 9 in the gate valve chamber 5 are detected as described above when transporting to the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F. It is possible to correct the positional deviation at.

  Further, depending on the case, the positional deviation may be detected and corrected when returning from the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F to the transfer chamber 3. Detection of misalignment and correction as necessary have already been performed when transported from the transport chamber 3 to the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F, but in the unlikely event, the processing chambers 1A, 1B, 1C, and 1D are performed. , 1E, 1F, detection and correction when returning to the transfer chamber 3 is effective when the substrate 9 is displaced. For example, when the substrate 9 is removed from the substrate holder 11 and the substrate 9 is displaced without being placed at a predetermined position on the end effector 23, detection and correction when returning to the transfer chamber 3 are effective. It is.

  Next, an example of the configuration of the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F will be described. The configuration of the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F varies depending on the contents of the processing. As an example, a configuration in the case of performing a process of creating a barrier film that prevents mutual fouling between two layers will be described. For convenience of explanation, the processing chambers 1A, 1B, 1C, 1D, 1E, and 1F are referred to as a first processing chamber 1A, a second processing chamber 1B,.

  When the barrier film is formed, the first processing chamber 1A is configured as a preheating chamber for preheating the substrate 9 prior to film formation, and the second processing chamber 1B is a natural oxide film or protective film on the surface of the substrate 9 prior to film formation. It is configured as an etching chamber for performing etching for removing the film. The third processing chamber 1C and the fourth processing chamber 1D are configured to form a barrier film by sputtering. As the barrier film, a laminated film of titanium and titanium nitride is often employed. In this case, a titanium thin film is formed in the third processing chamber 1C, and a titanium nitride thin film 1D is formed in the fourth processing chamber. The fifth and sixth processing chambers 1E and 1F are spare chambers, and are configured as cooling chambers, for example, when the substrate needs to be cooled.

  The configuration of the fourth processing chamber D for performing sputtering will be described in more detail with reference to FIG. As shown in FIG. 2, in the fourth processing chamber 1D, there is a substrate holder 11 for holding the substrate 9 at a predetermined position, a sputter electrode 13 provided facing the substrate holder 11, and a predetermined amount in the processing chamber 1D. And a gas introduction system 14 for introducing the above gas.

The sputter electrode 13 is composed of a target made of a material to be deposited, a magnet capable of magnetron sputtering, and the like. A sputtering power supply (not shown) that applies a negative high voltage or a high frequency voltage to the target is provided.
A predetermined gas is introduced into the processing chamber 1D by the gas introduction system 14, and the exhaust system 100 is controlled to adjust the pressure in the processing chamber 1D to a predetermined value. In this state, a sputtering power source (not shown) is operated to generate sputter discharge. The target is sputtered to form a predetermined thin film on the surface of the substrate 9 on the substrate holder 11. When a titanium nitride film is formed, sputtering is performed using a titanium target and introducing nitrogen gas.

  The substrate holder 11 includes a plate 12 that can be raised and lowered. The plate 12 receives and lowers the substrate 9 positioned at the set position, and places the substrate 9 on the substrate holding surface of the substrate holder 11. Then, after the processing is completed, the substrate 9 is raised to the set arrangement position. Then, the substrate 9 is unloaded from the position to the transfer chamber 3.

  As shown in FIG. 1, an external cassette 400 is arranged on the atmosphere side. An unloading substrate 9 accommodated in the external cassette 400 is transferred to the cassette 41 in the load lock chamber 4, and an autoloader 401 is provided to transfer the processed substrate 9 from the cassette 41 to the external cassette 400. Yes. The cassette 41 in the load lock chamber 4 is provided with an elevating mechanism 42.

Next, the operation of the entire apparatus of this embodiment will be described with reference to FIG. The unprocessed substrate 9 is transferred from the external cassette 400 to the load lock chamber 4 by the autoloader 401 and accommodated in the cassette 41 in the load lock chamber 4. A predetermined number of unprocessed substrates 9 are accommodated in the cassette 41 in the load lock chamber 4.
The transfer robot 21 takes out one substrate 9 from the cassette 41 and transfers it to the first processing chamber 1. When the processing in the first processing chamber 1A is completed, the transfer robot 21 transfers the substrate 9 to the second processing chamber 1B. Then, the next substrate 9 is taken out from the cassette 41 in the load lock chamber 4 and transferred to the first processing chamber 1A. In this manner, the substrate 9 is taken out from the cassette 41 and sequentially transferred to the first processing chamber 1A, the second processing chamber 1B,..., The sixth processing chamber 1F to perform processing. Then, the substrate 9 that has been processed in the sixth processing chamber 1 </ b> F is returned to the cassette 41 in the load lock chamber 4. Then, it is taken out by the autoloader 401 to the external cassette 400 on the atmosphere side.

  In the above operation, as described above, the substrate 9 is transported between the load lock chamber 4 and the transport chamber 3, and the processing chambers 1A, 1B, 1C, 1D, 1E, 1F and the transport chamber 3 are transported. When the substrate 9 is transported, the position of the end effector 23 and the position of the substrate 9 in the gate valve chamber 5 are detected, and the positional deviation is obtained by comparing with the teaching data which is the correct position data. A misalignment correction operation is performed. For this reason, the board | substrate 9 is always arrange | positioned correctly in each setting arrangement position in an apparatus. Therefore, the reproducibility of processing in each processing chamber 1 is maintained high, and an accident such as the substrate 9 colliding with the wall of the cassette 41 in the gate valve chamber 5 is prevented.

  In addition, since necessary misalignment correction is performed at each point on the transfer path from the load lock chamber 4 to the load lock chamber 4 again, the end effector 23 and the substrate 9 always pass through the correct transfer path. The end effector 23 and the substrate 9 move. For this reason, the possibility that the end effector 23 and the substrate 9 collide with the structural portion of the edge of the opening of the gate valve chamber 5 is significantly reduced.

  For example, when the end effector 23 moves from a certain transport start position to a certain set position, it is assumed that the substrate 9 is displaced for some reason. Even if an accident such as a collision does not occur particularly during the conveyance to the set arrangement position, the end effector 23 moves to the next position in a state where the positional deviation is preserved. Therefore, if the distance to the structure is short at the next position, an accident such as a collision may occur. Therefore, it is preferable to detect and correct misalignment at as many locations as possible on the transport path and eliminate the transport operation that preserves the misalignment from the viewpoint of eliminating the risk of collision with the structure. The apparatus according to the present embodiment is very suitable in this respect because it is possible to detect and correct misalignment of the end effector 23 and the substrate 9 during all conveyance through the gate valve chamber 5.

  In the apparatus of the present embodiment, as described above, the end effector 23 has the two marks 25, and the mark detection system 6 detects the two marks 25. This configuration has an important significance in relation to the specification of the substrate reference point. If there is only one mark 25, it is almost impossible to determine the position of the substrate reference point from the position data of the mark 25. However, if there are two marks 25, the substrate reference point can be easily obtained as described above. As a method for obtaining the center of the substrate 9 without detecting the position of the mark 25, there is a method for obtaining the center of the substrate 9 by processing an image of the outline of the substrate 9, but there is a disadvantage that it is very complicated and takes time. Note that the number of marks 25 is not necessarily two, and may be three or more. Increasing the number of marks 25 is preferable because the calculation accuracy of the substrate reference point is improved, although the calculation is somewhat complicated.

  In the above embodiment, the detector 62 detects the mark 25 of the end effector 23 and the substrate 9 at the same time. However, only the mark 23 may be detected. For example, this configuration is sufficient when only the movement accuracy of the end effector 23 is a problem. Further, the position shift may be detected only for the substrate 9. When the accuracy of the substrate transport system is high and the position of the end effector 23 is essentially not misaligned, the position and misalignment of only the substrate 9 may be detected and necessary correction may be performed.

  The configuration in which the detector 62 is an image sensor has a meaning that the position of the end effector 23 and the substrate 9 can be easily detected. As another configuration of the detector 62, a configuration using a plurality of sets of photosensors can be considered. In this configuration, a plurality of sets of photosensors are arranged at predetermined positions, and position detection is performed in an on / off state. However, such a configuration tends to be large, and it is difficult to detect with high accuracy.

The configuration in which the mark detection system 6 and the vertical direction detection system 7 are provided in the gate valve chamber 5 has the following significance.
In order to correct the positional deviation at the set arrangement position, it is desirable to detect the positional deviation of the end effector 23 or the substrate 9 at the place where it is set. In practice, however, it is often difficult. For example, it is actually difficult to provide the mark detection system 6 and the vertical direction detection system 7 described above in the processing chamber 1. A gas having a deposition action is often used in the processing chamber 1, and the reason is that the detection window provided on the wall portion of the processing chamber 1 is clouded with deposits.
Further, when the mark detection system 6 and the vertical direction detection system 7 are provided on the wall portion of the processing chamber 1 or the transfer chamber 3, such a wall portion can be opened and closed for maintenance in the chamber. In many cases, the positions of the mark detection system 6 and the vertical direction detection system 7 may shift due to opening and closing. As a result, there arises a problem that the detection accuracy is lowered.

  Since the wall portion constituting the gate valve chamber 5 is rarely openable and closable, the configuration in which the mark detection system 6 and the vertical direction detection system 7 are provided in the gate valve chamber 5 does not have the above-described problem, and has high accuracy High detection is possible. Further, during the processing of the substrate 9, the opening 54 of the processing chamber 1 is hermetically closed by the gate valve 51, so that a gas having a deposition action reaches the gate valve chamber 5 and the mark detection window 56 or the vertical direction detection window. 57 will not be fogged.

  In the above embodiment, the vertical direction detection system 7 is provided separately from the mark detection system 6 to detect the vertical direction, but a configuration in which the mark detection system 6 detects the vertical direction is also conceivable. For example, a configuration in which an imaging optical system with a shallow depth of focus is adopted as the optical system 622 provided in the detector 62 and the position of the mark 25 in the Z direction is obtained from the degree of blurring (focused state) of the image is conceivable. .

  Further, the configuration in which the detector 62 is an image sensor that takes images of the end effector 23 and the substrate 9 and detects the positions of the end effector 23 and the substrate 9 without stopping the end effector 23 is a point of productivity of the apparatus. It has important significance. In other words, the position detection during the conveyance can be performed by stopping the end effector 23 during the conveyance. However, stopping the end effector 23 in the middle of the conveyance will lengthen the conveyance time and ultimately reduce the productivity of the apparatus. On the other hand, if the configuration in which the image is held under a certain condition as in the present embodiment and detection is performed without stopping the end effector 23 during the conveyance, the productivity of the apparatus does not decrease.

  As described above, in the apparatus according to the present embodiment, the position shift of the end effector 23 or the substrate 9 is detected while the substrate 9 is being transported from the transport start position to the set arrangement position. This configuration is important in terms of improving the productivity of the entire apparatus. In other words, if the configuration is such that the positional deviation is detected after being arranged at the set arrangement position and this is corrected, it is not possible to proceed to the next operation unless these operations are completed after the conveyance to the set arrangement position is completed. However, if the configuration is such that misalignment is detected in the middle of conveyance, it is possible to perform arithmetic processing necessary for detecting misalignment in parallel with the conveyance. Further, in some cases, by correcting the transport direction and transport distance in the middle, the substrate 9 can be correctly placed at the set placement position without requiring any excess time. For this reason, it can contribute to the improvement of productivity.

  Further, in the present embodiment, as described above, teaching is performed, and positional deviation is detected by comparison with teaching data. Therefore, highly accurate detection can be performed, and as a result, the substrate 9 is accurately placed at the set placement position. be able to. In addition, since the teaching data in the reference memory can be updated, it is possible to effectively cope with a case where the accuracy of the substrate transport system is gradually lowered due to wear or the like. That is, by frequently teaching, updating the teaching data so as to compensate for such a decrease in accuracy of the substrate transport system, it becomes possible to always arrange the substrate 9 at the set arrangement position with high accuracy. .

  Furthermore, as described above, since the positional deviation is corrected by driving the end effector 23 by the transport robot 21, the configuration and operation required for correcting the positional deviation are simplified. The correction of the positional deviation can be performed by another mechanism that is driven at the set arrangement position. However, when this is done, the mechanism becomes complicated and the time required increases.

1A Processing chamber 1B Processing chamber 1C Processing chamber 1D Processing chamber 1E Processing chamber 1F Processing chamber 21 Transfer robot 23 End effector 25 Mark 20 Control unit 3 Transfer chamber 4 Load lock chamber 41 Cassette 5 Gate valve chamber 51 Gate valve 6 Mark detection system 62 Detector 621 CCD
7 Vertical Direction Detection System 8 Signal Processing Unit 81 Hold Circuit 811 Hold Memory 812 Judgment Circuit 82 Image Processing Circuit 83 Reference Memory 84 Arithmetic Circuit 9 Substrate 91 Holder Jig 92 Substrate Jig

Claims (9)

A substrate processing apparatus comprising: a processing chamber that performs predetermined processing on a substrate inside; and a substrate transfer system that carries an unprocessed substrate into the processing chamber and unloads the processed substrate from the processing chamber. ,
The substrate transport system has an end effector that holds the substrate and a drive mechanism that drives the end effector, and the end effector is provided with at least two marks.
An image sensor that simultaneously detects these marks and the substrate;
A substrate processing apparatus, comprising: a signal processing unit that processes a signal from an image sensor to obtain a positional deviation from a correct position of the substrate. The signal processing unit obtains data on the positions of the marks from the detected marks, obtains data on the positions of the edges of the boards from the detected boards, and shifts the position of the board relative to the correct positions according to these data. The substrate processing apparatus according to claim 1, wherein: The positional deviation is a deviation of the reference point of the substrate with respect to a reference point on the end effector, and the arithmetic processing unit calculates the reference point of the substrate from the detected mark and substrate, and according to the calculation result 3. The substrate processing apparatus according to claim 1, wherein the displacement is obtained. The reference point of the substrate is the center of the substrate, and the arithmetic processing unit identifies two points on the edge of the substrate from the position of the mark, and the data of the positions of the two points on the edge and known 4. The substrate processing apparatus according to claim 3, wherein the position of the center of the substrate is calculated according to the value of the radius of the substrate as a reference point position. The correct position is such that the substrate is transported from a transport start position, which is the position of the substrate when transporting the substrate, to a set placement position that is finally set as a position where the substrate is to be placed. The substrate processing apparatus according to claim 1, wherein the substrate processing apparatus is a position taken by the substrate during transfer when the end effector moves. The signal processing unit obtains a positional deviation by comparing the reference memory in which the correct position data is stored with the substrate position data detected during actual conveyance and the correct position data. And the reference memory is arranged in the image sensor in the middle of returning the substrate or substrate jig to the transfer start position after actually correctly arranging the substrate or substrate jig at the set arrangement position. 6. The substrate processing apparatus according to claim 5, wherein an image signal obtained by imaging can be stored and updated. There are a plurality of the transfer start positions and the set arrangement positions, and the image sensors are arranged so that a positional deviation from the correct position is obtained in the middle of each transfer from the transfer start position to the set arrangement position. 7. The substrate processing apparatus according to claim 5, wherein the substrate processing apparatus is provided. 8. The drive mechanism of the substrate transport system drives the end effector so as to correct a required position shift. Substrate processing equipment. In the substrate processing apparatus, a transport step of transporting the substrate from the transport start position to the set arrangement position by holding the substrate on the end effector and moving the end effector by driving the drive mechanism;
An imaging step of imaging the substrate by an image sensor in the middle of conveying the substrate from the conveyance start position to a set arrangement position;
A signal processing step of processing an image signal from an image sensor to obtain a positional shift with respect to a correct position of the substrate,
The signal processing step processes the image signal from the image sensor to obtain the position of the peripheral edge of the substrate, calculates the position of the reference point of the substrate based on this, and then corrects the substrate based on the calculation result. A step of obtaining a positional deviation with respect to the position,
The movement of the end effector in the transporting process includes the movement of the end effector for correcting the positional deviation obtained in the signal processing process. Method.

Images