JP2011021916A - Position detector, method of detecting position, and substrate overlapping apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem wherein accurate position of a stage cannot be readily grasped, if the amount of rotation of the stage exceeds a prescribed amount when positioning and overlapping respective wafers, and it is necessary to strictly manage the position of a wafer initially placed on the stage for control so as to prevent the amount of rotation of the stage from exceeding a prescribed amount. <P>SOLUTION: A position detector includes a first imaging unit for imaging at least two first substrate indexes provided on a first substrate by single imaging operation; and a measurement section for measuring the attitude of the first substrate by measuring the position of the first substrate index, based on the images captured by the first imaging unit. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a position detection device, a position detection method, and a substrate overlaying device.

  One technique for improving the effective mounting density of semiconductor devices is a three-dimensional structure in which a plurality of semiconductor chips are stacked. In particular, a semiconductor chip manufactured by a procedure of laminating a plurality of wafers in the state of a wafer as a semiconductor substrate and joining them into individual pieces has recently attracted attention because of its high productivity. When two wafers are overlapped, alignment is performed while measuring the alignment marks of each other with a microscope (see, for example, Patent Document 1).

JP 2005-251972 A

  In order to align the alignment marks with each other, at least one of the stages holding the respective wafers is moved. The accuracy required for mutual alignment at this time depends on the line width of the circuit formed on the wafer surface, but this required accuracy tends to increase year by year as the mounting density increases. Therefore, a sensor that detects the movement of the stage is required to have high detection accuracy. One of the sensors that satisfy this requirement is an interferometer that is a laser light wave interference type length measuring device. However, for example, when the laser is projected from the interferometer installed at a fixed position with respect to the stage to the reflecting mirror fixed to the stage and the distance is measured using the reflected light, Sensitivity is very high. That is, even if the stage is slightly rotated, the interferometer cannot capture the reflected light of the laser, and the distance cannot be measured. Therefore, when the wafers are aligned and overlapped with each other, if the rotation amount of the stage exceeds a predetermined amount, it is difficult to grasp the exact position of the stage. In order to control the rotation amount of the stage so as not to exceed a predetermined amount, it is necessary to strictly manage the position of the wafer first placed on the stage.

  In order to solve the above problem, a position detection device according to a first aspect of the present invention includes a first imaging unit that images at least two first substrate indices provided on a first substrate by a single imaging operation; And a measurement unit that measures the position of the first substrate by measuring the position of the first substrate index based on the image captured by the first imaging unit.

  In order to solve the above-described problem, the position detection method according to the second aspect of the present invention images at least two first substrate indices provided on the first substrate by the first imaging unit by one imaging operation. The first imaging step includes a first measurement step in which the measurement unit measures the position of the first substrate index based on the image captured in the first imaging step, thereby measuring the posture of the first substrate.

  The above summary of the invention does not enumerate all necessary features of the present invention. In addition, a sub-combination of these feature groups can also be an invention.

It is the perspective view which looked down at the 1st substrate holder holding the 1st substrate from the upper part. It is the perspective view which looked down at the 3rd substrate holder holding the 2nd substrate from the upper part. It is a schematic structure figure of a substrate superposition system concerning this embodiment. It is a schematic diagram of the observation image of a 1st measurement microscope. It is a flowchart which shows the position alignment process which concerns on this embodiment.

  Hereinafter, the present invention will be described through embodiments of the invention, but the following embodiments do not limit the invention according to the claims. In addition, not all the combinations of features described in the embodiments are essential for the solving means of the invention.

  FIG. 1 is a perspective view of a first wafer holder 101 holding a first wafer 10 as a substrate, as viewed from above. The first wafer holder 101 has a holder body 110 and an adsorber 111, and has a disk shape whose diameter is slightly larger than that of the first wafer 10 as a whole. The holder body 110 is integrally formed of a highly rigid material such as ceramic or metal.

  The holder main body 110 has a region for holding the first wafer 10 on the surface thereof. This holding region is polished and has high flatness. The holding of the first wafer 10 is performed by suction using an electrostatic force. Specifically, a potential difference is applied between the first wafer holder 101 and the first wafer 10 by applying a voltage to the electrode embedded in the holder body 110 via a voltage application terminal provided on the back surface of the holder body 110. And the first wafer 10 is attracted to the first wafer holder 101.

  The adsorbers 111 are made of a magnetic material such as iron, and a plurality of adsorbers 111 are arranged on the outer surface of the first wafer 10 on the outer surface of the first wafer 10. Further, the adsorber 111 is disposed in a depressed region formed in the holder main body 110 so that the upper surface thereof is located in a plane substantially the same as the plane that holds the first wafer 10. In the case of the figure, a total of six adsorbers 111 are arranged every 120 degrees with two as one set.

  A plurality of first circuit regions 11 are built in the first wafer 10. As shown in the drawing, a plurality of first circuit regions 11 surrounded by a rectangle are two-dimensionally arranged on the upper surface of the first wafer 10. A first alignment mark 12 is provided inside each first circuit region 11. The first alignment mark 12 is an index used for position control of the first wafer 10. For each first alignment mark 12, for example, the design coordinate value is individually managed as a coordinate value with the center of the first wafer 10 as the origin.

  However, since the first alignment mark 12 is an index used for position control of the first wafer 10, it may not be provided in all the first circuit regions 11. It is sufficient that there is at least a number that can control the position of the first wafer 10. The first alignment mark 12 is provided inside the first circuit region 11, but does not need to be provided separately regardless of the circuit, and is a component of the circuit that can be used as an index. The first alignment mark 12 may be used. For example, in a stacked semiconductor element, TSV (Through Si Via) is often used for connection between layers, but the tip of the TSV can be used as the first alignment mark 12. Further, the first alignment mark 12 may be provided outside the first circuit region 11.

  The first notch 13 is a notch provided in the first wafer 10 and is an index used when the posture of the first wafer 10 is determined on the basis of the outer shape. For the first notch 13, for example, the design coordinate value is individually managed as a coordinate value with the center of the first wafer 10 as the origin.

  FIG. 2 is a perspective view of the second wafer holder 201 holding the second wafer 20 as a substrate as viewed from above. The second wafer holder 201 has a holder body 210 and a magnet 211, and has a disk shape whose diameter is slightly larger than that of the second wafer 20 as a whole. The holder body 210 is integrally formed of a highly rigid material such as ceramic or metal.

  The holder main body 210 includes a region for holding the second wafer 20 on the surface thereof. This holding region is polished and has high flatness. The holding of the second wafer 20 is performed by suction using an electrostatic force. Specifically, a potential difference is applied between the second wafer holder 201 and the second wafer 20 by applying a voltage to the electrode embedded in the holder main body 210 via a voltage application terminal provided on the back surface of the holder main body 210. And the second wafer 20 is attracted to the second wafer holder 201.

  A plurality of magnets 211 are arranged in an outer peripheral region that is outside the held second wafer 20 on the surface holding the second wafer 20. In addition, the magnet 211 is arranged so that its upper surface is located in a plane substantially the same as the plane that holds the second wafer 20. In the case of the figure, a total of six magnets 211 are arranged every 120 degrees with two as one set.

  The magnets 211 of the second wafer holder 201 are arranged so as to correspond to the attractors 111 of the first wafer holder 101, respectively. When the first wafer holder 101 holding the first wafer 10 and the second wafer holder 201 holding the second wafer 20 face each other and the magnet 211 and the attractor 111 act, the first wafer 10 and the second wafer The wafer 20 can be clamped and fixed in an overlapped state.

  A plurality of second circuit regions 21 are formed in the second wafer 20. As shown in the drawing, a plurality of second circuit regions 21 surrounded by a rectangle are two-dimensionally arranged on the upper surface of the second wafer 20. A second alignment mark 22 is provided inside each second circuit region 21. The second alignment mark 22 is an index used for position control of the second wafer 20. For each second alignment mark 22, for example, design coordinate values are individually managed as coordinate values with the center of the second wafer 20 as the origin.

  However, since the second alignment mark 22 is an index used for position control of the second wafer 20, it may not be provided in all the second circuit regions 21. It is sufficient that there is at least a number that can control the position of the second wafer 20. The second alignment mark 22 is provided inside the second circuit region 21, but need not be provided separately regardless of the circuit, and is a component of the circuit that can be used as an index. The second alignment mark 22 may be used. Further, the second alignment mark 22 may be provided outside the second circuit region 21.

  The second notch 23 is a notch provided in the second wafer 20 and is an index used when the posture of the second wafer 20 is determined on the basis of the outer shape. For the second notch 23, for example, the design coordinate value is individually managed as a coordinate value with the center of the second wafer 20 as the origin.

  FIG. 3 is a schematic configuration diagram of a substrate overlaying system including the substrate overlaying apparatus 30 according to the present embodiment. The substrate overlaying device 30 is a device that aligns and superimposes a plurality of substrates on each other's mounting surface, and particularly as a part of a system that performs three-dimensional mounting by superimposing wafers that are semiconductor substrates. Is preferred. Therefore, the substrate overlaying device is a device including a position detection device.

  The substrate overlaying system includes a transport robot 40 and a pre-aligner 50 in addition to the substrate overlaying apparatus 30. The transfer robot 40 includes a hand unit 401 that can hold the wafer and the wafer holder, and an arm unit 402 that carries the held wafer and wafer holder to a predetermined position in a predetermined posture by extending and contracting and rotating. The hand unit 401 includes a voltage supply terminal that applies a voltage to the voltage application terminal of the first wafer holder 101 or the second wafer holder 201 to be gripped. Accordingly, even when the hand unit 401 holds the first wafer holder 101 that holds the first wafer 10 or the second wafer holder 201 that holds the second wafer 20, the holding of each wafer can be continued. .

  The pre-aligner 50 includes an observation microscope 501 and a mounting table 502. The transfer robot 40 carries the first wafer holder 101 and the first wafer 10 or the second wafer holder 201 and the second wafer 20 onto the mounting table 502 sequentially or integrally. The observation microscope 501 has an imaging device connected to a control calculation unit (not shown) and an optical system that forms an overall image of the wafer placed on the mounting table 502 on the imaging device. For example, the control calculation unit acquires an entire image of the first wafer 10 mounted on the mounting table 502, and the attitude of the first wafer 10 on the mounting table 502 from the position of the first notch 13 in the image. Is detected. The attitude of the first wafer 10 detected here includes at least the center coordinates and the rotation amount of the first wafer 10 in the XY plane. Even when the second wafer 20 is mounted on the mounting table 502, the posture can be detected in the same manner.

  The substrate overlaying apparatus 30 includes a first measurement microscope 301 and a second measurement microscope 302 that face each other. The first measurement microscope 301 is fixed to the ceiling frame of the housing, and the second measurement microscope 302 is installed on the first stage 303. The first stage 303 is installed in the driving device 307 and moves in the six axis directions of the XYZ direction and the θxθyθz direction under the control of the control calculation unit. The movement distance in the XY direction and the rotation amount in the θz direction, which is the rotation direction around the Z axis, are measured by a reflection mirror 305 provided at the end of the first stage 303 by an interferometer 306 which is a laser light wave interference type length measuring device. Is detected. A plurality of interferometers 306 and reflection mirrors 305 are installed at appropriate positions so that the amount of movement in the XY direction and the amount of rotation in the θz direction can be detected.

  The first wafer holder 101 and the first wafer 10 held by the first wafer holder 101 are gripped by the transfer robot 40 after being identified by the pre-aligner 50, pass through the gate 308, and are transferred to the first stage 303. The first stage 303 fixes the first wafer holder 101 that holds the first wafer 10 by vacuum suction. The first stage 303 includes a voltage supply terminal that applies a voltage to the voltage application terminal of the first wafer holder 101. Accordingly, the first wafer 10 remains fixed to the first wafer holder 101, and the relative position and posture are maintained.

  Similarly, the second wafer holder 201 and the second wafer 20 held by the second wafer holder 201 are gripped by the transfer robot 40 after receiving the posture identification by the pre-aligner 50, pass through the gate 308, and transferred to the second stage 304. Is done. During this time, the transfer robot 40 rotates the hand unit 401 by 180 degrees so that the second wafer 20 faces downward. The second stage 304 is a stage that faces the first stage 303 and is fixed to the ceiling frame of the housing, and fixes the second wafer holder 201 holding the second wafer 20 downward by vacuum suction. The second stage 304 includes a voltage supply terminal that applies a voltage to the voltage application terminal of the second wafer holder 201. Accordingly, the second wafer 20 remains fixed to the second wafer holder 201, and the relative position and posture are maintained.

  The first measurement microscope 301 forms an image of a part of the first wafer 10 placed on the first stage 303 via the first wafer holder 101 with the image sensor connected to a control calculation unit (not shown). A first imaging unit including an optical system to be configured. Similarly, the second measurement microscope 302 captures an image of an image sensor connected to a control arithmetic unit (not shown) and a part of the second wafer 20 placed on the second stage 304 via the second wafer holder 201. A second imaging unit including an optical system for forming an image on the second imaging unit.

  When observing a predetermined portion of the first wafer 10 with the first measurement microscope 301, the control calculation unit drives the driving device 307 so that the observation portion is located within the field of view of the first measurement microscope 301. One stage 303 is moved. When observing a predetermined portion of the second wafer 20 with the second measurement microscope 302, the control calculation unit drives the driving device 307 so that the observation portion is located within the field of view of the second measurement microscope 302. One stage 303 is moved. Note that the optical axes of the first measurement microscope 301 and the second measurement microscope 302 are adjusted in advance so as to coincide with each other when facing each other.

  The first measurement microscope 301 has an angle of view that allows at least two or more of the first alignment marks 12 on the first wafer 10 to be observed. Further, the resolution is high enough to identify the positions of the first alignment marks 12. Specifically, for example, when the first alignment mark 12 is a mark that is orthogonal with a line width of 5 μm, the resolution and angle of view have a resolution that can identify this orthogonal point and can secure a field of view of about 1 mm square. is there. The resolution and angle of view of the second measuring microscope 302 are the same as those of the first measuring microscope 301.

  Although detailed positioning will be described later, the precise postures of the first wafer 10 and the second wafer 20 are grasped by the first measurement microscope 301 and the second measurement microscope 302, and the first stage 303 is moved based on this. By doing so, the faces facing each other are brought into contact with each other. Then, the first wafer holder 101 and the second wafer holder 201 are integrated with each other while the first wafer 10 and the second wafer 20 are sandwiched by the action of the adsorbent 111 of the first wafer holder 101 and the magnet 211 of the second wafer holder 201. And fix. The integrated wafer holder pair is gripped by the hand unit 401 of the transfer robot 40 and is carried out of the substrate superposing apparatus 30 toward the apparatus responsible for the next process.

  FIG. 4 is a schematic diagram of an observation image of the first measurement microscope 301. Although the figure is represented by a circle in accordance with the optical system, it is sufficient that at least two images of the first alignment mark 12 are formed on a rectangular image sensor.

  The control calculation unit captures such an observation image by an imaging operation with a single imaging operation. The captured image data is sent to a control calculation unit that functions as a measurement unit. The control calculation unit obtains the actual alignment mark coordinate values from the coordinates of at least two alignment mark images existing in the image data. Specifically, the reference pixel value of the first alignment mark 12 with the image center as a reference is determined by image processing, and the magnification with the real world by the magnification of the optical system of the first measurement microscope 301, the pixel pitch of the image sensor, and the like. After performing the conversion, the conversion to match the coordinate system of the first stage 303 is performed. Here, the reference pixel value refers to a coordinate value of a pixel where the reference point is the reference point in the case of an alignment mark represented by a cross, for example, and the reference point is present on the image data. Of course, the reference pixel value does not have to be an integer value, and may be a sub-pixel order value calculated as an intersection from the crosshairs.

  The first wafer 10 is fixed to the first stage 303 via the first wafer holder 101, and the first stage 303 can move in the XY plane direction relative to the first measurement microscope 301. Therefore, originally, in order to make the best use of the resolution of the image sensor, the position of the first alignment mark 12 is captured in the full field of view, the position is measured, and another first alignment mark is moved while the first stage 303 is moved. It seems that the posture can be measured more accurately by capturing 12. However, according to such a measurement method, it is necessary to move the first stage 303 each time, so that a stage movement error is included between the measurement coordinates of the two first alignment marks 12. In addition, there is a problem that it takes time for measurement because of the movement.

Therefore, as in this embodiment, the field of view of the first measurement microscope 301 is determined so that two or more first alignment marks 12 are included by one imaging operation. And the position coordinate of each 1st alignment mark 12 is measured at once. Specifically, as shown, for example, the coordinates P ₁ (x ₁ , y ₁ ) of the first alignment mark 14 and the coordinates P ₂ (x ₂ , y ₂ ) of the first alignment mark 15 are obtained.

Any number of calculation methods for identifying the posture of the first wafer 10 from the coordinates P ₁ and the coordinates P ₂ are assumed. As an example, the coordinates P ₁ and the coordinate P ₂ is the actual coordinate values, there is a method of comparing the nominal coordinate values. The design coordinate value of the first alignment mark 14 with respect to the center of the first wafer 10 and the design coordinate value of the first alignment mark 15 with respect to the center of the first wafer 10 are stored in advance in a memory (not shown). In this case, assuming that the first alignment mark 14 and the first alignment mark 15 are accurately marked, by measuring the _two coordinates P ₁ and P ₂ and comparing them with the design coordinate values stored in the memory, The attitude of the first wafer 10 can be identified. For example, when the first wafer 10 is placed on the first stage 303, it is assumed that the control target is to make the orthogonal coordinate system of the first wafer 10 coincide with the orthogonal coordinate system of the first stage 303. In this case, the design coordinate value matches the actual coordinate value if the control can be performed according to the control target. On the contrary, if they do not match, the error amount becomes a deviation from the control target value, so that the center position of the first wafer 10 with respect to the coordinate system of the first stage 303 and the rotation amount of the first wafer 10 can be determined. it can.

As another example, there is a method of observing alignment marks in a specific positional relationship. Since the pre-aligner 50 acquires the entire image of the first wafer 10, at this time, the first alignment mark 14 and the first alignment mark 15 that exist in parallel with any axis in the orthogonal coordinate system of the first wafer 10. Find it. When the first wafer 10 is placed on the first stage 303 and the control target is to make the orthogonal coordinate system of the first wafer 10 coincide with the orthogonal coordinate system of the first stage 303, these alignment marks are It can be determined that the inclination of the straight line connecting the two measured coordinates P ₁ and P ₂ is the rotation amount of the first wafer 10 with respect to the coordinate system of the first stage 303. At the same time, the center position of the first wafer 10 relative to the coordinate system of the first stage 303 can be obtained from the intersection of the straight line connecting the _two coordinates P ₁ and P ₂ and the orthogonal coordinate system of the first stage 303. With such a method, the design coordinate values of the first alignment marks 14 and 15 need not be stored in advance. As a control target, even when the rotation amount of the orthogonal coordinate system of the first wafer 10 is given to the orthogonal coordinate system of the first stage 303, the rotation amount of the first stage 303 is taken into account by taking the rotation amount into consideration. The center position of the first wafer 10 with respect to the coordinate system and the rotation amount of the first wafer 10 can be determined.

  However, the first alignment mark 14 and the first alignment mark 15 may actually be marked with a deviation from the design coordinate value. In this case, for example, the coordinates to be measured are further increased, a more probable approximate straight line is obtained by using the least square method or the like, the center position of the first wafer 10 with respect to the coordinate system of the first stage 303, and the rotation of the first wafer 10 The amount may be determined.

  Similarly, by measuring the second alignment mark 22, the center position of the second wafer 20 with respect to the coordinate system of the second stage 304 and the rotation amount of the second wafer 20 can be determined. However, if the posture identification of the first wafer 10 and the posture identification of the second wafer 20 are performed independently, the following problems may occur.

  That is, in order to align the first alignment mark 12 and the second alignment mark 22 corresponding to each other, the first stage 303 is moved. The accuracy required for mutual alignment at this time depends on the line widths of the first circuit region 11 and the second circuit region 21. A laser light wave interferometer 306 is used as a sensor that can detect the movement of the first stage 303 with the accuracy of the line width. The interferometer 306 is very sensitive to the rotation of the first stage 303. Specifically, when the first stage 303 rotates by ± 0.5 mrad or more, the amount of rotation cannot be detected. Therefore, even if the precise postures of the first wafer 10 and the second wafer 20 can be identified, if the rotation amount of the first stage 303 necessary for alignment exceeds 1 mrad, alignment can be performed as a result. There will be no.

  In order to avoid such a problem, the first wafer 10 and the second wafer 20 are aligned by the following steps. FIG. 5 is a flowchart showing the alignment process according to the present embodiment. Unless otherwise specified, the processing of each step is executed directly by the control calculation unit or the control of each element.

  When the alignment process is started by the substrate superposition system, first, in step S101, the posture of the first wafer 10 carried into the pre-aligner 50 by the transfer robot 40 and held by the first wafer holder 101 is measured. The posture measurement here is based on the center coordinates and rotation of the first wafer 10 in the XY plane based on the position of the first notch 13 in the image from the entire image of the first wafer 10 acquired using the observation microscope 501. Find the amount.

  In step S102, a transport path to the target position on the first stage 303 is generated based on the center coordinates and the rotation amount of the first wafer 10 measured in step S101. The target position includes the amount of rotation of the first wafer 10 on the first stage 303.

  The transfer robot 40 loads the first wafer holder 101 holding the first wafer 10 to the target position on the first stage 303 according to the generated transfer path. The first stage 303 fixes the first wafer holder 101 by vacuum suction.

  When the first wafer 10 is loaded and fixed on the first stage 303 in step S102, the posture of the first wafer 10 on the first stage 303 is accurately measured in step S103.

  The posture measurement based on the first notch 13 by the observation microscope 501 of the pre-aligner 50 is executed by observing the entire first wafer 10, and thus the accuracy is coarser than the posture measurement based on the first alignment mark 12. Further, the error with respect to the target position is also accumulated in the transfer control of the transfer robot 40 along the transfer path generated in step S102. Therefore, the attitude of the first wafer 10 is accurately measured again on the first stage 303 on the basis of the first alignment mark 12 in order to drive the accuracy required for the alignment of the first wafer 10 and the second wafer 20.

  Specifically, as described above, the first stage 303 is moved so that two or more first alignment marks 12 are included in the field of view of the first measurement microscope 301. Then, after obtaining the coordinates of the first alignment mark 12 from the image data acquired by one imaging operation, the center position of the first wafer 10 and the rotation amount of the first wafer 10 are measured.

In step S104, the actual posture of the first wafer 10 is compared with the target posture. Specifically, it is determined whether or not the rotation error amount R ₁ for the target posture of the actual posture is smaller than the rotation allowable error amount R ₀ for the predetermined target posture. If it is smaller, the process proceeds to step S106, and if it is larger, the process proceeds to step S105. Note that the rotation allowable error amount R ₀ is set to, for example, a value that is half of the rotation amount at which the interferometer 306 cannot measure the position of the first stage 303.

  In step S <b> 105, the first wafer holder 101 placed on the first stage 303 is returned to the pre-aligner 50 together with the first wafer 10. Then, the carrying-in process from step S101 to the first stage 303 is performed again.

  From step S106, the posture of the second wafer 20 is identified. In step S <b> 106, the posture of the second wafer 20 held in the second wafer holder 201 carried into the pre-aligner 50 by the transfer robot 40 is measured. The posture measurement here is based on the center coordinates and rotation of the second wafer 20 in the XY plane, based on the position of the second notch 23 in the image, from the entire image of the second wafer 20 acquired using the observation microscope 501. Find the amount.

In step S107, based on the center coordinates and rotation amount of the second wafer 20 measured in step S106 and the center coordinates and rotation amount of the first wafer 10 measured in step S103, the target position on the second stage 304 is reached. Generate a transport path. That, considering the first wafer 10 on the first stage 303, if it is detected to have a rotation error amount R ₁ relative to the target position, so as to cancel this, the target posture of the second wafer 20 To generate a transport path. This transfer path includes an operation of rotating the second wafer holder 201 holding the second wafer 20 180 degrees so that the second wafer 20 faces downward.

  The transfer robot 40 loads the second wafer holder 201 holding the second wafer 20 to the target position on the second stage 304 according to the generated transfer path. Then, the second stage 304 fixes the second wafer holder 201 by vacuum suction.

  When the second wafer 20 is loaded onto the second stage 304 and fixed in step S107, the posture of the second wafer 20 on the second stage 304 is accurately measured in step S108. Specifically, the first stage 303 is moved so that two or more second alignment marks 22 are included in the field of view of the second measurement microscope 302. Then, after obtaining the coordinates of the second alignment mark 22 from the image data acquired by one imaging operation, the center position of the second wafer 20 and the rotation amount of the second wafer 20 are measured.

In step S109, the actual posture of the second wafer 20 is compared with the target posture. Specifically, it is determined whether or not the rotation error amount R ₂ for the target posture of the actual posture is smaller than the rotation allowable error amount R ₀ for the predetermined target posture. If it is smaller, the process proceeds to step S111. If it is larger, the process proceeds to step S110.

  In step S <b> 110, the second wafer holder 201 placed on the second stage 304 is returned to the pre-aligner 50 together with the second wafer 20. Then, the carrying-in process from step S106 to the second stage 304 is performed again.

  Through the steps up to step S109, the position and orientation of the first wafer 10 on the first stage 303 and the position and orientation of the second wafer 20 on the second stage 304 can be accurately grasped. Further, it is ensured that the relative deviation between the first wafer 10 and the second wafer 20 is within the allowable rotation angle of the first stage 303. In step S111, the first stage 303 is moved so that each of the second alignment marks 22 on the second wafer 20 is opposed to the corresponding first alignment mark 12 on the first wafer 10.

  In step S112, the first stage 303 is moved in the Z-axis direction, the first wafer 10 and the second wafer 20 are brought into contact with each other, and temporary bonding is performed. When the temporary bonding is completed, in step S113, the attracting member 111 of the first wafer holder 101 and the magnet 211 of the second wafer holder 201 are operated to perform clamping. The first wafer holder 101 and the second wafer holder 201 that have been clamped are sandwiched between the first wafer 10 and the second wafer 20 and fixed together.

  The integrated wafer holder pair leaves the second stage 304 and is pulled down as the first stage 303 is lowered. In step S114, the wafer holder pair is held by the hand unit 401 of the transfer robot 40, and the substrate stacking apparatus 30 is moved. Is carried out toward the device responsible for the next process. Thus, a series of processing ends.

  As described above, a plurality of first alignment marks 12 and a plurality of second alignment marks 22 corresponding to the first alignment marks 12 are provided. However, these are not necessarily provided at positions according to design values. Therefore, in the mutual alignment, for example, the alignment is performed so that the distance between the alignment marks corresponding to each other becomes the minimum value as a whole.

In the present embodiment, since the first wafer 10 is moved by the first stage 303 and aligned with the second wafer 20, the positional deviation of the first alignment mark 12 with respect to the second alignment mark 22 is set as a relative error. This relative error is approximately expressed by a rotation amount θ due to the rotation of the first wafer 10 and an offset amount (x ₀ , y ₀ ) due to an offset in the x direction and the y direction as main error parameters. Hereinafter, an example in which these are used as error parameters will be described.

ただし、ｎ＝１，２，３
（１） Three second alignment marks 26, 27, 28 to be measured are provided on the second wafer 20, and the respective coordinate values are (x ₂₁ , y ₂₁ ), (x ₂₂ , y ₂₂ ), ( x ₂₃ , y ₂₃ ). In addition, first alignment marks 16, 17, 18 corresponding to these are provided on the first wafer 10, and the coordinate values thereof are (x ₁₁ , y ₁₁ ), (x ₁₂ , y ₁₂ ), ( x ₁₃ , y ₁₃ ). The coordinate values obtained by converting the coordinate values of the first alignment marks 16, 17, and 18 by the coordinate conversion formula using the rotation amount θ and the offset amount (x ₀ , y ₀ ) are respectively (x _c1 , y _c1 ), ( x _c2 , y _c2 ), (x _c3 , y _c3 ). At this time, the coordinate conversion formula is expressed by formula (1).

However, n = 1, 2, 3
(1)

（２） The error parameters θ and (x ₀ , y ₀ ) are those in which the residual sum of squares expressed by the following equation (2) is minimized by using the coordinate values of the second alignment marks 26, 27, and 28. As required.

(2)

  Specifically, since it is calculated by a known statistical method such as a least square method, the details are omitted.

When θ and (x ₀ , y ₀ ), which are error parameters, are determined in this way, the coordinate conversion formula (1) is determined. Thereby, in addition to the relative attitude difference between the first wafer 10 and the second wafer 20, the degree of coincidence between the alignment marks can be increased by moving the first stage 303 in consideration of the calculated error parameter. Higher alignment can be achieved. When performing alignment with such an error parameter in mind, the above calculation process is added to step S111 and executed. At this time, a plurality of alignment marks corresponding to each other as targets include an operation of sequentially measuring the positions using the first measurement microscope 301 and the second measurement microscope 302 while moving the first stage 303.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

  It should be noted that the execution order of the above-described processes can be realized in an arbitrary order unless otherwise specified. Even if the operation flow in the claims, the specification, and the drawings is described using “next” or the like for convenience, it does not mean that the operation is essential in this order.

  10 1st wafer, 11 1st circuit area, 12 1st alignment mark, 13 1st notch, 14, 15, 16, 17, 18 1st alignment mark, 20 2nd wafer, 21 2nd circuit area, 22 2nd Alignment mark, 23 Second notch, 26, 27, 28 Second alignment mark, 30 Substrate overlay device, 40 Transfer robot, 50 Pre-aligner, 101 First wafer holder, 110 Holder body, 111 Adsorber, 201 Second wafer holder, 210 holder main body, 211 magnet, 301 first measurement microscope, 302 second measurement microscope, 303 first stage, 304 second stage, 305 reflection mirror, 306 interferometer, 307 driving device, 308 gate, 401 hand unit, 402 arm 501 Observation microscope, 50 Mounting table

Claims (19)

A first imaging unit that images at least two first substrate indices provided on the first substrate by one imaging operation;
A position detection apparatus comprising: a measurement unit that measures the position of the first substrate by measuring a position of the first substrate index based on an image captured by the first imaging unit. A calculation unit that calculates a first error between the posture of the first substrate measured by the measurement unit and the target posture of the first substrate;
2. The position according to claim 1, further comprising: a control unit that carries in the second substrate opposite to the first substrate, taking into account the first error calculated by the calculation unit with a target posture of the second substrate. Detection device.   The position detection device according to claim 2, wherein when the first error is larger than a predetermined allowable error, the control unit does not carry in the second substrate.   4. The position detection device according to claim 3, wherein when the first error is larger than the predetermined allowable error, the control unit causes the first substrate to be carried in again. A second imaging unit that images at least two second substrate indices provided on the second substrate by one imaging operation;
The said measurement part measures the attitude | position of the said 2nd board | substrate by measuring the position of the said 2nd board | substrate parameter | index based on the image imaged by the said 2nd imaging unit, The any one of Claim 2 to 4 The position detection apparatus described in 1. The calculating unit calculates a second error between the posture of the second substrate measured by the measuring unit and a target posture of the second substrate;
The control unit moves at least one of a first stage holding the first substrate and a second stage holding the second substrate based on the second error calculated by the calculation unit, and The position detection device according to claim 5, wherein the first substrate and the second substrate are opposed to each other.   The position detection device according to claim 6, wherein when the second error is larger than a predetermined allowable error, the control unit does not move the first stage and the second stage.   The position detection device according to claim 6 or 7, wherein when the second error is larger than a predetermined allowable error, the control unit causes the second substrate to be carried in again. The controller is
A transformed coordinate value obtained by converting measured coordinate values of the plurality of first substrate indices measured by the measuring unit using a coordinate transformation formula defined by a predetermined error parameter, and a plurality of the second measured by the measuring unit. The error parameter is calculated so that the difference from the measured coordinate value of the substrate index is minimized,
The first substrate and the second substrate are moved by moving at least one of the first stage holding the first substrate and the second stage holding the second substrate in consideration of the calculated error parameter. The position detection device according to any one of claims 6 to 8, wherein   A substrate superposition apparatus comprising the position detection device according to claim 1. A first imaging step of imaging at least two first substrate indices provided on the first substrate by the first imaging unit by one imaging operation;
A position detection method comprising: a first measurement step of measuring a posture of the first substrate by measuring a position of the first substrate index based on an image captured in the first imaging step. A first error calculating step for calculating a first error between the posture of the first substrate measured in the first measuring step and the target posture of the first substrate;
The loading step of loading the second substrate opposite to the first substrate by adding the first error calculated in the first error calculation step to the target posture of the second substrate. The position detection method as described in.   The position detection method according to claim 12, further comprising a carry-in prohibiting step for prohibiting execution of the carry-in step when the first error is larger than a predetermined allowable error.   The position detection method according to claim 13, further comprising a first re-loading step of reloading the first substrate when the first error is larger than the predetermined allowable error. A second imaging step of imaging at least two second substrate indices provided on the second substrate by a second imaging unit by one imaging operation;
And a second measurement step of measuring the posture of the second substrate by the measurement unit measuring the position of the second substrate index based on the image captured in the second imaging step. The position detection method according to any one of 12 to 14. A second error calculating step of calculating a second error between the posture of the second substrate measured in the second measuring step and the target posture of the second substrate;
Based on the second error calculated in the second error calculating step, at least one of the first stage holding the first substrate and the second stage holding the second substrate is moved to move the first stage. The position detection method according to claim 15, further comprising a moving step of causing the substrate and the second substrate to face each other.   The position detection method according to claim 16, further comprising a movement prohibiting step for prohibiting the execution of the moving step when the second error is larger than a predetermined allowable error.   18. The position detection method according to claim 16, further comprising a second re-loading step of re-loading the second substrate when the second error is larger than a predetermined allowable error. A transformed coordinate value obtained by converting measured coordinate values of the plurality of first substrate indices measured by the measuring unit using a coordinate transformation formula defined by a predetermined error parameter, and a plurality of the second measured by the measuring unit. A calculation step for calculating the error parameter so that a difference from the measured coordinate value of the substrate index is minimized;
In consideration of the error parameter calculated in the calculating step, at least one of the first stage holding the first substrate and the second stage holding the second substrate is moved to move the first substrate and the first substrate. The position detection method according to any one of claims 16 to 18, further comprising an opposing step of opposing the two substrates.

Images