JP5504840B2 - Superposition apparatus, superposition method and device manufacturing method for superposing a plurality of substrates - Google Patents

Description

  The present invention relates to a superposition apparatus, a superposition method, and a device manufacturing method for superposing a plurality of substrates.

  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

  There is a demand for an overlay apparatus that can move the other wafer with respect to one wafer accurately and at high speed in accordance with the alignment command. However, when superposition is performed at high speed, the wafer load is applied to the wafer support stage after superposition at the same time, and the stage supporting the respective wafers collides after being separated once. Has occurred. That the stages collide means that the supported wafer is struck by the other stage. Such a phenomenon has caused physical damage and electrical failure of the wafer.

  In order to solve the above-described problem, in the first aspect of the present invention, there is provided a superposing apparatus for superposing a plurality of substrates, the first stage holding the first substrate, and the first substrate being opposed to the first substrate. When the second substrate holding the second substrate to be moved and the first substrate and the second substrate are overlapped and separated from the first stage together with the second stage, the first substrate collides with the first stage by reaction. A collision prevention unit for preventing the collision.

  In order to solve the above-mentioned problem, in the second aspect of the present invention, there is provided a superposition method for superposing a plurality of substrates, a first holding step for holding the first substrate on the first stage, and a second stage And a second holding step for holding the second substrate facing the first substrate, a moving step for moving at least the second stage toward the first stage by the driving unit, and a second step from the second stage to the first stage. A detection step that detects that the driving force of the two stages can be applied, and after the detection unit detects that the driving force can be applied to the first stage in the detection step. Until the first stage releases the holding of the first substrate and the load on the first substrate acts on the second stage, at least a driving force corresponding to the load on the first substrate is applied from the second stage. And a control step of controlling the driving unit so as to act on the first stage.

  In order to solve the above problems, in a third aspect of the present invention, there is provided a superposition method for superposing a plurality of substrates, a first holding step for holding the first substrate on the first stage, and a second stage. A second holding step for holding the second substrate facing the first substrate, a moving step for moving the second stage toward at least the first stage by the drive unit, and the first stage holding the first substrate. In synchronism with the timing of releasing the gas, an air blowing step for injecting gas between the first stage and the first substrate is provided.

  In order to solve the above problems, in a fourth aspect of the present invention, there is provided a superposition method for superposing a plurality of substrates, a first holding step for holding the first substrate on the first stage, and a second stage. A second holding step for holding the second substrate facing the first substrate, a moving step for bringing the second stage and the first stage closer by the driving unit, and the first stage releases the holding of the first substrate. And a control step for controlling the drive unit so that the first stage and the second stage are separated from each other by a predetermined distance until the load on the first substrate acts on the second stage. .

  In order to solve the above problems, in a fifth aspect of the present invention, there is provided a superposition method for superposing a plurality of substrates, a first holding step for holding the first substrate on the first stage, and a second stage. In addition, a second holding step for holding the second substrate facing the first substrate, a moving step for bringing the second stage and the first stage closer by the driving unit, and an overlap for overlapping the first substrate and the second substrate The feedback gain value of the driving unit is set to a predetermined value at a predetermined time between the alignment step, the non-contact detection step for detecting non-contact between the first substrate and the first stage, and the overlay step and the non-contact detection step. And a gain changing step for making the value smaller than a value before the time point.

  In order to solve the above problems, in a sixth aspect of the present invention, there is provided a superposition method for superposing a plurality of substrates, a first holding step for holding the first substrate on the first stage, and a second stage. In addition, a second holding step for holding the second substrate facing the first substrate, a moving step for bringing the second stage and the first stage closer by the driving unit, and the first substrate and the second substrate are overlapped. At the predetermined time between the overlay step, the non-contact detection step for detecting non-contact between the first substrate and the first stage, and the overlap step and the non-contact detection step, the control of the driving unit is opened from the feedback control to the open loop. A control switching step for switching to control.

  In order to solve the above-described problem, in a seventh aspect of the present invention, there is provided a device manufacturing method manufactured by stacking a plurality of substrates, wherein the step of stacking the plurality of substrates is performed on the first stage. A first holding step for holding one substrate, a second holding step for holding the second substrate facing the first substrate on the second stage, and moving at least the second stage toward the first stage by the drive unit A moving step, a detecting step for detecting that the driving force of the second stage can be applied from the second stage to the first stage, and an operation of the driving force on the first stage being possible in the detecting step After the detection unit detects that the first substrate is released, the first stage releases the holding of the first substrate and the load on the first substrate acts on the second stage. And a driving force of the magnitude corresponding to the weight of the second stage control step of controlling the driving unit so as to act on the first stage.

  In order to solve the above problems, according to an eighth aspect of the present invention, there is provided a device manufacturing method manufactured by stacking a plurality of substrates, wherein the step of stacking the plurality of substrates is performed on the first stage. A first holding step for holding one substrate; a second holding step for holding the second substrate opposite to the first substrate; and a movement for causing the second stage and the first stage to approach each other by the driving unit. The first stage and the second stage are larger than a predetermined distance at a predetermined time between the step and the first stage releases the holding of the first substrate and the load of the first substrate acts on the second stage. And a control step of controlling the drive unit so as to be separated.

  In order to solve the above-described problem, in a ninth aspect of the present invention, there is provided a device manufacturing method manufactured by stacking a plurality of substrates, wherein the step of stacking the plurality of substrates is performed on the first stage. A first holding step for holding one substrate; a second holding step for holding the second substrate opposite to the first substrate; and a movement for causing the second stage and the first stage to approach each other by the driving unit. The first stage and the second stage are larger than a predetermined distance at a predetermined time between the step and the first stage releases the holding of the first substrate and the load of the first substrate acts on the second stage. And a control step of controlling the drive unit so as to be separated.

  In order to solve the above problems, according to a tenth aspect of the present invention, there is provided a device manufacturing method manufactured by stacking a plurality of substrates, wherein the step of stacking the plurality of substrates is performed on the first stage. A first holding step for holding one substrate; a second holding step for holding the second substrate opposite to the first substrate; and a movement for causing the second stage and the first stage to approach each other by the driving unit. A step of superimposing the first substrate and the second substrate, a non-contact detection step for detecting non-contact between the first substrate and the first stage, and a predetermined step between the overlay step and the non-contact detection step. And a gain changing step of making the feedback gain value of the drive unit smaller than a value before a predetermined time point.

  In order to solve the above problems, in an eleventh aspect of the present invention, there is provided a device manufacturing method manufactured by stacking a plurality of substrates, wherein the step of stacking the plurality of substrates is performed on the first stage. A first holding step for holding one substrate; a second holding step for holding the second substrate opposite to the first substrate; and a movement for causing the second stage and the first stage to approach each other by the driving unit. A step of superimposing the first substrate and the second substrate, a non-contact detection step for detecting non-contact between the first substrate and the first stage, and a predetermined step between the overlay step and the non-contact detection step. And a control switching step of switching the control of the drive unit from feedback control to open loop control at the time.

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

It is sectional drawing which shows a superposition apparatus roughly. It is a top view which shows a superposition apparatus roughly. It is explanatory drawing which shows the vicinity of an upper table roughly. It is a perspective view which shows a coarse movement stage part. It is a schematic sectional drawing in alignment with BB of FIG. It is a perspective view which shows a self-weight canceller. It is a schematic sectional drawing in alignment with CC of FIG. FIG. 6 is a cross-sectional view taken along the line BB in FIG. 2 and is a partial detail view seen from the opposite direction to FIG. 5. FIG. 3 is a partial detailed cross-sectional view along the line AA in FIG. 2. It is a top view which shows a fine movement stage part roughly. It is a schematic sectional drawing which shows the other structure of a superimposition apparatus. It is a perspective view which shows a mode that the 1st board | substrate holder was looked down from upper direction. It is a perspective view which shows a mode that the 1st board | substrate holder was looked up from the downward direction. It is a control flow figure of a superposition device. It is a control flow figure concerning the 1st control of a superposition device. It is a control flow figure concerning the 2nd control of a superposition device. It is explanatory drawing which shows roughly the vicinity of the upper table which concerns on a 1st modification. It is a control flow figure of the superposition device concerning the 1st modification. In the self-weight canceller which concerns on a 2nd modification, it is a schematic sectional drawing along the sectional line corresponded to CC of FIG. It is a control flow figure of the superposition device concerning the 2nd modification. It is a control flow figure of the superposition device concerning the 3rd modification. It is a block diagram which shows simply the position control system in the gravity direction of the lower table which concerns on a 3rd modification. It is a figure which shows the response characteristic when the value of the proportional gain Kp which concerns on a 3rd modification is changed.

  Hereinafter, the present invention will be described through embodiments of the invention and modifications thereof, but these do not limit the invention according to the claims. In addition, not all combinations of features described in the embodiment and its modifications are necessarily essential to the solution means of the invention.

  FIG. 1 is a cross-sectional view schematically showing a superposition apparatus 10 according to the present embodiment. As an overall configuration of the superimposing apparatus 10, first, on a coarse motion base member 13 installed on a floor 15 via a vibration isolation member 16, a coarse motion stage portion 14, and a lower table above the coarse motion stage portion 14. A fine movement stage unit 18 including 35 is provided. Further, in the recess 19 provided in the coarse motion base member 13, the upper table 24 is positioned above the lower table 35 and opposed to the lower table 35 in the body 21 installed via the vibration isolation member 20. Is provided. Therefore, the vibration isolation member 20 functions as a vibration isolator, and prevents the upper table 24 from vibrating.

  The coarse movement base member 13 is provided with an X-drive unit 33 and a Y-drive unit 34 on the upper surface thereof, and the XY-stage member 32 is driven in the xy directions. Therefore, the XY-stage member 32 can be moved to a predetermined position in the xy direction by an instruction from a control unit (not shown). The X-drive unit 33, the Y-drive unit 34, and the XY-stage member 32 constitute the coarse movement stage unit 14. The vertical direction of the apparatus is the z-axis direction, and the plane orthogonal to the z-axis direction is the xy plane.

  The Y-drive unit 34 includes a pair of Y-linear motors 53 that apply a moving force along the Y-axis direction to the XY-stage member 32 and the X-drive unit 33, and the XY-stage member 32 and the X-drive unit 33. And a Y-guide member 56 for guiding the movement of. That is, the XY-stage member 32 and the X-drive unit 33 move integrally in the y-axis direction. The magnets 54 of the pair of Y-linear motors 53 are fixed to a pair of edges 13b facing each other in the x-axis direction of the coarse motion base member 13 so as to extend along the y-axis direction and be parallel to each other. Yes. The coils 55 of the pair of Y-linear motors 53 are provided on the connecting members 50 and 51 and engage with the magnets 54 in a non-contact state.

  The Y-guide member 56 is fixed to the coarse motion base member 13 so as to extend along the y-axis direction between the pair of magnets 54. An engaging portion 57 provided on one connecting member 50 is engaged with the Y-guide member 56 through an air bearing 58 in a non-contact state. When each Y-linear motor 53 is driven, the engaging portion 57 is guided by the Y-guide member 56, whereby the XY-stage member 32 and the X-driving portion 33 move along the y-axis direction.

  The connecting members 50 and 51 are members for connecting and integrating the two X-guide members 46 constituting the X-driving unit 33 with each other at their respective ends. Air pads 52 are provided on the bottom surfaces of the connecting members 50 and 51, respectively. Thereby, the connection members 50 and 51 are installed at intervals without contacting the upper surface 13a of the coarse motion base member 13, respectively.

  The body 21 includes a fine movement base member 17, an upper plate member 22 disposed in parallel therewith, and a column member 23 that supports the upper plate member 22 on the fine movement base member 17, and these are integrated with high rigidity. It functions as a frame. The upper table 24 fixed to the upper plate member 22 holds the first substrate 11 that is a wafer on the lower surface thereof. A specific configuration near the upper table 24 will be described later.

  The column member 23 is provided with an X-interferometer 29 at a predetermined position with respect to the lower surface of the upper table 24. Specifically, on the surface of the column member 23 processed so as to be parallel to the yz plane, two locations separated in the y-axis direction and two locations separated in the z-axis direction with respect to each other Four X-interferometers 29 are installed at a total of four locations. From the outputs of these four interferometers, the position of the lower table 35 in the x-axis direction, the rotation direction around the y-axis, and the rotation amount around the z-axis can be measured. The X-interferometer 29 constitutes a position measurement unit that measures the position of the lower table 35.

  The fine movement stage unit 18 is supported by a self-weight canceller 36 that penetrates the XY-stage member 32. The self-weight canceller 36 is installed on the fine movement base member 17 of the body 21. Then, the second substrate 12 which is a wafer is held on the upper surface of the lower table 35 of the fine movement stage unit 18. A specific configuration in the vicinity of fine movement stage 18 will be described later. Note that the first substrate 11 and the second substrate 12 to be overlaid have semiconductor elements formed on the respective surfaces to be overlaid, and are joined after being overlaid so that a three-dimensional electric circuit is obtained. Realize the configuration. The first substrate 11 and the second substrate 12 may be a single wafer, or may be a wafer in which one or both are already laminated.

  FIG. 2 is a top view schematically showing the superimposing apparatus 10. FIG. 2 shows a state in which the upper plate member 22 and the column member 23 and the structures installed on them are removed from the body 21.

  The X-drive unit 33 guides the movement of the XY-stage member 32 by a pair of X-linear motors 43 that apply a moving force along the x-axis direction to the XY-stage member 32 and the X-linear motors 43. A pair of X-guide members 46.

  The magnets 44 of the pair of X-linear motors 43 are arranged on the coarse movement base member 13 so as to straddle the fine movement base member 17 disposed in the recess 19 of the coarse movement base member 13 in the x-axis direction and to be parallel to each other. It arrange | positions at the upper surface 13a. The coil 45 of the X-linear motor 43 is attached to the XY-stage member 32 and engages with each magnet 44 in a non-contact state. The pair of X-guide members 46 are disposed on the upper surface 13a of the coarse motion base member 13 so as to extend along the x-axis direction and be parallel to each other between the pair of magnets 44. .

  The connecting members 50 and 51 that connect the two X-guide members 46 and the two magnets 44 are installed on the upper surface 13 a of the coarse motion base member 13 via the air pad 52. Therefore, the magnet 44 and the X-guide member 46 are in a floating relationship on the coarse movement base member 13 and the fine movement base member 17, respectively.

  As described above, the X-interferometer 29 for measuring the position and the like of the lower table 35 in the x-axis direction is installed on the column member 23. Similarly, the position and the like of the lower table 35 in the y-axis direction are measured. The Y-interferometer 30 is also installed at a predetermined position on the column member 23. Specifically, on the surface of the pillar member 23 processed so as to be parallel to the xz plane, two locations separated in the x-axis direction and two locations spaced in the z-axis direction with respect to each other Four Y-interferometers 30 are installed for a total of four locations. From the outputs of these four interferometers, the distance of the lower table 35 in the y-axis direction, the rotation direction around the x-axis, and the rotation amount around the z-axis can be measured. The Y-interferometer 30 constitutes a position measurement unit that measures the position of the lower table 35. Therefore, the position measuring unit composed of the X-interferometer 29 and the Y-interferometer 30 includes the lower table 35 in the x-axis direction, the y-axis direction, the rotation direction around the x-axis, the rotation direction around the y-axis, and the z-axis. The position can be measured with respect to 5 degrees of freedom in the surrounding rotation direction.

  An X-moving mirror 74 and a Y-moving mirror 75 are attached to the corresponding lower table 35. The X-moving mirror 74 is a mirror for reflecting light from the X-interferometer 29, and the Y-moving mirror 75 is a mirror for reflecting light from the Y-interferometer 30. The mirror surface of the X-moving mirror 74 is a plane parallel to the yz plane, and the mirror surface of the Y-moving mirror 75 is a plane parallel to the xz plane.

  If the position measuring unit is configured in this way, the position of the lower table 35 can be grasped relatively accurately with reference to the upper table 24. When the second substrate 12 is superimposed on the first substrate 11 while being aligned, position measurement on the second substrate 12 side using the first substrate 11 side as a positioning reference is important. Satisfies such a relative relationship. In particular, since the body 21 has an integral frame structure with high rigidity, it is preferable to install the position measuring unit on the surface of the frame structure that faces the side surface of the lower table 35. In the present embodiment, the position measurement unit is configured using an interferometer. However, the position measurement unit is not limited thereto, and a position sensor or a distance sensor that realizes the same function, such as a laser length meter and a microsensor, is used. Also good.

  FIG. 3 is an explanatory diagram schematically showing the vicinity of the upper table 24. As described above, the upper table 24 is fixed to the upper plate member 22. On the lower surface 24 a of the upper table 24, a first substrate holder 25 that holds the first substrate 11 is placed. Specifically, a plurality of suction ports connected to the exhaust device are provided on the lower surface 24a, whereby the first substrate holder 25 is vacuum-sucked to realize the placement. The lower surface 24a serves as a reference surface in the alignment control described later.

  The first substrate 11 is attracted to the first substrate holder 25 by electrostatic force. The ESC terminal 27 is provided on the upper plate member 22, and when the first substrate holder 25 is placed on the upper table 24, the ESC terminal 27 and the terminal of the first substrate holder 25 are in contact with each other. ing. Then, after the first substrate holder 25 is placed on the upper table 24, electric power for generating an electrostatic force is supplied to the first substrate holder 25 via the ESC terminal 27.

  A plurality of load cells 26 for detecting a load acting on the upper table 24 are provided between the upper table 24 and the upper plate member 22. Specifically, at least three load cells that are not arranged in a straight line between the upper table 24 and the upper plate member 22 so that the magnitude and position of the load applied in the two-dimensional region of the lower surface 24a can be detected. 26 is installed. The outputs of these load cells 26 are input to a control unit as an analysis unit. The control unit analyzes from the output of each load cell 26 whether or not the second substrate 12 is in contact with the first substrate 11, and further reaches the pressure necessary for temporary bonding with each other. Analyzes whether or not excessive pressure is applied. At the same time, the control unit also analyzes whether the first substrate 11 or the first substrate holder 25 holding the first substrate 11 has contacted the upper table 24 from the output of each load cell 26, or Then, the downward force received by the upper table 24 by the first substrate 11 or the first substrate holder 25 being attracted to the lower surface 24a is analyzed to determine the mounting. As described above, by performing various analyzes using the output of the load cell 26, it is possible to simplify the apparatus as compared to performing detection and analysis by providing a single sensor. As a result, it contributes to downsizing and cost reduction of the apparatus. The sum of the pressure applied as temporary bonding and the load applied by the substrate holder is controlled to be smaller than the destructive force for destroying the element formed on any substrate.

  As an example of a specific load detected by the load cell 26, when the number of installed cells is three and the second substrate 12 is brought into contact with the first substrate 11, each load cell 26 is About 10N is detected as an upward force. When the first substrate holder 25 is mounted on the upper table 24, each load cell 26 detects about 5N as a downward force. Similarly, when only the first substrate 11 is mounted on the upper table 24, each load cell 26 detects about 0.5 N as a downward force.

  Whether the second substrate 12 is in contact with the first substrate 11, whether the pressure has reached a pressure necessary for temporary bonding with each other, or is an excessive pressure applied? And the determination that the first substrate 11 or the first substrate holder 25 is attracted to the upper table 24 can be executed even when only one load cell 26 is installed. . In addition to these analyzes and determinations, by arranging a plurality of load cells 26 in a plane, further analysis of the position of the second substrate 12 in contact with the first substrate 11 and pressure at the time of temporary bonding Analysis of distribution and analysis of which position on the upper table 24 the first substrate 11 or the first substrate holder 25 is in contact with can be performed. Therefore, for example, when the first substrate holder 25 is attracted to the upper table 24, it can be determined that the surfaces are in contact with each other without being inclined if the same output is detected from all the load cells 26. On the other hand, if different outputs are detected from the respective load cells 26, it can be determined that the first substrate holder 25 is tilted and contacted or attracted to the upper table 24.

  Moreover, the upper microscope 28 for observing the 2nd board | substrate 12 hold | maintained at the fine movement stage part 18 at the time of alignment is provided. A specific alignment operation will be described later.

  FIG. 4 is a perspective view of the coarse movement stage portion when the respective structures are specifically assembled. The same number is attached | subjected to the same structure and an example of the specific shape is shown.

  As shown in the drawing, the XY-stage member 32 is made of a plate-like member. A through hole 37 for allowing the self-weight canceller 36 to pass therethrough is formed at the center.

  FIG. 5 is a partial schematic cross-sectional view along BB in FIG. Specifically, it is a diagram schematically showing a cross section of the structure around the XY-stage member 32, the self-weight canceller 36, and the lower table 35 disposed above the fine movement base member 17.

  One X-guide member 46 of the pair of X-guide members 46 supports the XY-stage member 32 via an air pad 47 provided on the XY-stage member 32. An engaging portion 48 provided on the XY-stage member 32 is engaged with the other X-guide member 46 through an air bearing 49 in a non-contact state. When each X-linear motor 43 is driven, the engaging portion 48 is guided by the other X-guide member 46, whereby the XY-stage member 32 moves on the fine movement base member 17 along the x-axis direction.

  A plurality of air pads 38 for supporting the self-weight canceller 36 from the side thereof in a non-contact manner are provided on the inner peripheral portion of the through hole 37. Since each air pad 38 can cope with a slight inclination, even if the self-weight canceller 36 is inclined, it can be supported following the inclination within the allowable range.

  The self-weight canceller 36 is placed on the fine movement base member 17 via a plurality of air pads 70. As described above, the self-weight canceller 36 is supported on the XY-stage member 32 via the plurality of air pads 38. Then, when the XY-stage member 32 is driven on the xy plane by the X-drive unit 33 and the Y-drive unit 34, the self-weight canceller 36 follows the xy-stage member 17 and floats on the fine movement base member 17. Will move in the direction.

  The self-weight canceller 36 includes a cylinder 59, a piston 61, and a spherical seat 63. The cylinder 59 is made of a cylindrical member having a rectangular cross section, and has a pressure chamber 60 inside. The control unit controls the pressure adjusting device connected to the pressure chamber 60 to control the pressure in the pressure chamber 60 so as to balance the weight of the lower table 35 and the structure attached thereto. The air pad 38 supports the cylinder 59 in a non-contact manner. Therefore, the cylinder 59 is penetrated so as to open upward with respect to the through hole 37 of the XY-stage member 32. An arm member 89 is attached to the upper end of the cylinder 59, and a Z-interferometer 67 for detecting the position of the lower table 35 with respect to the self-weight canceller 36 is installed at the tip of the arm member.

  The piston 61 is inserted into the cylinder 59 through the air bearing 62 in a non-contact state, and can move along the z-axis direction according to the pressure in the pressure chamber 60. That is, the piston 61 is an expansion / contraction mechanism of the self-weight canceller 36 in the illustrated example. The upper part of the piston 61 is formed in a flat plate shape, and a spherical seat 63 is placed on the upper surface 61a via an air bearing 66. The spherical seat 63 is a pedestal that comes into contact with the spherical member 64 that is integrally attached to the lower surface 35 b of the lower table 35. The convex spherical surface 77 is a surface of the spherical member 64 that faces the concave spherical surface 65 of the spherical seat 63 via the air bearing 78. The piston 61 supports the spherical seat 63 and the spherical seat 63 supports the spherical member 64 in the z-axis direction in a non-contact state by interposing the air bearing 66 and the air bearing 78. Further, the lower table 35 can be tilted by guiding the spherical member 64 along the concave spherical surface 65 of the spherical seat 63.

  The lower table 35 is made of a plate member and is disposed above the XY-stage member 32. As described above, the second substrate holder 41 that holds the second substrate 12 is placed on the upper surface 35 a of the lower table 35. A plurality of suction ports connected to the exhaust device are provided on the upper surface 35a, whereby the second substrate holder 41 is vacuum-sucked to realize placement. The second substrate 12 is attracted to the second substrate holder 41 by electrostatic force. The ESC terminal 73 is provided on the lower table 35, and when the second substrate holder 41 is placed on the lower table 35, the ESC terminal 73 and the terminal of the second substrate holder 41 are in contact with each other. Yes. After the second substrate holder 41 is placed on the lower table 35, electric power for generating an electrostatic force is supplied to the second substrate holder 41 via the ESC terminal 73.

  The lower table 35 is formed with insertion holes 71 through which the plurality of push-up pins 42 are inserted. A plurality of insertion holes 72 are also formed in the second substrate holder 41 so as to correspond to a part of these insertion holes 71. With this configuration, the push-up pin 42 is extended so as to pass through the insertion hole 72 and the insertion hole 71 corresponding to the insertion hole 72 to lift the second substrate 12, and the second substrate holder 41 can be detached. Further, if the insertion hole 71 not corresponding to the insertion hole 72 is extended so as to pass the push-up pin 42, the second substrate holder 41 can be lifted and detached from the lower table 35. The plurality of push-up pins 42 have their main bodies fixed on the XY-stage member 32, and the extension and storage of the pins are controlled by commands from the control unit.

  The XY-stage member 32 includes an XY-eddy current sensor 39 that measures a relative distance between the lower table 35 and the XY-stage member 32 along the x-axis and y-axis directions, and a relative position along the z-axis direction. A Z-eddy current sensor 40 for measuring the distance is provided. These form a measurement unit that detects relative positions of the XY-stage member 32 and the lower table 35 that are driven together on the xy plane by the X-drive unit 33 and the Y-drive unit 34.

  Between the XY-stage member 32 and the lower table 35, three Z-VCMs 86 as voice coil motors as actuators are installed so that a driving force acts in the z-axis direction. Magnets 87 of the respective Z-VCMs 86 are fixed to the lower table 35, and coils 88 that engage with the magnets 87 are fixed to the XY-stage member 32, respectively. The control unit individually controls these three Z-VCMs 86, whereby the lower table 35 can be translated in the z-axis direction or rotated about the x-axis and the y-axis. Specific operations will be described later.

  In the lower table 35, an AF sensor 79 and a lower microscope 76 are installed around the mounting surface of the second substrate holder 41. The AF sensor 79 is a sensor that detects the relative inclination of the upper table 24 and the lower table 35. The lower microscope 76 is a microscope for observing the first substrate 11 held on the upper table 24 during alignment.

  FIG. 6 is a perspective view of the self-weight canceller 36 when each structure is specifically assembled. The same number is attached | subjected to the same structure as the self-weight canceller 36 shown in FIG. 5, and an example of the specific shape is shown. A difference from the self-weight canceller 36 shown in FIG. 5 is that a leg member 68 is provided between the cylinder 59 and the air pad 70. The leg member 68 extends from the lower part of the cylinder 59 toward the fine movement base member 17.

  FIG. 7 is a schematic cross-sectional view taken along the line CC of FIG. By passing screws through the illustrated screw holes, the cylinder 59 and the leg member 68, the top plate portion having the upper surface 61a of the piston 61 and the piston portion, the lower table 35 and the spherical member 64 are integrated. Air bearings 62, 66, and 78 are provided between the cylinder 59 and the piston 61, between the piston 61 and the spherical seat 63, and between the spherical seat 63 and the spherical member 64, respectively, and are not in contact with each other. It is kept.

  8 is a cross-sectional view taken along the line BB in FIG. 2, and is a partial detail view seen from the opposite direction to FIG. Between the XY-stage member 32 and the lower table 35, two Y-VCMs 83 serving as voice coil motors as actuators are installed so that a driving force acts in the y-axis direction. The magnets 84 of the respective Y-VCMs 83 are fixed to the lower table 35, and the coils 85 that engage with the magnets 84 are fixed to the XY-stage member 32, respectively. When the control unit individually controls these two Y-VCMs 83, the lower table 35 can be translated in the y-axis direction or rotated around the z-axis.

  FIG. 9 is a partial detailed cross-sectional view along the line AA of FIG. Between the XY-stage member 32 and the lower table 35, one X-VCM 80, which is a voice coil motor as an actuator, is installed so that a driving force acts in the x-axis direction. The magnets 81 of the respective X-VCMs 80 are fixed to the lower table 35, and the coils 82 that engage with the magnets 81 are fixed to the XY-stage member 32, respectively. As the control unit controls the X-VCM 80, the lower table 35 can be translated in the x-axis direction.

  When the driving force of the X-VCM 80 and Y-VCM 83 is applied to move the lower table 35 in the x-axis direction, the y-axis direction, or the rotation direction around the z-axis, the lower table 35 is moved through the spherical member 64 and the air bearing 78. The spherical seat 63 moves on the upper surface 61a of the piston 61 in the xy plane by the force received in this manner. At this time, since the air bearing 66 is formed between the spherical seat 63 and the piston 61, the lower table 35 is guided on the upper surface 61a of the piston 61 in the x-axis direction and the y-axis direction.

  Further, when the driving force of the Z-VCM 86 is applied to move the lower table 35 in the rotational direction around the x axis or the rotational direction around the y axis, the spherical member 64 is provided with the concave spherical surface of the spherical seat 63 via the air bearing 78. 65, so that the lower table 35 tilts about the x axis or the y axis. Furthermore, when the lower table 35 is moved in the z-axis direction by applying the driving force of the Z-VCM 86, the piston 61 of the self-weight canceller 36 moves in the z-axis direction according to the movement amount of the Z-VCM 86. The lower table 35 and the weight of the structure integrally attached thereto are supported. Therefore, the lower table 35 can be moved in the z-axis direction even if the driving force of the Z-VCM 86 is not so large. An air pad 70 for forming an air bearing 69 is provided between the lower surface of the leg member 68 of the self-weight canceller 36 and the fine movement base member 17. When the control unit rotates and tilts the lower table 35 around the x-axis or the y-axis, the rotation axis passes through the center of gravity of the surface of the second substrate 12 that faces the first substrate 11. Drive the Z-VCM86.

  FIG. 10 is a top view schematically showing the fine movement stage portion. In particular, here, the arrangement and operation of actuators that drive the lower table 35 will be described.

  As described above, a structure such as the lower microscope 76 is attached to the lower table 35. Further, although the mass of the lower table 35 and the structure attached to the lower table 35 is supported by the self-weight canceller 36, it is spatially isolated from other structures by the action of the air bearings 66 and 78. Can be seen as a unit. At this time, the center of gravity of the unit comprising the lower table 35 and the entire structure attached to the lower table 35 is defined as the center of gravity G. One X-VCM 80 is installed so that the driving force acts in the x-axis direction, two Y-VCMs 83 are installed so that the driving force acts in the y-axis direction, and Z-VCM 86 is Three are provided so that a driving force acts in the z-axis direction. As a specific installation position, the X-VCM 80 is installed so that the driving force in the x-axis direction passes through the center of gravity G. The Y-VCM 83 is installed so that no moment is generated around the z-axis when the same driving force is output in the y-axis direction. For example, if it is assumed that there is no mass distribution bias in the unit, the units may be installed so that the respective driving forces are output in two linear directions parallel to the y axis passing through the center of gravity G and symmetric. The Z-VCM 86 is installed so that no moment is generated around the x-axis or the y-axis when the same driving force is output in the z-axis direction. For example, if it is assumed that the unit has no mass distribution bias, each driving force is output in three linear directions that pass through the apex of an equilateral triangle parallel to the z axis and centered on the center of gravity G on the xy plane. Can be installed as required. The control unit drives the X-VCM 80 and the Y-VCM 83 so as to pass through the center of gravity G when the lower table 35 is rotated around the z axis.

  As described above, according to the present embodiment, each actuator directly applies the driving force to the lower table 35, whereby the three degrees of freedom of translation in the x-axis direction, the y-axis direction, and the z-axis direction, and x The lower table 35 can be driven with a total of 6 degrees of freedom of rotation around the axis, around the y axis and around the z axis. The control unit independently drives each actuator to control the posture of the lower table 35 with six degrees of freedom, and the position of the mounted second second substrate 12 is a stationary reference plane. The lower surface 24a of the upper table 24 is controlled. In particular, by directly applying the driving force of the actuator to the lower table 35, it is possible to improve the responsiveness that is the time taken to reach the target position in response to the driving force output instruction. As a result, the alignment apparatus that aligns the substrates on the submicron order while moving the lower table 35 with respect to the upper table 24a reduces the time required for alignment and improves the alignment accuracy. In the present embodiment, it has been described that each independently driven actuator is a single actuator, but two or more actuators may function as a single actuator. In this case, since the output can be shared by two or more actuators, an actuator having only a small output with respect to the required output can be used. The control unit controls these two or more actuators so as to function as one actuator.

  Installing the actuator as described above with respect to the center of gravity G of the unit is preferable not only for driving the lower table 35 with a small driving force but also from the viewpoint of responsiveness. However, the installation of the actuator is not limited to this position, and the number of actuators is not limited to this. It only needs to be installed so that the control unit can control the posture of the lower table 35 with the above-described degree of freedom. In that sense, care should be taken that actuators that output a driving force in the z-axis direction are not arranged in a straight line. In this embodiment, the voice coil motor is used as the actuator, but the present invention is not limited to this. Any actuator that can directly apply a driving force to the lower table 35 may be used. For example, even if the driving output is a rotational force, such as a rotary motor, an actuator that directly applies the driving force to the lower table 35 by incorporating a conversion unit that converts it into a linear output may be realized. Note that a linear actuator such as a voice coil motor that directly outputs a driving force in a linear direction is preferable because it has excellent responsiveness and controllability to the lower table 35.

  Note that the drive amounts of the X-drive unit 33 and the Y-drive unit 34 in the x-axis direction and the y-axis direction are larger than the drive amounts of the X-VCM 80 and the Y-VCM 83 in the x-axis direction and the y-axis direction. Therefore, when the XY plane member 32 is moved largely in the xy plane, it is preferable to move the XY-stage member 32 integrally with the lower table 35 by the X-drive unit 33 and the Y-drive unit 34. It is preferable to move only the lower table 35 by the VCM 80 and the Y-VCM 83.

  FIG. 11 is a schematic cross-sectional view showing another configuration of the overlaying apparatus. Specifically, the positions of the cylinder 59 and the piston 61 of the self-weight canceller 36 are reversed in the above-described superimposing apparatus. That is, the cylinder 59 passes through the through hole 37 of the XY-stage member 32 so as to open downward. The same structure is given the same number.

  The spherical seat 63 is supported on the upper surface 59 b of the cylinder 59 through the air bearing 66 in a non-contact state. The piston 61 is inserted into the cylinder 59 via the air bearing 62 from the lower side in a non-contact state. An arm member 91 that extends through a through-hole 90 formed in the XY-stage member 32 is attached to the lower end portion 61 b of the piston 61. A Z-interferometer 67 is installed at the tip of the arm member 91. Further, since the air bearing 66 is formed between the spherical seat 63 and the cylinder 59, when the lower table 35 moves in the x-axis direction and the y-axis direction by driving the X-VCM 80 and the Y-VCM 83, Due to the force received from the table 35 via the spherical member 64 and the air bearing 78, the spherical seat 63 moves on the upper surface 59b of the cylinder 59 within the xy plane. Accordingly, the lower table 35 is guided on the upper surface 59b of the cylinder 59 in the x-axis direction and the y-axis direction.

  FIG. 12 is a perspective view showing a state in which the second substrate holder 41 is looked down from above. In the figure, the second substrate 12 is held on the upper surface of the second substrate holder 41. FIG. 13 is a perspective view showing a state where the same second substrate holder 41 is looked up from below.

  The second substrate holder 41 includes a holder main body 110, an adsorber 111, and a voltage application terminal 113, and has a disk shape whose diameter is slightly larger than that of the second substrate 12 as a whole. The holder body 110 is integrally formed of a highly rigid material such as ceramic or metal. 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 second substrate 12 on the outer surface of the second substrate 12. In the case of the figure, a total of six adsorbers 111 are arranged every 120 degrees with two as one set. The voltage application terminal 113 is embedded in the back surface of the surface holding the second substrate 12.

  The holder main body 110 is in contact with the second substrate 12 to be held with a high flatness in the region where the second substrate 12 is held on the surface thereof. In addition, the holder body 110 has a plurality of positioning holes 112 that are formed so as to penetrate the holder body 110 on the front and back sides outside the region where the held second substrate 12 is sucked. Furthermore, the holder main body 110 has a plurality of insertion holes 72 formed through the holder main body 110 on the front and back sides inside the region for adsorbing the held second substrate 12. The push-up pin 42 is inserted into the insertion hole 72, and the second substrate 12 is removed from the second substrate holder 41.

  The positioning hole 112 is fitted to a positioning pin or the like to give positioning to the second substrate holder 41. The adsorber 111 is disposed in a depressed region formed in the holder main body 110 so that the upper surface is located in a plane substantially the same as the plane that holds the second substrate 12. The voltage application terminal 113 is embedded in the holder main body 110 on the back surface with respect to the front surface holding the second substrate 12. By applying a voltage via the voltage application terminal 113, a potential difference is generated between the second substrate holder 41 and the second substrate 12, and the second substrate 12 is attracted to the second substrate holder 41. To do. When the second substrate holder 41 is placed on the lower table 35, the voltage application terminal 113 is in contact with the ESC terminal 73 and is supplied with power.

  The first substrate holder 25 has substantially the same configuration as the second substrate holder 41. When the first substrate holder 25 is placed on the upper table 24, the voltage application terminal 113 is in contact with the ESC terminal 27 and receives power.

  FIG. 14 is a control flow diagram for explaining the alignment process in the superposition apparatus 10. In particular, a process until the first substrate 11 and the second substrate 12 are in surface contact with each other is shown. Each step receives an output from each sensor or a control unit controls each driving unit.

  In step S <b> 101, the first substrate holder 25 is placed on the upper table 24 and the second substrate holder 41 is placed on the lower table 35. The first substrate 11 is placed on the first substrate holder 25 and electrostatically adsorbed in the pre-aligner device before being carried into the stacking apparatus 10. Therefore, the first substrate 11 and the first substrate holder 25 are integrated at the stage of being carried into the overlapping apparatus 10. The pre-aligner apparatus further detects a plurality of alignment marks provided on the first substrate 11 when the first substrate 11 is placed on the first substrate holder 25, and this information is connected to the control unit. Stored in the storage unit. Similarly, the second substrate 12 is also integrated with the second substrate holder 41 in the pre-aligner apparatus. At the same time, when the second substrate 12 is placed on the second substrate holder 41, a plurality of alignment marks provided on the second substrate 12 are detected, and this information is stored in the memory connected to the control unit. Store in the department.

  The integrated first substrate 11 and first substrate holder 25, and second substrate 12 and second substrate holder 41 are carried into the stacking apparatus 10, respectively. Carrying into the superposition apparatus 10 is performed by a robot arm. A power supply terminal that supplies power to the substrate holder when the substrate holder is held is provided in the hand unit that holds the substrate holder, which is the tip of the robot arm. Thereby, the substrate holder can maintain the electrostatic attraction force, and the substrate can be fixed until it is placed on the upper table 24 or the lower table 35.

  In particular, when the first substrate holder 25 is placed on the upper table 24, the control unit determines the contact and placement using the output of the load cell 26. Thus, for example, when it is determined that the first substrate holder 25 is in point contact with the upper table 24, the control unit corrects the inclination of the hand portion of the robot arm so that the first substrate holder 25 and the upper substrate 24 The table 24 can be controlled to come into surface contact. Further, by observing the output of the load cell 26 after the upper table 24 sucks the first substrate holder 25, it can be determined whether the first substrate holder 25 is normally mounted.

  In step S102, the relative positions of the XY-stage member 32 and the lower table 35 are detected using the XY-eddy current sensor 39 and the Z-eddy current sensor 40. Information on the detected relative position is stored in a storage unit connected to the control unit. Thereby, even if the XY-stage member 32 is coarsely moved by the X-drive unit 33 and the Y-drive unit 34, the position of the lower table 35 can be calculated according to the amount of movement. At this time, the inclination of the upper table 24 relative to the lower table 35 is also detected using the AF sensor 79.

  In step S103, a target position of the lower table 35 with respect to the lower surface 24a of the upper table 24, which is a reference surface, and a drive path to the target position are determined based on each position information stored in the storage unit. As for the target position, for example, when the plurality of alignment marks on the first substrate 11 and the plurality of alignment marks on the second substrate 12 are overlapped, the amount of positional deviation is statistically minimized. It is calculated and determined by using the determined global alignment method or the like. Regarding the drive path, fine adjustment by the X-VCM 80, Y-VCM 83, and Z-VCM 86 is performed after the movement in the xy direction by the X-drive unit 33 and the Y-drive unit 34 and the movement in the z direction by the self-weight canceller 36. Is determined by a combination of driving for.

  When the target position and the drive path are determined in step S103, the X-drive unit 33 and the Y-drive unit 34 move in the xy direction and the Z-VCM 86 moves in the z direction. The lower table 35 is moved until. In step S104, the X-interferometer 29 and the Y-interferometer 30 detect the accurate positions of the lower table 35 in the x-axis direction and the y-axis direction. Then, fine adjustment is performed by the X-VCM 80 and the Y-VCM 83 in step S105 while calculating the difference between the target position and the current position.

  In step S106, the Z-interferometer 67 detects the exact position of the lower table 35 in the z-axis direction. Then, while calculating the difference between the target position and the current position, the lower table 35 is gradually advanced toward the upper table 24 by the Z-VCM 86 in step S107. At this time, the control unit adjusts the pressure in the pressure chamber 60 of the self-weight canceller 36 in accordance with the movement of the lower table 35 driven by the Z-VCM 86. That is, the self-weight canceller 36 is feedback-controlled so as to support the self-weight of the unit including the lower table 35.

  In step S108, the output of the load cell 26 is monitored, and the time point when the lower table 35 approaches the upper table 24 and the second substrate 12 first contacts the first substrate 11 is detected. While detection by the load cell 26 is not performed, step S104 to step S107 are repeated.

  In step S109, the inclination of the lower table 35 is detected. A plurality of load cells 26 are provided between the upper table 24 and the upper plate member 22, and the contact state of the second substrate 12 with respect to the first substrate 11 can be detected as a pressure distribution. When the lower table 35 is tilted, the second substrate 12 comes into point contact with the first substrate 11, so that the load concentrates on a specific region. The load cell 26 outputs this, and the control unit detects the inclination of the lower table.

  If it is determined in step S109 that the lower table 35 has an inclination, the process proceeds to step S110, and the inclination of the lower table 35 is corrected by the Z-VCM 86 using the output of the load cell 26. When the correction of the inclination is completed, the process proceeds to step S109 again.

  Here, the control in the z-axis direction by the self-weight canceller 36 and the Z-VCM 86 will be further described. As described above, the self-weight canceller 36 is feedback-controlled so as to support the self-weight of the unit including the lower table 35. Specifically, the pressure of the pressure chamber 60 corresponding to the support weight is set in advance, and the pressure is controlled so that the pressure is kept constant even when the lower table 35 is driven by the Z-VCM 86. . For example, the weight of the lower substrate 35 and the structure integrally attached thereto, the weight of about 15N of the second substrate holder 41 to be placed, and the weight of the second substrate, which is a single wafer, are reduced. By adding 1.5 N, the pressure in the pressure chamber 60 is controlled to balance these weights.

  The self-weight canceller 36 is controlled by the pressure chamber 60 by extracting and injecting the fluid in the pressure chamber 60 via a pressure adjusting device connected to the pressure chamber 60. Therefore, the response speed is generally slow compared to the response speed of Z-VCM86. Therefore, the response to the disturbance is first adjusted to be performed by the position control by the Z-VCM 86. That is, when any sudden change occurs, feedback control is performed so as to drive the Z-VCM 86 and mitigate the change.

  Although details will be described later, when the first substrate 11 and the second substrate 12 come into surface contact, provisional bonding is performed thereafter, the first substrate holder 25 and the second substrate holder 41 are clamped, and the upper table 24 is clamped. Releases the suction of the first substrate holder 25. Then, a weight of about 16.5 N of the first substrate 11 and the first substrate holder 25 is added to the self-weight canceller 36. At this time, if no countermeasure is taken, the weight addition of the first substrate 11 and the first substrate holder 25 serves as a step input, and the response generally involves an overshoot. Then, the lower table 35 on which the clamped first substrate holder 25 and the second substrate holder 41 are placed is once separated from the upper table 24 and then again bounced by driving of the Z-VCM 86. It is pulled back to the 24 side. Then, the target position, which is the initial position, is exceeded, and the placed first substrate holder 25 and second substrate holder 41 collide with the upper table 24. Such a phenomenon not only causes physical damage to the first substrate 11 and the second substrate 12 that are temporarily bonded, but also induces misalignment to cause an electrical failure. Become.

  In order to avoid such a phenomenon, the present embodiment can take two control flows. FIG. 15 is a control flowchart for explaining the alignment process of the superposition apparatus 10 according to the first control subsequent to the alignment process of FIG. Each step receives an output from each sensor or a control unit controls each driving unit.

  When the first substrate 11 and the second substrate 12 are in surface contact in step S109, the process proceeds to step S211. In step S211, the Z-VCM 86 is driven so that a pressing force corresponding to the weight of the first substrate 11 and the first substrate holder 25 is added at this time. In step S212, the self-weight canceller 36 pressurizes the lower table 35 toward the upper table 24 in order to temporarily join the first substrate 11 and the second substrate 12. The main joining may be performed by a heating / pressurizing device that is an external device of the overlaying apparatus 10 or may be configured to be performed in the overlaying apparatus 10.

  When temporary bonding is completed in step S212, the adsorbent 111 of the first substrate holder 25 and the adsorbent 111 of the second substrate holder 41 are acted on in step S213 to clamp the two substrate holders together. That is, the first substrate holder 25 and the second substrate holder 41 sandwich the first substrate 11 and the second substrate 12 that are accurately aligned to create a state in which they are integrally fixed.

  In step S214, the upper table 24 releases the state where the first substrate holder 25 is sucked by vacuum suction, and the support of the first substrate holder 25 and the first substrate 11 by the upper table 24 is finished. To do. At this time, the first substrate holder 25 and the second substrate holder 41 which are in a pressurized state and clamped more than support the own weight by the action of the own weight canceller 36 are sandwiched between the upper table 24 and the lower table 35. Yes.

  The pressure is gradually released from this state, and the lower table 35 is clamped and integrated with all of the first substrate 11, the second substrate 12, the first substrate holder 25, and the second substrate holder 41. Transition to a state that supports its own weight. Step S215 is this process, and at the same time, the pressing force of the Z-VCM 86 added in step S211 is reduced. In other words, the state of step S214 is such that the Z-VCM 86 exerts a pressing force that supports the weight of the first substrate 11 and the first substrate holder 25, and the weight canceller 36 supports the pressing force that supports other weights. When the welding pressure is exerted, the support of the Z-VCM 86 is gradually reduced to 0, and all weights are shifted to a state where the weight canceller 36 supports them. In other words, the driving force for supporting the weight of the first substrate 11 and the first substrate holder 25 is shifted so that the self-weight canceller 36 is generated. Since the applied pressure is also reduced at the same time, the acting force acting between the upper table 24 and the first substrate holder 25 eventually becomes zero.

  When the acting force between the upper table 24 and the first substrate holder 25 becomes zero, and the self-weight canceller 36 is further driven to lower the lower table 35, the upper table 24 and the first substrate holder 25 are separated (step). S216). In this way, the pressing force of the self-weight canceller 36 acts on the upper table 24 side, and the load cell 26 detects this until the upper table 24 and the first substrate holder 25 are separated from each other. If a driving force having a magnitude corresponding to the load of the substrate 11 and the first substrate holder 25 is controlled to act from the lower table 35 side to the upper table 24 side, the first substrate 11 and the first substrate abruptly are separated at the time of separation. The weight of one substrate holder 25 is not added to the lower table 35, and inadvertent driving of the Z-VCM 86 having high response sensitivity can be prevented. Therefore, even if the above flow is processed at a relatively high speed, an inadvertent collision due to overshoot can be avoided.

  In step S217, the control unit lowers the lower table 35 to a predetermined position by the self weight canceller 36. Specifically, the predetermined position is a position where the hand portion of the robot arm can be inserted so that the first substrate holder 25 and the second substrate holder 41 can be carried out in a clamped state. Then, in step S218, the push-up pins 42 are operated to make them slightly lifted from the lower table 35, and are carried out of the stacking apparatus 10 by the robot arm. As described above, a series of positioning steps is completed.

  FIG. 16 is a control flowchart for explaining the alignment process of the superposition apparatus 10 according to the second control following the alignment process of FIG. Each step receives an output from each sensor or a control unit controls each driving unit.

  When the first substrate 11 and the second substrate 12 come into surface contact in step S109, the process proceeds to step S311. In step S <b> 311, the lower table 35 is pressurized toward the upper table 24 by the self-weight canceller 36 in order to temporarily bond the first substrate 11 and the second substrate 12.

  When the temporary bonding is completed in step S311, the adsorbent 111 of the first substrate holder 25 and the adsorbent 111 of the second substrate holder 41 are acted on to clamp the two substrate holders in step S312. That is, the first substrate holder 25 and the second substrate holder 41 sandwich the first substrate 11 and the second substrate 12 that are accurately aligned to create a state in which they are integrally fixed.

  In step S313, the upper table 24 releases the state where the first substrate holder 25 is sucked by vacuum suction, and the support of the first substrate holder 25 and the first substrate 11 by the upper table 24 is finished. To do. At this time, the first weight holder 25 and the second weight holder 41 are clamped between the upper table 24 and the lower table 35, because the pressure is higher than that of supporting the dead weight by the action of the dead weight canceller 36. ing.

  The pressure is gradually released from this state, and the lower table 35 is clamped and integrated with all of the first substrate 11, the second substrate 12, the first substrate holder 25, and the second substrate holder 41. The process proceeds to a state in which its own weight is supported (step S314). That is, in the state of step S313, the self weight canceller 36 exerts the pressing force for supporting all the self weights including the lower table 35 and the pressure of temporary bonding. In step S314, the pressure of the temporary bonding is gradually increased. The acting force acting between the upper table 24 and the first substrate holder 25 is set to zero. Note that the Z-VCM 86 at this time is not responsible for either the applied pressure or the weight support.

  When the acting force between the upper table 24 and the first substrate holder 25 becomes zero, and the self-weight canceller 36 is further driven to lower the lower table 35, the upper table 24 and the first substrate holder 25 are separated (step). S315). By controlling in this way, the weight of the first substrate 11 and the first substrate holder 25 is not suddenly added to the lower table 35 at the time of the separation. In addition, by performing a series of processes relatively slowly, it is possible to prevent inadvertent driving of the Z-VCM 86 with high response sensitivity. Therefore, inadvertent collision due to overshoot can be avoided.

  In step S316, the control unit lowers the lower table 35 to a predetermined position by the self-weight canceller 36. Specifically, the predetermined position is a position where the hand portion of the robot arm can be inserted so that the first substrate holder 25 and the second substrate holder 41 can be carried out in a clamped state. In step S317, the push-up pins 42 are operated to slightly lift them from the lower table 35, and are carried out of the stacking apparatus 10 by the robot arm. As described above, a series of positioning steps is completed.

  Next, modified examples of the above embodiment will be described focusing on different parts. FIG. 17 is an explanatory diagram schematically showing the vicinity of the upper table 24 according to the first modification. On the lower surface 24 a of the upper table 24, a first substrate holder 25 that holds the first substrate 11 is placed and held by the attachment / detachment mechanism 121. Specifically, the attachment / detachment mechanism 121 includes a pipe 122 connected to the exhaust device, a valve 123 installed in the pipe, and one end connected to a plurality of openings provided on the lower surface 24a of the upper table 24 and the other end connected to the pipe. Constituted by a conduit 124 connected to 122. The control unit controls the exhaust device and the valve 123 to form an intake state, and vacuum-sucks the first substrate holder 25 and holds it on the upper table 24. When controlling the exhaust device in the intake state, the attachment / detachment mechanism 121 functions as a holding mechanism. Further, the attachment / detachment mechanism 121 also functions as a squirting mechanism when the control unit switches the exhaust device from the intake state to the squirt state. At this time, a plurality of openings provided in the lower surface 24a play a role of the fumarole. Air, nitrogen, inert gas, or the like is used as the gas to be blown. Of course, the holding mechanism and the blow mechanism can be configured separately. The lower surface 24a serves as a reference surface in the alignment control described later.

  In the first modification, following the control flow described in FIG. 14, a control flow different from that of the above embodiment is executed. FIG. 18 is a control flow diagram of the overlaying apparatus according to the first modification, which is executed following the control flow described in FIG. Each step receives an output from each sensor or a control unit controls each driving unit.

  When the first substrate 11 and the second substrate 12 come into surface contact in step S109, the process proceeds to step S111. In step S <b> 111, the lower table 35 is pressurized toward the upper table 24 by the self-weight canceller 36 in order to temporarily bond the first substrate 11 and the second substrate 12. The main joining may be performed by a heating / pressurizing device that is an external device of the overlaying apparatus 10 or may be configured to be performed in the overlaying apparatus 10.

  When temporary bonding is completed in step S111, the adsorbent 111 of the first substrate holder 25 and the adsorbent 111 of the second substrate holder 41 are acted on in step S112 to clamp the two substrate holders together. That is, the first substrate holder 25 and the second substrate holder 41 sandwich the first substrate 11 and the second substrate 12 that are accurately aligned to create a state in which they are integrally fixed.

  In step S <b> 113, the upper table 24 releases the state where the first substrate holder 25 is sucked by the vacuum suction of the attaching / detaching mechanism 121, and the first substrate holder 25 and the first substrate 11 by the upper table 24 are released. End of support. And a control part switches to fumarole control simultaneously with complete | finishing adsorption | suction control of the attachment / detachment mechanism 121. FIG. Furthermore, the control unit starts control to lower the lower table 35. At this time, the weights of the first substrate 11 and the first substrate holder 25 are added to the self-weight canceller 36 as a step input. However, even if the lower table 35 bounces, compressed air remains from the upper table 24. It is ejected, and this forms an air layer and exhibits a buffering function, so that the first substrate holder 25 placed on the lower table 35 is prevented from colliding with the upper table 24.

  And the attachment / detachment mechanism 121 continues to blow for a predetermined time until the bounce operation is settled. Control may be performed such that strong squirt occurs while the distance between the upper table 24 and the lower table 35 is short, and weak squirt as the distance increases.

  Note that the feedback gain of the feedback control in which the Z-VCM 86 tries to maintain the same position may be changed to a predetermined value smaller than the original value at the same time as switching to the fumarole control. By changing in this way, the response sensitivity of the Z-VCM 86 can be lowered, and the operation to maintain the current state can be slowed down. That is, while the self-weight canceller 36 attempts to lower the lower table 35, the Z-VCM 86 is prevented from operating to lift the lower table 35.

  In step S <b> 114, the control unit lowers the lower table 35 to a predetermined position by the self-weight canceller 36. Specifically, the predetermined position is a position where the hand portion of the robot arm can be inserted so that the first substrate holder 25 and the second substrate holder 41 can be carried out in a clamped state. In step S115, the push-up pins 42 are operated so that they are slightly lifted from the lower table 35, and are carried out of the stacking apparatus 10 by the robot arm. As described above, a series of positioning steps is completed.

  Furthermore, a second modification will be described focusing on the differences from the above embodiment. FIG. 19 is a schematic cross-sectional view along a cross-sectional line corresponding to CC in FIG. 6 in the self-weight canceller 36 according to the second modification. Unlike the self-weight canceller 36 in the above embodiment, a shock absorber 92 is provided inside the pressure chamber 60 and regulates the lower limit end of the piston 61. That is, the range in which the piston 61 can be moved up and down is set in advance, but the shock absorber 92 is provided as a buffer member so that the piston 61 does not collide with the bottom surface of the cylinder 59 even if the piston 61 is rapidly lowered. The shock absorber 92 is covered with a material that is not eroded by the fluid filled in the pressure chamber 60, and is formed by combining, for example, a spring and a damper.

  In the second modified example, following the control flow described in FIG. 14, a control flow different from the above embodiment is executed. FIG. 20 is a control flow diagram of the superposing apparatus according to the second modification, which is executed following the control flow described in FIG. Each step receives an output from each sensor or a control unit controls each driving unit.

  When the first substrate 11 and the second substrate 12 come into surface contact in step S109, the process proceeds to step S411. In step S411, the lower table 35 is pressurized toward the upper table 24 by the self-weight canceller 36 in order to temporarily bond the first substrate 11 and the second substrate 12 together. The main joining may be performed by a heating / pressurizing device that is an external device of the overlaying apparatus 10 or may be configured to be performed in the overlaying apparatus 10.

  When temporary bonding is completed in step S411, the adsorbent 111 of the first substrate holder 25 and the adsorbent 111 of the second substrate holder 41 are acted on to clamp the two substrate holders together in step S412. That is, the first substrate holder 25 and the second substrate holder 41 sandwich the first substrate 11 and the second substrate 12 that are accurately aligned to create a state in which they are integrally fixed.

  In step S413, the upper table 24 releases the state where the first substrate holder 25 is sucked by vacuum suction, and ends the support of the first substrate holder 25 and the first substrate 11 by the upper table 24. To do. At this time, the first weight holder 25 and the second weight holder 41 are clamped between the upper table 24 and the lower table 35, because the pressure is higher than that of supporting the dead weight by the action of the dead weight canceller 36. ing.

In synchronization with the operation of releasing the suction state of the upper table 24 in step S413, in step S414, the control unit performs control to instantaneously lower the lower table 35 by d ₀ mm or more. At the same time, in step S415, the feedback gain of feedback control in which the Z-VCM 86 tries to maintain the same position is changed to a predetermined value smaller than the original value. That is, step S414 and step S415 are steps of pulling the lower table 35 away from the upper table 24 from the state where the lower table 35 pressurizes the upper table 24. Synchronizing with the operation in which the upper table 24 releases the suction state does not necessarily coincide with the time if the lower table 35 is in a state of applying pressure to the upper table 24, and the suction state is released. It means that while maintaining the pressurized state.

  However, when temporary clamping is not performed as in the present processing flow but only alignment is performed and clamping is performed, the pressurized state may not be passed. In the state where the lower table 35 holds and balances the second substrate holder 41 on which the second substrate 12 is placed, not in the pressurized state, the first substrate holder 25 on which the first substrate 11 is placed. When the suction is released, the self-weights of the first substrate 11 and the first substrate holder 25 are added to the lower table 35 at this moment, so that synchronization requires time coincidence.

In the above description, the predetermined distance d ₀ mm is set in advance to a value equal to or greater than the distance corresponding to the overshoot that can occur when the weight addition of the first substrate 11 and the first substrate holder 25 acts as a step input. The The control of the self-weight canceller 36 that lowers the lower table 35 instantaneously by d ₀ mm or more can take the following two. First, the amount of subtraction obtained by subtracting the amount of fluid in the pressure chamber 60 that realizes the state after the reduction from the amount of fluid in the pressure chamber 60 that realizes the state before the reduction is controlled by the pressure regulator. This is a method of extracting from the pressure chamber 60. Here, the state after the lowering is a state in which the clamped first substrate holder 25 and the second substrate holder 41 are placed on the lower table 35 and are balanced below d ₀ mm compared to before the lowering. . In this case, the predetermined distance d ₀ mm itself is not meaningful, but it is significant that the distance is equal to or more than the distance corresponding to the overshoot. Therefore, the distance corresponding to the overshoot is not set, but the distance corresponding to the overshoot is lowered. A sufficient amount of subtraction may be set by volume or mass.

The second is a method in which the self-weight canceller 36 that is normally pressure-controlled is temporarily switched to position control and monitored until the lower table 35 is pulled down by d ₀ mm or more. As described above, the accurate position of the lower table 35 in the z-axis direction is detected by the Z-interferometer 67. Therefore, until the value lower than d ₀ mm is obtained by the output of the Z-interferometer 67, the pressure regulator is controlled to extract the fluid from the pressure chamber 60.

  The reason why the feedback gain is changed to a predetermined value smaller than the original value in step S415 is that it is desired to lower the response sensitivity of the Z-VCM 86 and to slow down the operation for maintaining the current state. That is, while the self-weight canceller 36 attempts to lower the lower table 35, the Z-VCM 86 is prevented from operating to lift the lower table 35. Even if such an operation is started, if the feedback gain is small, the speed at which the Z-VCM 86 attempts to lift the lower table 35 is slower than the speed at which the self-weight canceller 36 attempts to lower the lower table 35. Therefore, a situation where the lower table 35 collides with the upper table 24 can be avoided.

In addition, the operation | movement which cancels | releases the adsorption | suction state of the upper table 24 of step S413 detects the timing with the output of the pressure gauge provided in the exhaust pipe connected with the upper table 24. FIG. The pressure gauge is preferably provided as close to the upper table 24 as possible. Of course, the predetermined distance d ₀ mm is smaller than the distance to the receiving surface of the shock absorber 92.

In step S416, the control unit further lowers the lower table 35 to a predetermined position by the self-weight canceller 36. Specifically, the predetermined position is a predetermined distance d ₀ mm or more, and the hand portion of the robot arm is inserted so that the first substrate holder 25 and the second substrate holder 41 can be carried out in a clamped state. It is a position that can be done. Then, in step S417, the push-up pins 42 are operated to make them slightly lifted from the lower table 35, and are carried out of the stacking apparatus 10 by the robot arm. As described above, a series of positioning steps is completed.

In the above-described embodiment, the lower table 35 is controlled to be instantaneously pulled down by d ₀ mm or more by the self-weight canceller 36. However, since the tables need only be relatively spaced apart by d ₀ mm or more, the upper table 24 may be controlled to be lifted by d ₀ mm or more. Specifically, a mechanism for moving the upper table 24 in the z-axis direction with the upper plate member 22 as a base is installed. The moving mechanism is configured to include an actuator or an elastic member having a stroke of d ₀ mm or more and having responsiveness capable of avoiding a collision due to overshoot. Further, the upper table 24 and the lower table 35 may be moved simultaneously away from each other.

  In the above-described embodiment, the example in which the feedback gain of the feedback control in which the Z-VCM 86 tries to keep the same position is changed to a predetermined value smaller than the original value in step S415. However, in step S415, the control target of the Z-VCM 86 may be changed from the z-direction position of the lower table 35 and the inclination of the lower table 35 to only the inclination of the lower table 35.

Furthermore, in the above-described embodiment, the example in which the lower table 35 is pulled down by d ₀ mm or more by the self-weight canceller 36 is shown. However, the lower table 35 can be pulled down by the Z-VCM 86 instead of the self-weight canceller 36. In this case, it is not necessary to change the feedback gain in step S415.

  Further, a third modification will be described with a focus on differences from the above embodiment. In the third modified example, following the control flow described in FIG. 14, a control flow different from the above embodiment is executed. FIG. 21 is a control flow diagram of the superposition apparatus according to the third modification, which is executed following the control flow described in FIG. Each step receives an output from each sensor or a control unit controls each driving unit. In the third modification, the self-weight canceller 36 of FIG. 19 described in the second modification is used.

  When the first substrate 11 and the second substrate 12 are in surface contact in step S109, the process proceeds to step S511. In step S511, the lower table 35 is pressurized toward the upper table 24 by the self-weight canceller 36 in order to temporarily bond the first substrate 11 and the second substrate 12 together. The main joining may be performed by a heating / pressurizing device that is an external device of the overlaying apparatus 10 or may be configured to be performed in the overlaying apparatus 10.

  When the temporary bonding is completed in step S511, the adsorbent 111 of the first substrate holder 25 and the adsorbent 111 of the second substrate holder 41 are acted on to clamp the two substrate holders in step S512. That is, the first substrate holder 25 and the second substrate holder 41 sandwich the first substrate 11 and the second substrate 12 that are accurately aligned to create a state in which they are integrally fixed.

  In step S513, the upper table 24 releases the state where the first substrate holder 25 is sucked by vacuum suction, and the support of the first substrate holder 25 and the first substrate 11 by the upper table 24 is finished. To do. At this time, the first weight holder 25 and the second weight holder 41 are clamped between the upper table 24 and the lower table 35, because the pressure is higher than that of supporting the dead weight by the action of the dead weight canceller 36. ing.

In synchronization with the operation of releasing the suction state of the upper table 24 in step S513, in step S514, the control unit performs control to instantaneously lower the lower table 35 by d ₀ mm or more. In step S515, the feedback gain of feedback control in which the Z-VCM 86 tries to maintain the same position is changed to a predetermined value smaller than the original value. That is, step S514 and step S515 are steps of pulling the lower table 35 away from the upper table 24 from the state where the lower table 35 pressurizes the upper table 24. Synchronizing with the operation in which the upper table 24 releases the suction state does not necessarily coincide with the time if the lower table 35 is in a state of applying pressure to the upper table 24, and the suction state is released. It means that while maintaining the pressurized state.

  However, when temporary clamping is not performed as in the present processing flow but only alignment is performed and clamping is performed, the pressurized state may not be passed. In the state where the lower table 35 holds and balances the second substrate holder 41 on which the second substrate 12 is placed, not in the pressurized state, the first substrate holder 25 on which the first substrate 11 is placed. When the suction is released, the self-weights of the first substrate 11 and the first substrate holder 25 are added to the lower table 35 at this moment, so that synchronization requires time coincidence. Note that the change of the feedback gain of the feedback control is executed in synchronization with a predetermined timing, which will be described later in detail.

In the above description, the predetermined distance d ₀ mm is set in advance to a value equal to or greater than the distance corresponding to the overshoot that can occur when the weight addition of the first substrate 11 and the first substrate holder 25 acts as a step input. The The control of the self-weight canceller 36 that lowers the lower table 35 instantaneously by d ₀ mm or more can take the following two. First, the amount of subtraction obtained by subtracting the amount of fluid in the pressure chamber 60 that realizes the state after the reduction from the amount of fluid in the pressure chamber 60 that realizes the state before the reduction is controlled by the pressure regulator. This is a method of extracting from the pressure chamber 60. Here, the state after the lowering is a state in which the clamped first substrate holder 25 and the second substrate holder 41 are placed on the lower table 35 and are balanced below d ₀ mm compared to before the lowering. . In this case, the predetermined distance d ₀ mm itself is not meaningful, but it is significant that the distance is equal to or more than the distance corresponding to the overshoot. A sufficient amount of subtraction may be set by volume or mass.

The second is a method in which the self-weight canceller 36 that is normally pressure-controlled is temporarily switched to position control and monitored until the lower table 35 is pulled down by d ₀ mm or more. As described above, the accurate position of the lower table 35 in the z-axis direction is detected by the Z-interferometer 67. Therefore, until the value lower than d ₀ mm is obtained by the output of the Z-interferometer 67, the pressure regulator is controlled to extract the fluid from the pressure chamber 60.

In addition, the operation | movement which cancels | releases the adsorption | suction state of the upper table 24 of step S513 detects the timing with the output of the pressure gauge provided in the exhaust pipe connected with the upper table 24. FIG. The pressure gauge is preferably provided as close to the upper table 24 as possible. Of course, the predetermined distance d ₀ mm is smaller than the distance to the receiving surface of the shock absorber 92.

In step S516, the control unit further lowers the lower table 35 to a predetermined position by the self-weight canceller 36. Specifically, the predetermined position is a predetermined distance d ₀ mm or more, and the hand portion of the robot arm is inserted so that the first substrate holder 25 and the second substrate holder 41 can be carried out in a clamped state. It is a position that can be done. In step S517, the push-up pins 42 are operated so that they are slightly lifted from the lower table 35, and are carried out of the superposition apparatus 10 by the robot arm. As described above, a series of positioning steps is completed.

In the above-described embodiment, the lower table 35 is controlled to be instantaneously pulled down by d ₀ mm or more by the self-weight canceller 36. However, since the tables need only be relatively spaced apart by d ₀ mm or more, the upper table 24 may be controlled to be lifted by d ₀ mm or more. Specifically, a mechanism for moving the upper table 24 in the z-axis direction with the upper plate member 22 as a base is installed. The moving mechanism is configured to include an actuator or an elastic member having a stroke of d ₀ mm or more and having responsiveness capable of avoiding a collision due to overshoot. Further, the upper table 24 and the lower table 35 may be moved simultaneously away from each other.

  The reason why the feedback gain is changed to a predetermined value smaller than the original value in step S515 is that it is desired to lower the response sensitivity of the Z-VCM 86 to slow down the operation for maintaining the current state. That is, while the self-weight canceller 36 tries to pull down the lower table 35, the Z-VCM 86 does not work to lift the lower table 35. Even if such an operation is started, if the feedback gain is small, the speed at which the Z-VCM 86 attempts to lift the lower table 35 is slower than the speed at which the self-weight canceller 36 attempts to lower the lower table 35. Therefore, a situation where the lower table 35 collides with the upper table 24 can be avoided. Hereinafter, the change of the feedback gain will be described.

  FIG. 22 is a block diagram schematically showing a position control system in the direction of gravity of the lower table 35 according to the third modification. As described above, the control in the z-axis direction, which is the direction of gravity, is realized by the cooperative operation of the self-weight canceller 36 and the Z-VCM 86. Represents.

  The Z-VCM 86 drives the lower table 35 to be controlled based on the deviation between the actual position of the lower table 35 monitored by the Z-interferometer 67 and the given target position. At this time, the control system characteristics are determined by appropriately setting the integral gain Ki for the integral operation for obtaining the operation amount proportional to the integral value of the operation signal and the proportional gain Kp for the proportional operation for obtaining the operation amount proportional to the deviation.

FIG. 23 is a diagram showing response characteristics when the value of the proportional gain Kp is changed. In the figure, the horizontal axis represents elapsed time, and the vertical axis represents displacement. When the proportional gain Kp is set to a large value Kp ₁ , the target value is reached quickly as shown by the dotted line, but an overshoot exceeding the target value occurs. And the characteristic which converges to a target value is accompanied with a damped vibration. When the proportional gain Kp is set to Kp ₂ which is a value slightly smaller than Kp ₁ , the arrival at the target value is slightly delayed as shown by the alternate long and short dash line, and although there is no damped vibration, the characteristic is still that overshoot occurs. When overshoot occurs, as described above, the upper table 24 exists at a position exceeding the target value, so the first substrate holder 25 and the second substrate holder 41 placed on the second table. Will collide with the upper table 24. Therefore, the value of the proportional gain Kp ₃ that provides the characteristic indicated by the solid line is set. Kp ₃ is a smaller value than Kp ₂ . According to the characteristic of the proportional gain Kp ₃ , although the arrival time to the target value is slow, no overshoot occurs, so the first substrate holder 25 and the second substrate holder 41 do not collide with the upper table 24. .

  On the other hand, in a normal control state, it is preferable that the response characteristics reach the target value faster. Therefore, in this embodiment, the proportional gain Kp is changed from a large value to a small value at the timing described below. In any of the following timings, after the load cell 26 detects the contact between the first substrate 11 and the second substrate 12, the load cell 26 is moved to the first after the first substrate 11 and the second substrate 12 are overlaid. This is the timing until the non-contact between the first substrate holder 25 and the upper table 24 is detected.

  For the position control system in the gravity direction of the lower table 35, the weight addition of the first substrate 11 and the first substrate holder 25 when the upper table 24 releases the suction of the first substrate holder 25 is a step input. Works. Therefore, if the lower table 35 is not in a state of applying pressure to the upper table 24, at this point, the deviation that is the actual position with respect to the target position increases instantaneously. The return to the target position after the deviation has occurred depends on the magnitude of the proportional gain Kp as shown in FIG. Therefore, it is preferable to change the proportional gain Kp from a large value to a small value at the timing when the upper table 24 releases the suction of the first substrate holder 25.

  Since the timing at which the upper table 24 releases the suction of the first substrate holder 25 is a timing that is linked with other operations, other operations may be detected without directly detecting this timing. Another operation is, for example, the timing at which the first substrate 11 and the second substrate 12 are superposed, and the upper table 24 releases the drop-off preventing mechanism that prevents the first substrate holder 25 from dropping off. Timing is mentioned. As the drop-off prevention mechanism, a hook mechanism that supports the first substrate holder 25 and that is provided on the upper table 24 and that can be moved back and forth is used. These timings are preferably before the timing at which compressed air or compressed nitrogen is blown from the suction port provided in the upper table 24 to the first substrate holder 25.

  Further, even when the upper table 24 releases the suction of the first substrate holder 25, as long as the lower table 35 pressurizes the upper table 24 by the action of the self-weight canceller 36, it is still at this point in time. The release of suction does not act as a step input for the position control system of the lower table 35. This is because the self-weight canceller 36 pushes up the lower table 35, and as a result, the self-weight canceller 36 supports the weight of the first substrate 11 and the first substrate holder 25 and maintains the position of the lower table 35. In this case, when the pressurized state is released and the weight of the first substrate 11 and the first substrate holder 25 is applied to the lower table 35, it acts as a step input to the position control system of the lower table 35. . That is, this time may be set as a timing at which the proportional gain Kp is changed from a large value to a small value.

  Specifically, this timing can be a time when the output of the load cell 26 falls below a predetermined value. That is, if it can be detected that the upper table 24 is practically unloaded, it can be considered that the weight of the first substrate 11 and the first substrate holder 25 has moved to the lower table 35. This time is detected and the proportional gain Kp is changed.

  This timing can also be detected by monitoring the output of the Z-VCM 86. Immediately after acting as a step input to the position control system, the Z-VCM 86 rapidly increases its thrust to return to the target position. That is, the amount of increase in thrust per unit time exceeds a predetermined value set in advance. This time is detected and the proportional gain Kp is changed instantaneously.

  The proportional gain Kp changed at the above timing is restored when the situation where the lower table 35 collides with the upper table 24 can be avoided. Specifically, it is preferable that the lower table 35 is pulled down to a height at which the first substrate holder 25 and the second substrate holder 41 placed on the lower table 35 are carried out.

  In the above description, the first substrate 11 and the second substrate 12 are overlapped using the first substrate holder 25 and the second substrate holder 41, but the substrates are bonded to each other without using the substrate holder. A form of superimposing may be used. In this case, the weight addition to the lower table 35 is limited, but it is still effective because it does not change the step input.

  In the above description, the position control system of the lower table 35 in the direction of gravity has changed the proportional gain Kp so as to maintain the feedback control. However, feedback control may be terminated and switched to open loop control at the above timing. By switching in this way, even if there is a deviation between the target position and the actual position, an operation for canceling this does not occur. That is, the state in which the deviation has occurred is maintained. If the deviation is allowed in this way, a situation in which the lower table 35 collides with the upper table 24 can be avoided.

  In the above description, instead of changing the proportional gain Kp, it is also possible to feed forward the added weight at the above timing. In this case, the position displacement caused by the weight addition may be obtained experimentally and controlled so as to allow a predetermined deviation amount.

  In the above-described embodiment and its modification, the case where the upper table 24 and the lower table 35 are arranged along the direction of gravity has been described. For example, these two tables are opposed to each other in the horizontal direction, and two substrates are arranged. May be configured to overlap each other. Moreover, in said embodiment and its modification, although the timing which two board | substrates contact was detected, you may comprise so that not only this but the timing which board | substrate holders contact may be detected.

  Moreover, in said embodiment and its modification, the example in which the expansion-contraction mechanism of the self-weight canceller 36 was comprised with the piston 61 was shown. However, the expansion / contraction mechanism can be composed of, for example, a diaphragm or a bellows.

  Although the present invention has been described using the embodiment and its modifications, the technical scope of the present invention is not limited to the scope described in the embodiment. In particular, it is effective as a manufacturing method for a semiconductor device manufactured by applying the above 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.

DESCRIPTION OF SYMBOLS 10 Overlay apparatus, 11 1st board | substrate, 12 2nd board | substrate, 13 coarse movement base member, 13a upper surface, 13b edge part, 14 coarse movement stage part, 15 floor, 16 vibration isolation member, 17 fine movement base member, 18 Fine movement stage portion, 19 concave portion, 20 vibration isolation member, 21 body, 22 upper plate member, 23 column member, 24 upper table, 24a lower surface, 25 first substrate holder, 26 load cell, 27 ESC terminal, 28 upper microscope, 29 X-interferometer, 30 Y-interferometer, 32 XY-stage member, 33 X-drive unit, 34 Y-drive unit, 35 lower table, 35a upper surface, 35b lower surface, 36 self-weight canceller, 37 through-hole, 38 air pad, 39 XY-eddy current sensor, 40 Z-eddy current sensor, 41 2nd substrate holder, 42 push-up pin, 43 X-re Near motor, 44 magnet, 45 coil, 46 X-guide member, 47 air pad, 48 engaging part, 49 air bearing, 50, 51 connecting member, 52 air pad, 53 Y-linear motor, 54 magnet, 55 coil, 56 Y- Guide member, 57 engaging portion, 58 air bearing, 59 cylinder, 59b upper surface, 60 pressure chamber, 61 piston, 61a upper surface, 61b lower end portion, 62 air bearing, 63 spherical seat, 64 spherical member, 65 concave spherical surface, 66 air Bearing, 67 Z-interferometer, 68 Leg member, 68a Bottom surface, 69 Air bearing, 70 Air pad, 71 Insertion hole, 72 Insertion hole, 73 ESC terminal, 74 X-moving mirror, 75 Y-moving mirror, 76 Lower microscope, 77 Convex spherical surface, 78 Air bearing, 79 AF sensor, 0 X-VCM, 81 Magnet, 82 Coil, 83 Y-VCM, 84 Magnet, 85 Coil, 86 Z-VCM, 87 Magnet, 88 Coil, 89 Arm Member, 90 Through Hole, 91 Arm Member, 92 Shock Absorber, 110 Holder body, 111 adsorber, 112 positioning hole, 113 voltage application terminal, 121 attaching / detaching mechanism, 122 piping, 123 valve, 124 conduit

Claims (44)

An overlay device that superimposes a first substrate and a second substrate on each other,
A first stage for holding the first substrate;
A second stage that holds the second substrate and moves to overlap the first substrate and the second substrate;
When the holding of the first stage with respect to the first substrate overlaid on the second substrate is released, displacement occurs with the load movement of the first substrate from the first stage to the second stage. A collision prevention unit for preventing the first substrate from colliding with the first stage by the return operation of the second stage;
A drive unit for moving the second stage in the direction of the first stage;
The collision prevention unit includes a detection unit that detects that the driving force of the driving unit can be applied to the first stage via the second stage, and a control unit that controls the driving unit. Have
The control unit is after the detection unit detects that the driving force of the driving unit can be applied to the first stage, and the first stage and the first substrate are separated from each other. And a superposing apparatus for controlling the driving unit so that a driving force having a magnitude corresponding to at least a load of the first substrate is applied from the second stage to the first stage. The first stage and the second stage face each other in the direction of gravity, and the second stage is disposed below the first stage,
The drive unit includes an actuator that drives the second stage toward the first stage,
The superimposing apparatus according to claim 1, wherein the control unit generates a driving force having a magnitude corresponding to the load by the actuator.   The drive unit further includes a support mechanism that drives the second stage in the direction of the first stage, and the control unit transfers a drive force having a magnitude corresponding to the load from the actuator to the support mechanism. The superposition device according to claim 2 to be made.   The superimposing apparatus according to claim 3, wherein the transition is performed between the time when the first stage releases the holding of the first substrate and the time when the first stage and the first substrate are separated from each other. The first stage and the second stage oppose each other in the direction of gravity, and the second stage is disposed below the first stage,
The drive unit has a support mechanism for supporting the second stage in the direction of gravity,
The superimposing apparatus according to claim 1, wherein the control unit generates a driving force having a magnitude corresponding to the load by a support mechanism.   The control unit causes the support mechanism to generate a driving force that exceeds the force corresponding to the load and presses the second substrate toward the first substrate, and then reduces the driving force to a size corresponding to the load. The superposition apparatus according to claim 5. At least one of the first substrate and the second substrate is held by a substrate holder;
The collision prevention unit according to any one of claims 1 to 6, wherein when the first substrate is held by the substrate holder, the displacement prevents the substrate holder from colliding with the first stage. The overlay apparatus described.   The control unit has a size corresponding to a load of the substrate holder when the first substrate is held by the substrate holder and the first substrate is separated from the first stage integrally with the substrate holder. The superposition apparatus according to claim 7, wherein the driving unit is controlled to apply the driving force of the second stage so as to act on the first stage from the second stage.   The detection unit detects that the driving force of the driving unit can be applied to the first stage via the second stage by contacting the first substrate and the second substrate. The overlay apparatus according to any one of claims 1 to 8. An overlay device that superimposes a first substrate and a second substrate on each other,
A first stage for holding the first substrate;
A second stage that holds the second substrate and moves to overlap the first substrate and the second substrate;
When the holding of the first stage with respect to the first substrate overlaid on the second substrate is released, displacement occurs with the load movement of the first substrate from the first stage to the second stage. A collision preventing unit for preventing the first substrate from colliding with the first stage due to the return operation of the second stage,
The collision preventing unit is a superposition apparatus having a gas blowing mechanism for jetting gas between the first substrate and the first stage. The collision prevention unit includes a control unit that controls the fuming mechanism.
The superimposing apparatus according to claim 10, wherein the control unit ejects the gas in synchronization with a timing at which the first stage releases the holding of the first substrate.   The superimposing apparatus according to claim 11, wherein the control unit reduces the amount of gas ejection as the distance between the first stage and the second stage increases.   The said 1st stage is provided with the hole through which the said gas is ejected, The said air blowing mechanism serves as the holding mechanism which adsorb | sucks the said 1st board | substrate to the said 1st stage by sucking in from the said hole. The superposition apparatus according to 11 or 12. At least one of the first substrate and the second substrate is held by a substrate holder;
The collision prevention unit according to any one of claims 11 to 13, wherein when the first substrate is held by the substrate holder, the displacement prevents the substrate holder from colliding with the first stage. The overlay apparatus described. A detector for detecting that the first substrate and the second substrate are in contact with each other;
The superposition apparatus according to claim 11, wherein the control unit ejects the gas after the detection unit detects that the first substrate and the second substrate are in contact with each other. An overlay device that superimposes a first substrate and a second substrate on each other,
A first stage for holding the first substrate;
A second stage that holds the second substrate and moves to overlap the first substrate and the second substrate;
When the holding of the first stage with respect to the first substrate overlaid on the second substrate is released, displacement occurs with the load movement of the first substrate from the first stage to the second stage. A collision prevention unit for preventing the first substrate from colliding with the first stage by the return operation of the second stage;
A drive unit that relatively moves at least one of the first stage and the second stage along directions facing each other;
The collision prevention unit has a control unit for controlling the drive unit,
The controller controls the drive unit so that a distance between the second stage and the first stage on which the load of the first substrate released from being held on the first stage is applied is greater than a predetermined distance. Superposition device to control.   The control unit is configured to release the holding of the first substrate after the first substrate and the second substrate are overlapped with each other, and the load of the first substrate is applied to the second substrate. The superposition apparatus according to claim 16, wherein the driving unit is controlled so that an interval between the first stage and the second stage is larger than the predetermined distance at a predetermined time point before acting on the stage. The first stage and the second stage face each other in the direction of gravity, and the second stage is disposed below the first stage,
The drive unit includes a support mechanism that includes a telescopic mechanism that can expand and contract in the direction of gravity and supports the second stage in the direction of gravity.
18. The superposition apparatus according to claim 17, wherein the control unit separates the second stage from the first stage by a predetermined amount by extracting the fluid of the expansion / contraction mechanism by a predetermined amount at the predetermined time point. . The first stage and the second stage face each other in the direction of gravity, and the second stage is disposed below the first stage,
The drive unit includes a support mechanism that includes a telescopic mechanism that can expand and contract in the direction of gravity and supports the second stage in the direction of gravity.
The superposition apparatus according to claim 17, wherein the control unit switches the control of the expansion and contraction mechanism from pressure control to position control, and separates the second stage from the first stage by a predetermined distance. The first stage and the second stage face each other in the direction of gravity, and the second stage is disposed below the first stage,
The drive unit includes a support mechanism and an actuator for driving the second stage toward the first stage,
18. The superimposing apparatus according to claim 17, wherein the control unit makes a feedback gain value of the actuator smaller than a value before the predetermined time in synchronization with the predetermined time.   The overlay apparatus according to claim 17, wherein the predetermined time includes a time when the first substrate is separated from the first stage. At least one of the first substrate and the second substrate is held by a substrate holder;
The collision prevention unit according to any one of claims 17 to 21, wherein when the first substrate is held by the substrate holder, the displacement prevents the substrate holder from colliding with the first stage. The overlay apparatus described. A detector for detecting that the first substrate and the second substrate are in contact with each other;
The control unit is configured so that the interval between the first stage and the second stage is greater than a predetermined distance at a time after the detection unit detects that the first substrate and the second substrate are in contact with each other. The superposition apparatus according to any one of claims 17 to 22, which controls the drive unit. An overlay device that superimposes a first substrate and a second substrate on each other,
A first stage for holding the first substrate;
A second stage that holds the second substrate and moves to overlap the first substrate and the second substrate;
When the holding of the first stage with respect to the first substrate overlaid on the second substrate is released, displacement occurs with the load movement of the first substrate from the first stage to the second stage. A collision prevention unit for preventing the first substrate from colliding with the first stage by the return operation of the second stage;
A drive unit that moves at least one of the first stage and the second stage in a direction facing each other;
The collision prevention unit has a control unit for controlling the drive unit,
The control unit is configured to superimpose a feedback gain value of the driving unit at a predetermined time after the first substrate and the second substrate are overlapped with each other to be smaller than a value before the predetermined time. .   The predetermined time point is after the first substrate and the second substrate are overlapped with each other, and the first stage releases the holding of the first substrate and the load on the first substrate is changed to the first substrate. 25. The superimposing apparatus according to claim 24, which is a point in time until the two stages are acted on.   26. The overlay apparatus according to claim 25, wherein the predetermined time includes a time when the first substrate releases the holding of the first substrate or a time when the first substrate moves away from the first stage.   26. The superimposing apparatus according to claim 25, wherein the predetermined time is a time when superposition of the first substrate and the second substrate is completed. A detector for detecting contact between the first substrate and the second substrate;
26. The control unit, after the detection unit detects contact between the first substrate and the second substrate, makes a feedback gain value of the driving unit smaller than a value before the predetermined time point. 27. The superposition apparatus according to any one of 27.   26. The superimposing apparatus according to claim 25, wherein the predetermined time is a time when the amount of increase in thrust per unit time in the driving unit exceeds a predetermined value. An overlay device that superimposes a first substrate and a second substrate on each other,
A first stage for holding the first substrate;
A second stage that holds the second substrate and moves to overlap the first substrate and the second substrate;
When the holding of the first stage with respect to the first substrate overlaid on the second substrate is released, displacement occurs with the load movement of the first substrate from the first stage to the second stage. A collision prevention unit for preventing the first substrate from colliding with the first stage by the return operation of the second stage;
A drive unit that moves at least one of the first stage and the second stage in a direction facing each other;
The collision prevention unit is a control unit that controls the driving unit,
The controller is a superposition device that switches the control of the drive unit from feedback control to open loop control at a predetermined time after the first substrate and the second substrate are superposed on each other.   The predetermined time point is after the first substrate and the second substrate are overlapped with each other, and the first stage releases the holding of the first substrate, and the load on the first substrate becomes the second load. The superimposing apparatus according to claim 30, which is a time point until the stage is acted on. At least one of the first substrate and the second substrate is held by a substrate holder;
32. The collision prevention unit according to any one of claims 24 to 31, wherein the collision prevention unit prevents the substrate holder from colliding with the first stage due to the displacement when the first substrate is held by the substrate holder. The overlay apparatus described.   33. The apparatus according to claim 32, wherein when the first substrate is held by the substrate holder, the predetermined time point includes a time point when the substrate holder is separated from the first stage.   The overlay apparatus according to any one of claims 1 to 30, wherein at least one of the first substrate and the second substrate includes a plurality of substrates already stacked. An overlapping method for overlapping a plurality of substrates,
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step of moving at least the second stage toward the first stage by a driving unit;
A detecting step for detecting that the driving force of the second stage can act on the first stage from the second stage;
After detecting that the operation of the driving force to the first stage is enabled in the detection step, the first stage releases the holding of the first substrate and the load on the first substrate A control step for controlling the drive unit so that at least a driving force corresponding to the load of the first substrate is applied from the second stage to the first stage until the first stage is applied to the second stage. And a superposition method. An overlapping method for overlapping a plurality of substrates,
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step of moving the second stage toward at least the first stage by a driving unit;
A superposition method comprising a step of blowing air between the first stage and the first substrate in synchronization with a timing at which the first stage releases the holding of the first substrate. An overlapping method for overlapping a plurality of substrates,
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step for causing the second stage and the first stage to approach each other by a drive unit;
The first stage releases the holding of the first substrate and controls the driving unit so that the distance between the second stage and the first stage on which the load of the first substrate is applied is greater than a predetermined distance. A superimposing method. An overlapping method for overlapping a plurality of substrates,
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step for causing the second stage and the first stage to approach each other by a drive unit;
An overlaying step of overlaying the first substrate and the second substrate;
A superposition method comprising: a gain changing step of making a feedback gain value of the driving unit smaller than a value before the predetermine time at a predetermined time after the superposition step. An overlapping method for overlapping a plurality of substrates,
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step for causing the second stage and the first stage to approach each other by a drive unit;
An overlaying step of overlaying the first substrate and the second substrate;
And a control switching step of switching the control of the drive unit from feedback control to open loop control at a predetermined time after the superposition step. A device manufacturing method manufactured by stacking a plurality of substrates,
The step of superimposing the plurality of substrates includes:
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step of moving at least the second stage toward the first stage by a driving unit;
A detecting step for detecting that the driving force of the second stage can act on the first stage from the second stage;
After detecting that the operation of the driving force to the first stage is enabled in the detection step, the first stage releases the holding of the first substrate and the load on the first substrate A control step for controlling the drive unit so that at least a driving force corresponding to the load of the first substrate is applied from the second stage to the first stage until the first stage is applied to the second stage. A device manufacturing method including: A device manufacturing method manufactured by stacking a plurality of substrates,
The step of superimposing the plurality of substrates includes:
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step of moving the second stage toward at least the first stage by a driving unit;
A device manufacturing method, comprising: a blowing step of blowing a gas between the first stage and the first substrate in synchronization with a timing at which the first stage releases the holding of the first substrate. A device manufacturing method manufactured by stacking a plurality of substrates,
The step of superimposing the plurality of substrates includes:
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step for causing the second stage and the first stage to approach each other by a drive unit;
A control step of controlling the drive unit so that a distance between the second stage on which the load of the first substrate released from the holding on the first stage is applied and the first stage is larger than a predetermined distance; A device manufacturing method including: A device manufacturing method manufactured by stacking a plurality of substrates,
The step of superimposing the plurality of substrates includes:
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step for causing the second stage and the first stage to approach each other by a drive unit;
An overlaying step of overlaying the first substrate and the second substrate;
A device manufacturing method including a gain changing step of making a feedback gain value of the driving unit smaller than a value before the predetermined time at a predetermined time after the superposition step. A device manufacturing method manufactured by stacking a plurality of substrates,
The step of superimposing the plurality of substrates includes:
A first holding step for holding the first substrate on the first stage;
A second holding step for holding the second substrate opposite to the first substrate on the second stage;
A moving step for causing the second stage and the first stage to approach each other by a drive unit;
An overlaying step of overlaying the first substrate and the second substrate;
And a control switching step of switching the control of the drive unit from feedback control to open loop control at a predetermined time after the superposition step.

Images