Laminating device and method

Abstract

To provide a laminating device that can correct a misalignment in the observational position of an alignment mark of a substrate caused by a curvature of a substrate holder and a method.SOLUTION: A laminating device 100 includes a first stage that holds a first substrate, a second stage 223 that holds a second substrate 213 by curving it so that the second substrate is opposite to the first substrate, an observation unit that observes the position of an alignment mark 218 provided on the surface of the second substrate held on the second stage, and a control unit that controls at least one of the first stage and the second stage on the basis of the location of the alignment mark observed by the observation unit. The control unit controls at least one of the first stage and the second stage on the basis of a misalignment in the observational position of the alignment mark caused by the curvature.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present invention relates to a lamination apparatus and method. [Background technology]

[0002] Patent Document 1 describes an apparatus for aligning a pair of substrates by observing alignment marks formed on the substrates with a microscope. [Prior art document] [Patent documents] [Patent Document 1] JP 2005-251972 A Summary of the Invention

[0003] In a first aspect of the present invention, there is provided a stacking device including a first stage that holds a first substrate, a second stage that holds a second substrate in a curved state, and an observation unit that observes an alignment mark provided on the second substrate held in a curved state by the second stage, and controls at least one of the first stage in a state where it is holding the first substrate and the second stage in a state where it is holding the second substrate based on the position of the alignment mark observed by the observation unit and information regarding the curvature of the second substrate held in a curved state by the second stage.

[0004] In a second aspect of the present invention, there is provided a method comprising holding a first substrate on a first stage, bending a second substrate and holding it on a second stage, observing an alignment mark on the second substrate held in a curved state on the second stage, and controlling at least one of the first stage holding the first substrate and the second stage holding the second substrate based on the position of the alignment mark observed by the observation and information regarding the curvature of the second substrate held in a curved state on the second stage.

[0005] In a third aspect of the present invention, there is provided a method comprising holding a first substrate on a first stage, curving a second substrate and holding it on a second stage, observing an alignment mark provided on the second substrate held in a curved state on the second stage, and performing processing using an enhanced global alignment method based on the position of the alignment mark observed by the observation and information regarding the curvature of the second substrate held in a curved state on the second stage.

[0006] The above summary of the invention does not list all of the necessary features of the present invention. Also, subcombinations of these features may also be inventions. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a schematic plan view of a stacking device 100 according to a first embodiment. [Diagram 2] 2 is a schematic plan view of substrates 211 and 213 to be stacked in the stacking device 100. FIG. [Diagram 3] 1 is a schematic cross-sectional view of a substrate holder 221 that holds an upper substrate 211 of two substrates 211, 213 to be stacked in the stacking device 100. FIG. [Figure 4] 1 is a schematic cross-sectional view of a substrate holder 223 that holds a lower substrate 213 of two substrates 211, 213 that are stacked in the stacking device 100. FIG. [Diagram 5] 1 is a flowchart showing a procedure for laminating substrates 211 and 213 in the lamination device 100 to produce a laminate 230. [Figure 6] 13 is a diagram showing the structure of stacking section 300, as well as the state after substrate holders 221 and 223 holding substrates 211 and 213 have been carried into stacking section 300. FIG. [Figure 7] 11A and 11B are diagrams illustrating the operation of the stacking unit 300 in step S03. [Figure 8] 13(a) to 13(c) are schematic diagrams showing parallax in the XY directions caused by the curvature of the substrate holder 223. FIG. [Figure 9] 1A is a diagram showing the relationship between the position of alignment mark 218 and the amount of parallax for each amount of convexity of substrate holder 223, and FIG. 1B is a diagram showing the relationship between the position of alignment mark 218 and the amount of magnification change caused by parallax for each amount of convexity of substrate holder 223. [Figure 10] 2 is a schematic cross-sectional view of a laminated portion 300 in a state in which substrates 211 and 213 are aligned. FIG. [Figure 11] FIG. 2 is a diagram showing the state of substrates 211, 213 and substrate holders 221, 223 in an aligned state. [Figure 12] 13 is a schematic cross-sectional view of a laminated portion 300 at the start of bonding substrates 211 and 213. FIG. [Figure 13] FIG. 2 is a diagram showing the state of substrates 211, 213 and substrate holders 221, 223 when bonding is started. [Figure 14] 2A and 2B are diagrams showing the state of substrates 211 and 213 and substrate holders 221 and 223 in a state in which the holding of substrate 211 by substrate holder 221 is released. [Figure 15] 13(a) to 13(c) are diagrams showing the parallax in the Z direction of the alignment mark 218 caused by the curvature of the substrate holder 223 in the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] The present invention will be described below through the embodiments of the invention. The following embodiments do not limit the invention according to the claims. Not all of the combinations of features described in the embodiments are essential to the solution of the invention.

[0009] [First embodiment] 1 is a schematic plan view of a stacking device 100 in the first embodiment. The stacking device 100 is a device that stacks one substrate 211 with another substrate 213 to form a stack 230, and includes a housing 110, substrate cassettes 120 and 130 disposed outside the housing 110, a control unit 150, a transport unit 140 disposed inside the housing 110, a stacking unit 300, a holder stocker 400, a pre-aligner 500, and an activation device 600.

[0010] One substrate cassette 120 accommodates substrates 211, 213 to be stacked. The other substrate cassette 130 accommodates a plurality of stacks 230 formed by stacking the substrates 211, 213. By using the substrate cassette 120, the plurality of substrates 211, 213 can be loaded into the stacking device 100 all at once. In addition, by using the substrate cassette 130, the plurality of stacks 230 can be unloaded from the stacking device 100 all at once.

[0011] The transport section 140 transports any one of the substrates 211, 213, the substrate holders 221, 223, the substrate holders 221, 223 holding the substrates 211, 213, and the stack 230 formed by stacking the substrates 211, 213, within the stacking device 100. The holder stocker 400 accommodates a plurality of the substrate holders 221, 223.

[0012] The substrate holders 221, 223 are an example of a holding member, and are disk-shaped members having dimensions slightly larger than the substrates 211, 213 and high rigidity. Each of the substrate holders 221, 223 has a substrate adsorption function such as an electrostatic chuck or a vacuum chuck, and holds the substrates 211, 213 individually inside the stacking device 100. In this way, the substrate holders 221, 223 are transported together with the substrates 211, 213 by the transport unit 140 while protecting the fragile substrates 211, 213.

[0013] The pre-aligner 500 detects the centers of the substrates 211, 213, and holds them after aligning them with the centers of the substrate holders 221, 223. The pre-aligner 500 also aligns the relative positions and directions of the substrates 211, 213 and the substrate holders 221, 223 to reduce the load of the alignment process in the stacking unit 300, which will be described later. The pre-aligner 500 is also used when separating the stack 230, which has been removed from the stacking unit 300, from the substrate holders 221, 223.

[0014] The control unit 150 is composed of a processor such as a CPU, FPGA, or ASIC, and memories such as a ROM, RAM, and controls each unit of the stacking device 100 in a coordinated manner based on a control program. The control unit 150 also receives instructions from an external user and sets manufacturing conditions for manufacturing the stacked body 230. The control unit 150 also has a user interface that displays the operating state of the stacking device 100 to the outside.

[0015] The activation device 600 generates plasma that activates the upper surface of the substrate 211. The substrates 211 and 213 activated by the activation device 600 autonomously adhere to each other and are bonded together by contacting or approaching each other.

[0016] The stacking unit 300 has a pair of opposing stages, each holding the substrates 211, 213, and aligns the substrates 211, 213 held on the stages with respect to each other, and then stacks the substrates in contact with each other to form the stack 230. Thus, the stacking unit 300 functions as an alignment device that aligns the substrates 211, 213 with respect to each other. In the stacking unit 300, the substrates 211, 213 are carried in while being held by the substrate holders 221, 223, and stacked.

[0017] The stack 230 formed in the stacking device 100 having the above-mentioned structure is finally separated from the substrate holders 221, 223 and carried out from the stacking device 100. Therefore, the substrate holders 221, 223 separated from the substrates 211, 213 or the stack 230 remain inside the stacking device 100 and are used repeatedly. Therefore, the substrate holders 221, 223 can be considered as parts of the stacking device 100.

[0018] The substrates 211, 213 to be stacked in the stacking device 100 may be unprocessed silicon wafers, compound semiconductor wafers, glass substrates, etc., in addition to the substrates 211, 213 on which elements, circuits, terminals, etc. are formed. The combination of the substrates 211, 213 to be stacked may be a circuit board and an unprocessed substrate, or two unprocessed substrates. Furthermore, the substrates 211, 213 to be stacked may themselves be a stack 230 already formed by stacking a plurality of substrates.

[0019] 2 is a schematic plan view of substrates 211 and 213 to be laminated in the lamination apparatus 100. The substrates 211 and 213 each have a notch 214, a plurality of circuit regions 216, and a plurality of alignment marks 218.

[0020] The circuit regions 216 are periodically arranged on the surfaces of the substrates 211, 213 in the planar direction of the substrates 211, 213. Each of the circuit regions 216 is provided with a semiconductor device, wiring, a protective film, etc., formed by photolithography technology or the like. The circuit regions 216 also include structures including connecting portions such as pads and bumps that serve as connecting terminals when electrically connecting the substrates 211, 213 to other substrates 211, 213, lead frames, etc.

[0021] The alignment marks 218 are an example of a structure formed on the surfaces of the substrates 211, 213, and are arranged so as to overlap the scribe lines 212 arranged between the circuit regions 216. The alignment marks 218 are used as indicators when aligning the substrates 211, 213 with other substrates 211, 213 to be stacked.

[0022] 3 is a schematic cross-sectional view of a substrate holder 221 that holds the upper substrate 211 of the two substrates 211, 213 to be stacked in the stacking device 100. The substrate holder 221 has a flat holding surface 225 and has a function of adsorbing and holding the substrate 211, such as an electrostatic chuck or a vacuum chuck.

[0023] As described below, one substrate 211 of the two substrates 211, 213 is released from the hold by the substrate holder 221 when it is stacked on the other substrate 213. In this embodiment, one substrate 211 is held by the substrate holder 221 having a flat holding surface 225.

[0024] 4 is a schematic cross-sectional view of a substrate holder 223 that holds the lower substrate 213 of the two substrates 211, 213 to be stacked in the stacking device 100. In the illustrated example, the holding surface 225 of the substrate holder 223 has a shape with a raised center. This creates a height difference Z between the center and the edge of the holding surface 225 of the substrate holder 223. _WH In the following description, the shape of the substrate holder 223 is referred to as a convex shape, and the height difference Z _WH This is called the median convexity amount.

[0025] The substrate holder 223 has a function of adsorbing and holding the substrate 211, such as an electrostatic chuck or a vacuum chuck. Therefore, the substrate 213 held by the substrate holder 223 is curved along the shape of the holding surface 225, and is deformed convexly with the center of the substrate 213 as the apex.

[0026] FIG. 5 is a flow chart showing a procedure for stacking the substrates 211 and 213 in the stacking apparatus 100 to produce the stack 230. As shown in FIG.

[0027] The control unit 150 activates the bonding surfaces of the substrates 211, 213 to be stacked (step S01). The control unit 150 causes the surfaces of the substrates 211, 213 to be scanned with plasma generated by the activation device 600. This cleans the surfaces of the substrates 211, 213, and increases their chemical activity. Therefore, the substrates 211, 213 autonomously adhere to each other and bond to each other by coming into contact with or approaching each other.

[0028] In addition to the method of exposing the substrates 211, 213 to plasma, the substrates 211, 213 can be activated by sputter etching using an inert gas, ion beams, fast atomic beams, or mechanical processing such as polishing. When using an ion beam or fast atomic beam, the laminated portion 300 can be generated under reduced pressure. Furthermore, the substrates 211, 213 can also be activated by ultraviolet irradiation, an ozone asher, or the like. Furthermore, the substrates 211, 213 may be activated by chemically cleaning the surfaces thereof using, for example, a liquid or gaseous etchant.

[0029] Next, in the pre-aligner 500, the substrate holders 221, 223 are caused to hold the substrates 211, 213, and the substrate holders 221, 223 individually holding the substrates 211, 213 to be superimposed on each other are carried into the stacking section 300 (step S02).

[0030] 6 is a diagram showing the structure of the stacking unit 300, as well as a state after the substrate holders 221, 223 holding the substrates 211, 213 are carried into the stacking unit 300. The stacking unit 300 in the stacking device 100 includes a frame 310, a fixed stage 322, and a movable stage 332.

[0031] The frame 310 has a bottom plate 312 and a top plate 316 that are parallel to the floor surface 301 , and a plurality of support columns 314 that are perpendicular to the floor surface 301 .

[0032] Fixed stage 322, which is fixed facing downward to the lower surface of top plate 316 in the figure, has a holding function such as a vacuum chuck or an electrostatic chuck. As shown in the figure, fixed stage 322 holds substrate 211 together with substrate holder 221 having flat holding surface 225.

[0033] A microscope 324 is fixed to the lower surface of the top plate 316. The microscope 324 can observe the upper surface of the substrate 213 held on a moving stage 332 disposed opposite the fixed stage 322.

[0034] The moving stage 332 is mounted on a Y-direction drive unit 333 that moves in the direction indicated by the arrow Y in the figure. The Y-direction drive unit 333 is stacked on an X-direction drive unit 331 arranged on the bottom plate 312. The X-direction drive unit 331 moves in the direction indicated by the arrow X in the figure, parallel to the bottom plate 312. This allows the moving stage 332 to move two-dimensionally in the X and Y directions. The moving stage 332 shown in the figure holds a substrate 211 held by a substrate holder 223. Although not shown, the substrate holder 223 has a curved holding surface 225, and the substrate 213 is also held in a curved state along the holding surface 225.

[0035] Incidentally, the fixed stage 322 or the moving stage 332 on which the substrates 211, 213 are placed in the stacking section 300 of the stacking device 100 may directly hold the substrates 211, 213 without using the substrate holders 221, 223. In this case, the fixed stage 322 or the moving stage 332 serves as a holding member.

[0036] Furthermore, the moving stage 332 is raised and lowered relative to the Y direction driving section 333 by a Z direction driving section 335 which is raised and lowered in the direction indicated by the arrow Z.

[0037] The amount of movement of the moving stage 332 by the X-direction drive unit 331, the Y-direction drive unit 333, and the Z-direction drive unit 335 is precisely measured using an interferometer or the like. The X-direction drive unit 331 and the Y-direction drive unit 333 may also be configured in two stages, a coarse movement unit and a fine movement unit. This allows both highly accurate alignment and high throughput, and allows the substrate 211 mounted on the moving stage 332 to be bonded at high speed without reducing control accuracy.

[0038] A microscope 334 is further mounted on each of the Y direction drive units 333 on the sides of the moving stage 332. The microscope 334 can observe the lower surface of the substrate 211 held by the fixed stage 322 and facing downward.

[0039] The stacking unit 300 may further include a rotation drive unit that rotates the moving stage 332 around a rotation axis perpendicular to the bottom plate 312, and a swing drive unit that swings the moving stage 332. This makes it possible to adjust the tilt angle of the moving stage 332 to make the moving stage 332 parallel to the fixed stage 322, and to rotate the substrate 213 held by the moving stage 332, thereby improving the alignment accuracy of the substrates 211, 213.

[0040] The control unit 150 calibrates the microscopes 324, 334 in advance by adjusting the focus of the microscopes 324, 334 with respect to each other and by having them observe a common index. This allows the relative positions of the pair of microscopes 324, 334 in the stacking unit 300 to be measured. Next, referring again to Fig. 2, in the stacking unit 300, the alignment marks 218 formed on each of the substrates 211, 213 are detected (step S03).

[0041] 7 is a diagram illustrating the operation of the stacking unit 300 in step S03. In step S03, the control unit 150 operates the X-direction driving unit 331 and the Y-direction driving unit 333 to cause the microscopes 324 and 334 to detect the alignment marks 218 provided on the substrates 211 and 213, respectively.

[0042] In this way, the control unit 150 detects the positions of the alignment marks 218 on the substrates 211, 213 using the microscopes 324, 334 whose relative positions are known, and calculates the relative positions of the substrates 211, 213, and calculates the amount of movement of the moving stage 332 (step S04). That is, in the stacking unit 300, the control unit 150 calculates the amount of movement of the moving stage 332 so that the corresponding circuit regions 216 overlap each other. Note that, if the moving stage 332 is tilted and adjustment of the tilt angle is required, the control unit 150 also calculates the amount of adjustment of the tilt angle.

[0043] The amount of movement of the moving stage 332 calculated in step S04 can be calculated by performing a method of measuring and statistically processing the positions of the alignment marks 218 on the substrates 211 and 213 at multiple points. As the method, for example, Enhanced Global Alignment (EGA) can be given.

[0044] Here, it is assumed that statistical processing by the enhanced global alignment method (hereinafter also referred to as EGA) is performed based on the position of the alignment mark 218 when the substrates 211, 213 are in a flat state. However, when the position of the alignment mark 218 is measured in a curved state while the substrate 213 is held by a substrate holder 223 having a convex shape as shown in Fig. 4, the position of the alignment mark 218 deviates from the flat position of the substrate 213, generating parallax, which is the amount of deviation observed, and thus causing deviation in the processing result of the EGA. Therefore, it is desirable to correct the parallax caused by the curvature of the substrate holder 223.

[0045] 8(a) to (c) are schematic diagrams showing parallax in the XY directions caused by the curvature of the substrate holder 223. The XY directions are directions along the surface of the arranged substrate 213. Note that in FIGS. 8(a) to (c), the amount of movement of the alignment mark 218 and characteristic parts such as the shape of the substrate holder 223 are exaggerated. As shown in FIG. 8(a), when the substrate 213 is flat, the alignment mark 218 is observed at position F1 in the XY plane.

[0046] As shown in FIG. 8(b), when the substrate 213 is held by the substrate holder 223 having a convex shape and curved, the position of the alignment mark 218 on the observation is moved by a distance of δxy to a position F2 on the XY plane, and the alignment mark 218 is observed at the position F2 on the XY plane. Therefore, an observational shift amount δxy (also called a parallax amount) of the alignment mark 218 in the XY direction is generated. Therefore, as shown in FIG. 8(c), it is desirable to derive a difference between the position F1 and the position F2 when the substrate 213 is virtually returned to a flat surface, and correct the position of the alignment mark 218 based on the difference. The state before the substrate 213 is curved may be used as the virtual state, instead of the state in which the substrate 213 is flat. For convenience, the alignment mark 218 is depicted in FIG. 8(b) as being shifted only in the X direction.

[0047] FIG. 9(a) is a diagram showing the relationship between the position of alignment mark 218 and the amount of parallax for each amount of convexity of substrate holder 223. In FIG. 9(a), the horizontal axis indicates the distance of alignment mark 218 from the center of convexity of substrate holder 223, and the vertical axis indicates the amount of parallax in the XY directions of alignment mark 218. Note that the center of convexity of substrate holder 223 refers to the position at which substrate holder 223 protrudes most due to the curvature of substrate holder 223. In FIG. 9(a), the amount of convexity of substrate holder 223 is smallest for curve S1 and largest for curve S5. That is, the amount of convexity of substrate holder 223 is smallest for curve S1 and largest for curve S5. <S2<S3<S4<S5である。

[0048] The amount of parallax of the alignment mark 218 is determined according to the distance of the alignment mark 218 from the convex center of the substrate holder 223 and the amount of convexity of the substrate holder 223. It can be seen from Fig. 9(a) that the amount of parallax of the alignment mark 218 increases as the distance of the alignment mark 218 from the convex center of the substrate holder 223 increases, and also increases as the amount of convexity of the substrate holder 223 increases.

[0049] Here, the magnification of the substrate 213 also changes due to the parallax of the alignment mark 218, and there is a possibility that a deviation occurs in the EGA processing result. Here, the magnification of the substrate 213 refers to the amount of deviation that gradually increases radially from a certain point (for example, the center) on the substrate 213 in the surface direction. FIG. 9(b) is a diagram showing the relationship between the position of the alignment mark 218 for each amount of convexity of the substrate holder 223 and the amount of change in magnification due to parallax. Here, the amount of change in magnification of the substrate 213 refers to the amount of change in the magnification of the substrate 213 in the radial direction. In FIG. 9(b), the horizontal axis indicates the distance from the center of the convexity of the substrate holder 223 to the alignment mark 218, and the vertical axis indicates the amount of change in magnification due to parallax. In FIG. 9(b), the amount of convexity of the substrate holder 223 is smallest for the curve S6, and largest for the curve S10. That is, the amount of convexity of the substrate holder 223 is smallest for the curve S6, and largest for the curve S10. <S7<S8<S9<S10である。

[0050] The amount of change in magnification caused by parallax is determined according to the distance of the alignment mark 218 from the center of the convexity of the substrate holder 223 and the amount of convexity of the substrate holder 223. It can be seen from Fig. 9(b) that the amount of change in magnification caused by parallax increases as the distance of the alignment mark 218 from the center of the convexity of the substrate holder 223 increases, and also increases as the amount of convexity of the substrate holder 223 increases.

[0051] From the above, it can be seen that the parallax amount of the alignment mark 218 and the amount of change in magnification caused by the parallax are determined according to the distance of the alignment mark 218 from the convex center of the substrate holder 223 and the convex amount of the substrate holder 223. Hereinafter, a method of deriving the deviation amount δxy using the distance of the alignment mark 218 from the convex center of the substrate holder 223 and the convex amount of the substrate holder 223 will be described. In the following description, it is assumed that the convex center of the substrate holder 223 coincides with the center position of the substrate holder 223. The center position of the substrate holder 223 is, for example, the geometric center of gravity of the surface on which the substrate is placed. Note that instead of the distance of the alignment mark 218 from the convex center of the substrate holder 223, the deviation amount δxy may be derived using the coordinates of the alignment mark 218 in a coordinate system whose origin is the convex center of the substrate holder 223. Note that the coordinates of the convex center of the substrate holder 223 when calculating the parallax may be the design origin of the substrate holder 223, the convex apex of the substrate holder 223, or the observation origin of the substrate holder 223. The distance from the convex center of the substrate holder 223 may be measured along the convex or free-form surface shape, or may be measured linearly along the XY coordinates of the device. For each of the above, the measurement may be based only on the design coordinates, and the alignment results may be ignored.

[0052] The alignment mark 218, which is measured by the microscope 324 as being at a position R [μm] from the center position of the substrate holder 223, is aligned with the coordinate S along the deformed substrate holder 223 when the substrates 211 and 213 are bonded. _(Z_WH) (R) [mm]. The substrate holder 223 is displaced to the position Z _WH In the following calculation, the diameter of the substrate holder 223 is set to 300,000 [μm]. First, the coordinates S _(Z_WH) (R) [mm] is calculated. When the shape of the substrate holder 223 is a parabola, the vertical height Z is expressed by the following formula 1. Note that, although the analysis is performed by approximating the shape of the substrate holder 223 to a parabola in this embodiment, the analysis may be performed by approximating the shape of the substrate holder 223 to another curve (e.g., a circular arc).

number

[0053] In Equation 1, M is the magnification [ppm], and 2t is the thickness [μm] of the object to be deformed. _WH is the value of Z when R = 150000 [μm], but when M>0, Z _WH Since it is defined as >0, the sign must be inverted. _WH When expressed as:

number

[0054] By substituting Equation 2 into Equation 1, we obtain Equation 3 below.

number

[0055] Coordinate S of the alignment mark 218 when it is joined _(Z_WH) (R) is expressed as the integral shown in Equation 4 below.

number

[0056] The differentiation within the radical sign is expressed by the following number 5.

number

[0057] Here, as shown in the following formula 6, a _(Z_WH) is the amount of convexity Z _WH It is a constant determined in [μm].

number

[0058] The integral of Equation 4 is expressed by Equation 7 below.

number

[0059] By subjecting the root sign and logarithm in Equation 7 to a Taylor expansion and leaving only the terms up to the third order of R, we obtain Equation 8 below.

number

[0060] S _(Z_WH) (R) to parallax-induced error dR para When this is calculated, the following Equation 9 is obtained.

number

[0061] In Equation 9, R is the distance from the center, so when it is expressed using orthogonal coordinates, the following Equation 10 is obtained.

number

[0062] Here, the design coordinates (design coordinates) of the alignment mark 218, the deviation (virtual deviation) of the alignment mark 218 from the design coordinates when the substrate holder 223 is in a flat state, the parallax amount (parallax), and the deviation (actual deviation) of the alignment mark 218 from the design coordinates including the actually measured parallax amount are defined as shown in the following formula 11.

number

[0063] The sum of the coordinates of the moving stage 332 during measurement and the virtual displacement is expressed by the following formula 12. This is also called the virtual displacement.

number

[0064] Similarly, the sum of the coordinates of the moving stage 332 during measurement and the actual measurement deviation is expressed by the following formula 13. This is also called the actual measurement deviation.

number

[0065] The relationship between the quantities in Equations 11, 12, and 13 above can be expressed as follows:

number

[0066] From the above formulas 10 and 13, the parallax amount (dx i para , dy i para ) is the actual measurement deviation (x i WF+Meas , y i WF+Meas ) is given by the following Equation 15.

number

[0067] As described above, a in the above formula 15 _(Z_WH) is the amount of convexity Z _WH It is a constant determined in [μm].

number

[0068] After parallax correction, the virtual displacement of the alignment mark 218 (x i WF+Mark ,y i WF+Mark ) is expressed by the following Equation 17 from Equation 14 and Equation 15.

number

[0069] Virtual deviation of alignment mark 218 (x iWF+Mark ,y i WF+Mark ) to calculate the shift, rotation, magnification, and orthogonality of the EGA parameters. i WF ,y i WF ) [μm], the shift (ε x ,ε y )[μm], rotation θz[rad], magnification (γ x ,γ y ) [ppm], and the orthogonality ω [rad], (x i WF+Mark ,y i WF+Mark ) is expressed by the following Equation 18.

number

[0070] (ε x ,ε y ,θ z ,γ x ,γ y , ω) are constants, the following formulas 19 and 20 hold when n alignment marks 218 are measured.

number

number

[0071] In the above formula 20, the pseudo inverse matrix B + (=(B T B) (-1) B T ) is applied to obtain the following formula 21. Note that in this embodiment, the pseudoinverse matrix is ​​directly calculated when solving the least squares method, but a method using singular value decomposition may be used, or another algorithm that analytically or numerically solves the least squares method may be used.

number

[0072] As described above, each parameter of EGA is derived from the displacement amount δxy in the XY direction caused by the curvature of the substrate holder 223, and the movement amount of the moving stage 332 is calculated (step S04). The control unit 150 may calculate the movement amount of the moving stage 332 so that the substrate displacement amount between the substrates 211 and 213 becomes zero based on the displacement amount δxy, or, differently, may calculate the movement amount of the moving stage 332 so that the substrate displacement amount between the substrates 211 and 213 is within a predetermined threshold based on the displacement amount δxy. The threshold may be a displacement amount at which electrical conduction is possible between the substrates 211 and 213 when the bonding of the substrates 211 and 213 is completed, a displacement amount at which structures provided on the substrates 211 and 213 are at least partially in contact with each other, or a displacement amount at which a predetermined bonding strength is obtained between the connection structures. Although an example has been shown in which the movement amount of the moving stage 332 is calculated based on the displacement amount δxy, the movement amount may be calculated using the actually measured displacement amount between the bonded substrates 211, 213 in addition to or instead of the displacement amount δxy.

[0073] Referring again to Fig. 5, next, the control unit 150 aligns the substrates 211, 213 by moving the moving stage 332 based on the amount of movement calculated in step S04 (step S05), as shown in Fig. 10. Fig. 11 is a diagram showing the state of the substrates 211, 213 and the substrate holders 221, 223 in the aligned state.

[0074] Referring again to Fig. 5, next, as shown in Fig. 12, the control unit 150 causes the Z-direction driving unit 335 to raise the moving stage 332, brings the aligned substrates 211, 213 into contact with each other, and starts bonding the substrates 211, 213 (step S06). Fig. 13 is a diagram showing the state of the substrates 211, 213 and the substrate holders 221, 223 when bonding has started.

[0075] 13, at the time of contact in step S07, one flat substrate 211 and the other curved substrate 213 are partially in contact with each other. As a result, a starting point of partial bonding of the substrates 211 and 213 is formed approximately at the center of the substrates 211 and 213, as shown by the dotted line C in the figure.

[0076] 14, after portions of the substrates 211, 213 come into contact with each other, the control unit 150 releases the substrate 211 from the substrate holder 221 on the fixed stage 322. The substrate 211 on the upper side in the figure, which is now free, autonomously expands the bonded area due to its own weight and the intermolecular forces of the activated substrates 211, 213, and eventually the entire surfaces are bonded together. In this way, the stack 230 of the substrates 211, 213 is formed in the stacking unit 300.

[0077] In stack 230, substrate 213 on the lower side in the figure continues to be held by substrate holder 223 with curved holding surface 225 throughout the stacking process from step S103 onwards. Therefore, since substrate 213 is joined to substrate 211 in a corrected state by substrate holder 223, the magnification difference between substrates 211 and 213 is corrected.

[0078] In addition, in the process of expanding the contact area of ​​the substrates 211, 213 as described above, the control unit 150 may release a part or all of the substrate 213 from the substrate holder 223. Also, the substrate holder 223 may be released from its hold by the fixed stage 322. When the substrate 213 is released from its hold, the lower substrate 213 is lifted up from the substrate holder 223 and curved due to the pulling force from the upper substrate 211 in the process of expanding the contact area. As a result, the shape of the surface of the lower substrate 213 is changed so as to stretch, and the difference with the amount of stretch of the surface of the upper substrate 211 is reduced by the amount of this stretch.

[0079] This suppresses misalignment caused by different amounts of deformation between the two substrates 211, 213. Since the amount of lift of the substrate 213 from the substrate holder 223 can be adjusted by adjusting the holding force of the substrate holder 223, if a difference occurs between the amount of correction preset for the multiple substrate holders 223 and the amount of correction actually required, the difference can be compensated for by adjusting the holding force of the substrate holder 223.

[0080] Furthermore, the bonding of the substrates 211, 213 may be advanced by releasing the substrate 213 held by the movable stage 332 without releasing the substrate 211 held by the fixed stage 322. Furthermore, the substrates 211, 213 may be bonded by moving the fixed stage 322 and the movable stage 332 closer to each other while the substrates 211, 213 are held by both the fixed stage 322 and the movable stage 332.

[0081] The stack 230 thus formed is carried out from the stacking section 300 by the transport section 140 (step S07) and stored in the substrate cassette .

[0082] [evaluation] The amount of convexity of substrate holder 223 was set to a predetermined value, the diameter of substrate holder 223 was set to 300,000 [μm], and four alignment marks 218 with coordinates of (+100000, +100000), (+80000, -80000), (-90000, -100000), and (-110000, +90000) were measured to calculate the parallax, rotation, magnification change, and orthogonality at each point when parallax correction was performed and when parallax correction was not performed.

[0083] When parallax correction was not performed, deviations from the set values ​​occurred for parallax, rotation, magnification change, and orthogonality, and magnification change in particular deviated significantly from the set value. Therefore, it was found that deviations occurred in each parameter when using the substrate holder 223 with a convex shape and parallax correction was not performed. In contrast, when parallax correction was performed, the deviations were eliminated and the values ​​were close to the set values.

[0084] As described above, according to the first embodiment, each parameter of EGA is corrected from the deviation amount δxy in the XY directions caused by the curvature of substrate holder 223. This makes it possible to perform EGA based on the ideal position of alignment mark 218 when substrate 213 is in a flat state, and makes it possible to perform highly accurate EGA even when substrate holder 223 with a convex shape is used.

[0085] [Second embodiment] In the above first embodiment, each parameter of the EGA is derived from the deviation amount δxy based on the parallax in the XY directions of the alignment mark 218 due to the curvature of the substrate holder 223. However, the curvature of the substrate holder 223 generates not only the parallax in the XY directions of the alignment mark 218 but also the parallax in the Z direction.

[0086] In the second embodiment, an embodiment will be described in which the tilt angle of the substrate holder 223 is adjusted based on the parallax in the Z direction. Figures 15(a) to (c) are diagrams showing the parallax in the Z direction of the alignment mark 218 caused by the curvature of the substrate holder 223. The Z direction is a direction perpendicular to the XY plane.

[0087] As shown in FIG. 15(a), when the substrate 213 is flat, the alignment mark 218 is observed at a position F3 in the Z direction. The position of the alignment mark 218 in the Z direction can be measured by measuring the focus position of the alignment mark 218 with the microscope 324. As shown in FIG. 15(b), when the substrate 213 is held by the substrate holder 223 having a convex shape and curved, the alignment mark 218 is observed at a position F4 in the Z direction that is moved by δz from the position F3 in the Z direction. Therefore, a deviation amount δz (parallax) is generated in the observation of the alignment mark 218 in the Z direction. Therefore, as shown in FIG. 15(c), it is desirable to virtually return the substrate holder 223 to a flat surface and adjust the difference between the positions F3 and F4.

[0088] Similar to the parallax in the XY directions of alignment mark 218 described above, the parallax in the Z direction of alignment mark 218 is determined according to the distance of alignment mark 218 from the convex center of substrate holder 223 and the convex amount of substrate holder 223. Hereinafter, a method of deriving deviation amount δz in the Z direction using the distance of alignment mark 218 from the convex center of substrate holder 223 and the convex amount of substrate holder 223 will be described.

[0089] The misalignment amount (virtual misalignment) of the alignment mark 218 when the substrate holder 223 is in a flat state, the parallax amount (parallax), and the misalignment amount including the parallax amount actually measured by focus measurement of the alignment mark 218 (actual misalignment) are defined as shown in the following equation 22.

number

number

number

number

number

number

number

[0090] The control unit 150 can adjust the tilt angle of the moving stage 332 using the obtained values ​​of the tilt amounts θx [rad], θy [rad] and the shift amount εz [μm] of the moving stage 332. The tilt angle of the moving stage 332 is adjusted by driving the swing drive unit that swings the moving stage 332. As described above, the tilt angle of the moving stage 332 can be adjusted using the deviation amount δz [μm] of the alignment mark 218 in the Z direction.

[0091] As described above, according to the second embodiment, the tilt angle of the moving stage 332 can be adjusted from the deviation amount Δz of the alignment mark 218 in the Z direction caused by the curvature of the substrate holder 223.

[0092] [Variation 1] In the above embodiment, the shape of substrate holder 223 is assumed to be a convex shape, and the deviation amount δxy in the XY directions and the deviation amount δz in the Z direction are calculated from the distance of alignment mark 218 from the center of the convexity of substrate holder 223 and the amount of convexity of substrate holder 223. However, the shape of substrate holder 223 does not have to be a convex shape, and may be, for example, a free curved surface. The processing in this case is as follows.

[0093] When a free-form surface is used for the shape of the substrate holder 223, the amount of deviation δxy in the XY directions is calculated as follows. The length of the curve along the radial direction from the center of the substrate 213 to the alignment mark 218 is calculated by analytical or numerical integration to derive the amount of deviation δxy in the XY directions. The derived amount of deviation δxy is multiplied by the amount of parallax (dx i para , dy i para ) is substituted. The subsequent processing is the same as that in Equation 16 and subsequent steps. Note that the shape of the substrate 213 used in the above processing may be either a designed shape or a shape based on actual measurements.

[0094] When a free-form surface is used for the shape of the substrate holder 223, the displacement amount δz in the Z direction is calculated as follows. The difference in height (displacement amount δz) in the Z direction from the flat state of the alignment mark 218 of the substrate 213 is derived. The derived displacement amount δz is multiplied by the parallax amount dz in the above formula 24. i para Subsequent processing is the same as that in and after Expression 25. Note that the shape of the substrate 213 used in the above processing may be either a designed shape or a shape based on actual measurement.

[0095] As an example of a free-form surface, a shape obtained by deforming the substrates 211, 213 so that the heights of the substrates 211, 213 in the 45-degree, 135-degree, 225-degree, and 315-degree directions are different from the heights in other angular directions may be used. In this case, the substrates 211, 213 may be deformed so that the heights of the substrates 211, 213 in the 45-degree, 135-degree, 225-degree, and 315-degree directions are equal, and the heights in the other angular directions are equal.

[0096] [Variation 2] In the above embodiment, the upper substrate holder 221 is a flat holder, and the lower substrate holder 223 is a convex holder. However, the present invention is applicable to cases where either of the substrate holders 221, 223 is a convex holder. Therefore, for example, the upper substrate holder 221 may be a convex holder and the lower substrate holder 223 may be a flat holder, or both the upper substrate holder 221 and the lower substrate holder 223 may be convex holders.

[0097] [Variation 3] In the above embodiment, the substrates 211, 213 are adsorbed and held by the substrate holders 221, 223. However, a stage having a convex shape may hold the substrates 211, 213 without using the substrate holders 221, 223. In this case, the amount of deviation in the XY directions and the Z direction may be calculated based on the shape of the stage.

[0098] [Variation 4] In the first embodiment, the control unit 150 calculated the amount of deviation δxy of the alignment mark 218 in the XY directions to control the moving stage 332. In the second embodiment, the control unit 150 adjusted the tilt angle of the moving stage 332 from the amount of deviation δz of the alignment mark 218 in the Z direction caused by the curvature of the substrate holder 223. However, it is possible to combine the first and second embodiments.

[0099] [Variation 5] Also, a relational equation for calculating the amount of deviation may be stored in advance in the control unit 150. Furthermore, instead of the relational equation for calculating the amount of deviation, a lookup table of numerical values ​​may be stored in advance in the control unit 150, and the control unit 150 may calculate the amount of deviation by referring to the lookup table.

[0100] [Variation 6] In the above first embodiment, the misalignment amount δxy is derived using the distance of alignment mark 218 from the convex center of substrate holder 223 and the convex amount of substrate holder 223. However, the misalignment amount δxy may be derived using the radius of curvature of substrate holder 223 instead of the convex amount of substrate holder 223. In this case, since the radius of curvature of substrate holder 223 can be approximated using the convex amount of substrate holder 223, the radius of curvature of substrate holder 223 may be derived based on the convex amount of substrate holder 223. Also in this case, instead of the distance of alignment mark 218 from the convex center of substrate holder 223, the coordinates of alignment mark 218 in a coordinate system with the convex center of substrate holder 223 as the origin may be used.

[0101] [Variation 7] Furthermore, in the statistical processing by the enhanced global alignment method (EGA), the EGA may be performed by selecting alignment marks 218 in which the gradient, parallax amount, or height of the holding surface 225 of the substrate holder 223 is at the same position (or the difference is within a predetermined range), or alignment marks 218 in symmetric positions with respect to the centers of the substrates 211 and 213. This can shorten the time required to calculate the correction amount.

[0102] [Variation 8] In addition, in the above XY parallax correction, a pair of alignment marks 218 having the same parallax amount but opposite orientations may be selected and measured, and in the above Z parallax correction, alignment marks 218 having the same Z parallax amount may be selected and measured.

[0103] [Variation 9] In the above first embodiment, the amount of convexity of substrate holder 223 is used to derive the amount of deviation δxy. However, instead of the amount of convexity of substrate holder 223, information on other convex shapes such as the curvature of substrate holder 223, the height of the central part relative to the outer edge, a 3D model of the shape of substrate holder 223, etc. may be used.

[0104] Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It is clear to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the description of the claims that such modifications and improvements can also be included in the technical scope of the present invention.

[0105] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and may be realized in any order unless the output of a previous process is used in a later process. Even if the operational flow in the claims, specifications, and drawings is explained using "first," "next," etc. for convenience, it does not mean that it is essential to perform the process in this order. [Explanation of symbols]

[0106] 100, stacking device, 110 housing, 120, 130 substrate cassette, 140 transport unit, 150 control unit, 211, 213 substrate, 212 scribe line, 214 notch, 216 circuit area, 218 alignment mark, 221, 223 substrate holder, 225 holding surface, 230 stacked body, 300 stacking unit, 301 floor surface, 310 frame, 312 bottom plate, 314 support, 316 top plate, 322 fixed stage, 324, 334 microscope, 331 X-direction drive unit, 332 moving stage, 333 Y-direction drive unit, 335 Z-direction drive unit, 400 holder stocker, 500 pre-aligner, 600 activation device

Claims

1. a first stage that holds a first substrate; a second stage that curves and holds the second substrate; an observation unit that observes an alignment mark provided on the second substrate that is held in a curved state by the second stage, A stacking device that controls at least one of the first stage while holding the first substrate and the second stage while holding the second substrate based on the position of the alignment mark observed by the observation unit and information regarding the curvature of the second substrate held in a curved state on the second stage.

2. 2. The stacking device according to claim 1, wherein the information includes a deviation between a position of the alignment mark when the surface of the second substrate is flat and a position of the alignment mark provided on the second substrate held in a curved state by the second stage.

3. The stacking device according to claim 2 , wherein the amount of deviation includes a first amount of deviation in a direction parallel to the plane.

4. The stacking device according to claim 1 , wherein the second stage includes a holding member that holds the second substrate by a curved holding surface.

5. The stacking device according to claim 4 , wherein the information includes at least one of information on a shape of the holding member and information on a distance of the alignment mark from a reference position of the holding member and information on the position of the alignment mark.

6. The stacking device of claim 5 , wherein the information regarding the shape of the holding member includes at least one of a radius of curvature of the holding member, a curvature of the holding member, a height of a center portion relative to an outer edge portion of the holding member, and a 3D model of the shape of the holding member.

7. The stacking device according to claim 2 , wherein the amount of deviation includes a second amount of deviation in a direction perpendicular to the plane.

8. The stacking device according to claim 7 , further comprising: a tilt angle of at least one of the first stage and the second stage being controlled based on the second amount of deviation.

9. the second stage includes a holding member that holds the second substrate by a curved holding surface, 9. The stacking device according to claim 8, wherein the second deviation amount is set based on at least one of a radius of curvature and a curvature of the second substrate, and at least one of a distance of the alignment mark from a center of the holding member and a coordinate of the alignment mark in a coordinate system having an origin at the center of the holding member.

10. The stacking device according to claim 1 , further comprising: a parameter used for aligning the first substrate and the second substrate, the parameter being set based on the amount of deviation.

11. The stacking device according to claim 2 , further comprising: a control unit configured to control at least one of the first stage and the second stage so that the amount of deviation is within a predetermined threshold value.

12. 10. The stacking device according to claim 1, wherein the controlling includes controlling at least one of a position of the first stage when holding the first substrate and a position of the second stage when holding the second substrate based on the position and the information.

13. holding a first substrate on a first stage; bending the second substrate and holding it on a second stage; Observing an alignment mark provided on the second substrate held in a curved state by the second stage; A method for controlling at least one of the first stage while holding the first substrate and the second stage while holding the second substrate based on the position of the alignment mark observed by the observation and information regarding the curvature of the second substrate held in a curved state on the second stage.

14. holding a first substrate on a first stage; bending the second substrate and holding it on a second stage; Observing an alignment mark provided on the second substrate held in a curved state by the second stage; performing processing by an enhanced global alignment method based on the position of the alignment mark observed by the observation and information regarding the curvature of the second substrate held in a curved state on the second stage; method.

15. The method according to claim 13 or 14, wherein the information includes a deviation between a position of the alignment mark when the surface of the second substrate is flat and a position of the alignment mark provided on the second substrate held in a curved state by the second stage.

16. The method of claim 15 , wherein the offset comprises a first offset in a direction parallel to the plane.

17. The method of claim 15 , wherein the offset includes a second offset in a direction perpendicular to the plane.

18. the deviation amount includes a second deviation amount in a direction perpendicular to the plane, The method of claim 15 , wherein the controlling comprises controlling a tilt angle of at least one of the first stage and the second stage based on the second amount of deviation.

Images