Method for processing wafer units, and method for manufacturing semiconductor chips.

Abstract

This suppresses misalignment of the modified layer formed on each wafer of the bonded wafer. [Solution] In the processing of a wafer unit (30) formed by joining a surface (13) of a first wafer (10) containing a first pattern (12) and a surface (23) of a second wafer (20) containing a second pattern (22), the following steps are performed: a first alignment step in which the first pattern (12) or the second pattern (22) is imaged by an imaging unit (42) to identify a first planned division line (17) which will be the boundary of the division of the first wafer; a first modified layer formation step in which a modified layer (18) is formed on the first wafer along the first planned division line; a second alignment step in which the modified layer of the first wafer is imaged by an imaging unit to identify a second planned division line (27) which will be the boundary of the division of the second wafer; and a second modified layer formation step in which a modified layer (28) is formed on the second wafer along the second planned division line.

Description

【Technical Field】 【0001】 The present invention relates to a method for processing a wafer unit and a method for manufacturing a semiconductor chip. 【Background Art】 【0002】 As a method for dividing a semiconductor wafer into device chips (semiconductor chips), a method is known in which a modified layer is formed inside the wafer by irradiating a laser beam along a planned division line and the wafer is divided starting from this modified layer (see, for example, Patent Document 1). 【Prior Art Documents】 【Patent Documents】 【0003】 【Patent Document 1】 Japanese Patent No. 3408805 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 Bonded wafers in which two or more wafers are bonded together are known. When applying the above-described processing method when dividing a bonded wafer, a modified layer may be formed in a plurality of steps separately for each of the bonded wafers. When the modified layer is formed in a plurality of steps, there is a risk that the positions of the modified layers formed on each wafer will shift. If the positions of the modified layers of each wafer shift, there is a problem that the quality of the diced device chips deteriorates. 【0005】 In order to prevent such problems, there is a problem of suppressing the positional deviation of the modified layers formed on each of the wafers constituting a bonded wafer (wafer unit) in which two or more wafers are bonded together. 【Means for Solving the Problems】 【0006】 One aspect of the present disclosure is a wafer unit processing method for processing a wafer unit in which the first surface and the third surface of a first wafer are joined together, the first wafer having a first pattern formed on a first base wafer having a first street region including a first division planned line which serves as a boundary for division when fragmented, and the first wafer having a first surface including the first pattern and a second surface opposite to the first surface, and the second wafer having a second pattern formed on a second base wafer having a second street region, and the third surface including the second pattern and a fourth surface opposite to the third surface. The wafer unit processing method comprises: a first alignment step of imaging the first pattern or the second pattern with light of a wavelength that passes through the first base wafer and the second base wafer using an imaging unit to identify the first planned division line; a first modified layer formation step of forming a modified layer on the first wafer along the first planned division line identified by the first alignment step; a second alignment step of imaging the modified layer of the first wafer formed in the first modified layer formation step using the imaging unit to identify the second planned division line that will be the boundary of the division of the second wafer; and a second modified layer formation step of forming a modified layer on the second wafer along the second planned division line identified by the second alignment step. 【0007】 One form of the second alignment process includes: a first focusing step in which the imaging unit focuses on the second pattern of the second wafer; a focus area setting step in which the imaging unit images the second pattern with the second pattern in focus in the first focusing step and sets a focus area smaller than the second street area in the second street area within the imaging area of ​​the imaging unit; and a second focusing step in which focusing is performed inside the focus area set in the focus area setting step. This makes it possible to image the modified layer of the first wafer formed in the first modified layer formation step with the imaging unit and to identify the second planned division line based on the modified layer of the first wafer. 【0008】 The focus area setting step may further include a pattern identification step of identifying the position of a predetermined shape included in the second pattern from an image captured of the second pattern. In this case, based on the position of the predetermined shape identified by the pattern identification step, a focus area smaller than the second street area can be set in the second street area within the imaging area of ​​the imaging unit. 【0009】 One form of the second alignment process includes a first focusing step in which the imaging unit focuses on the second pattern on the second wafer, and a focus shifting step in which the focus position of the imaging unit, which was aligned with the second pattern in the first focusing step, is moved by a predetermined distance. This makes it possible to image the modified layer of the first wafer formed in the first modified layer formation step with the imaging unit and to identify the second planned division line based on the modified layer of the first wafer. 【0010】 Preferably, in the first alignment step, the imaging unit images the first pattern or the second pattern from the second surface side of the wafer unit; in the first modification layer formation step, a laser beam is incident from the second surface side of the wafer unit to form the modification layer; in the second alignment step, the imaging unit images the modification layer of the first wafer from the fourth surface side of the wafer unit; and in the second modification layer formation step, a laser beam is incident from the fourth surface side of the wafer unit to form the modification layer. 【0011】 The wafer unit may further include a ring frame having an opening in the center, and tapes that are fixed to the second and / or fourth surface of the wafer unit and the ring frame. 【0012】 The process may further include a quality confirmation step of imaging the modified layer formed on the first wafer to confirm the quality of the modified layer. 【0013】 One aspect of the present disclosure is a method for manufacturing a semiconductor chip, comprising: a first wafer having a first street region including a first division line which serves as a boundary for division during fragmentation, formed on a first base wafer made of semiconductor, and having a first surface including the first pattern and a second surface opposite to the first surface; and a second wafer having a second pattern having a second street region formed on a second base wafer made of semiconductor, and having a third surface including the second pattern and a fourth surface opposite to the third surface, wherein the first surface and the third surface of the wafer unit are joined together and the wafer unit is divided along the first division line. The method for manufacturing this semiconductor chip comprises: a first alignment step in which an imaging unit images the first pattern or the second pattern with light of a wavelength that penetrates the first base wafer and the second base wafer to identify the first planned division line; a first modified layer formation step in which a modified layer is formed on the first wafer along the first planned division line identified by the first alignment step; a second alignment step in which the imaging unit images the modified layer of the first wafer formed in the first modified layer formation step to identify the second planned division line which will be the boundary of the division of the second wafer; a second modified layer formation step in which a modified layer is formed on the second wafer along the second planned division line identified by the second alignment step; and a division step in which, after the second modified layer formation step, an external force is applied to the wafer unit to divide the wafer unit using the modified layer as the division starting point. [Effects of the Invention] 【0014】 According to this disclosure, it is possible to suppress misalignment of the modified layer formed on each wafer constituting the bonded wafer (wafer unit). [Brief explanation of the drawing] 【0015】 [Figure 1] This diagram shows the configuration of a wafer unit. [Figure 2] This diagram shows the initial holding process. [Figure 3] This is a diagram showing the first alignment process. [Figure 4]It is a diagram showing the first modification layer formation step. [Figure 5] It is a diagram showing the second alignment step. [Figure 6] It is a diagram showing the pattern identification step in the second alignment step. [Figure 7] It is a diagram showing the second modification layer formation step. [Figure 8] It is a diagram showing the dicing step. [Figure 9] It is a diagram showing the dicing step. 【BEST MODE FOR CARRYING OUT THE INVENTION】 【0016】 The wafer unit processed by the processing method of the present disclosure is a bonded wafer in which two or more wafers are bonded. The bonded wafer may be referred to as a bonding substrate or the like. A wafer is a plate-shaped or sheet-shaped workpiece, and the direction of its thickness is defined as the thickness direction. The bonded wafer is formed by bonding a plurality of wafers laminated in the thickness direction. As shown in FIG. 1, in the present embodiment, a wafer unit 30 is formed by laminating and bonding two wafers, a first wafer 10 and a second wafer 20, in the thickness direction. For convenience of explanation, the thickness of the wafer unit 30 and the like are exaggeratedly shown in the drawings. 【0017】 The first wafer 10 includes a disk-shaped first base wafer 11 made of a semiconductor such as silicon, and a first pattern 12 is formed on one side of the first base wafer 11 in the thickness direction. The second wafer 20 includes a disk-shaped second base wafer 21 made of a semiconductor such as silicon, and a second pattern 22 is formed on one side of the second base wafer 21 in the thickness direction. In the first wafer 10, the surface including the first pattern 12 is defined as the first surface 13, and the surface opposite to the first surface 13 is defined as the second surface 14. In the second wafer 20, the surface including the second pattern 22 is defined as the third surface 23, and the surface opposite to the third surface 23 is defined as the fourth surface 24. 【0018】 The wafer unit 30 is formed by facing and joining the first surface 13 of the first wafer 10 and the third surface 23 of the second wafer 20. That is, the second surface 14 of the first wafer 10 constitutes one surface (outer surface) in the thickness direction of the wafer unit 30, and the fourth surface 24 of the second wafer 20 constitutes the other surface (outer surface) in the thickness direction of the wafer unit 30. FIG. 1 separately shows the state of the wafer unit 30 in which the first wafer 10 and the second wafer 20 are joined as viewed from the side on the outer peripheral side, the state of the first wafer 10 (before joining) as viewed from the front on the first surface 13 side, and the state of the second wafer 20 (before joining) as viewed from the front on the third surface 23 side. 【0019】 As shown in FIG. 1, the first pattern 12 in the first wafer 10 has a first street region 15 formed in a grid pattern and a plurality of rectangular device regions 16 partitioned by the respective first street regions 15. Finally, by dividing the first wafer 10 along the first street region 15, the ranges of the plurality of device regions 16 are fragmented into individual semiconductor chips (device chips). The first street region 15 includes a first planned division line 17 that becomes a division boundary when the device regions 16 are fragmented. The first planned division line 17 is specified by a first alignment process described later. Generally, the planned division line of a wafer is set at the center in the width direction of the street. However, when the first planned division line 17 is specified based on the second pattern 22 of the second wafer 20 in the first alignment process described later, the first planned division line 17 may be set at a position slightly deviated from the center in the width direction of the first street region 15 within a range that does not hinder the division of the wafer unit 30. 【0020】 As shown in Figure 1, the second pattern 22 on the second wafer 20 has a grid-like second street region 25 and a plurality of rectangular device regions 26 separated by each of the second street regions 25. Finally, by dividing the second wafer 20 along the second street region 25, the range of the plurality of device regions 26 is fragmented into individual semiconductor chips (device chips). The second street region 25 includes a second division line 27 that serves as the boundary for division when fragmenting the device regions 26. The second division line 27 is identified by a second alignment process described later. In the second alignment process, the second division line 27 is identified based on the modified layer 18 formed on the first wafer 10 side. Therefore, the second division line 27 may be set at a position slightly off-center in the width direction of the second street region 25, within a range that does not interfere with the division of the wafer unit 30. 【0021】 The first wafer 10 and the second wafer 20 have the same diameter and material as the first base wafer 11 and the second base wafer 21, and the configurations of the first pattern 12 and the second pattern 22 are also identical. In other words, the configuration (number, dimensions, arrangement, etc.) of the first street region 15 and the device region 16 in the first wafer 10 is the same as the configuration (number, dimensions, arrangement, etc.) of the second street region 25 and the device region 26 in the second wafer 20. When forming the wafer unit 30, the first wafer 10 and the second wafer 20 are aligned so that the relative positions of their first street region 15 and second street region 25 and their relative positions of their device region 16 and device region 26 coincide when facing each other, and their first surface 13 and third surface 23 are joined together. Therefore, the wafer unit 30 is configured such that the first street region 15 and the device region 16 overlap in the thickness direction of the wafer unit 30, and the second street region 25 and the device region 26 overlap. 【0022】 The processing method of this disclosure involves processing a wafer unit 30, which is formed by joining a first wafer 10 and a second wafer 20, at positions along the first street region 15 and the second street region 25. More specifically, modified layers 18 (see Figure 4) and 28 (see Figure 7), which serve as the starting points for splitting, are formed inside the thickness direction of the first base wafer 11 and the second base wafer 21, respectively. Figures 4, 5, and 7 show a state in which multiple modified layers 18 and multiple modified layers 28 are formed at predetermined intervals along a specific (one) first street region 15 and second street region 25 extending in the left-right direction in the figures. Similarly, multiple modified layers 18 and multiple modified layers 28 are formed in all other regions along the first street region 15 and second street region 25. Modification in the modified layer means that the density, refractive index, mechanical strength, and other physical properties inside the first wafer 10 and the second wafer 20 become different from the surrounding areas. As a result of the modification, the area where the modified layer is formed becomes a region with lower strength than the surrounding area. When a predetermined external force is applied to the wafer unit 30 with modified layers 18 and 28 formed on the first wafer 10 and the second wafer 20 respectively, the first wafer 10 and the second wafer 20 are separated into individual pieces starting from the modified layers 18 and 28. 【0023】 Figures 2 to 7 show the schematic configuration of a laser processing apparatus 40, which is a processing apparatus used to form a modified layer on a wafer unit 30, and the various processes performed in the laser processing apparatus 40. The laser processing apparatus 40 includes a laser irradiation unit 41 that irradiates the wafer unit 30 with a laser beam Lz (Figures 4 and 7), an imaging unit 42 that images the wafer unit 30, and a holding unit 43 that holds the wafer unit 30. The laser irradiation unit 41 is used in the process of forming the modified layers 18 and 28 (first modified layer formation process and second modified layer formation process). The imaging unit 42 is used in the process of identifying the first planned division line 17 and the second planned division line 27, which are the boundaries of the division (first alignment process and second alignment process), and for aligning the laser irradiation unit 41 with respect to the wafer unit 30 in the process of forming the modified layers 18 and 28. 【0024】 The holding section 43 has a holding surface 431 with an area on which the wafer unit 30 can be placed. By applying negative pressure to the holding surface 431 with the suction force of a suction mechanism (not shown), the wafer unit 30 can be held in place by suction. Around the holding surface 431, a frame holding section 432 is provided for holding the ring frame 32, which will be described later. The ring frame 32 can be fixedly held by a plurality of clamps 433 that can open and close relative to the frame holding section 432. The direction parallel to the holding surface 431 is defined as the horizontal direction, and the direction perpendicular to the holding surface 431 is defined as the vertical direction. 【0025】 The laser irradiation unit 41 and the holding unit 43 are relatively movable in the horizontal direction along the holding surface 431 by a horizontal drive mechanism 44. The horizontal drive mechanism 44 is composed of a well-known feed screw mechanism or an air cylinder mechanism. The horizontal drive mechanism 44 may move both the laser irradiation unit 41 and the holding unit 43, or it may move only one of the laser irradiation unit 41 or the holding unit 43. In Figures 2 to 5 and 7, the horizontal drive mechanism 44 is shown as moving the holding unit 43, but the horizontal drive mechanism 44 may also move the laser irradiation unit 41. With the wafer unit 30 held in the holding unit 43, a laser beam Lz is irradiated from the laser irradiation unit 41 toward the wafer unit 30, and by moving the laser irradiation unit 41 and the holding unit 43 relatively in the horizontal direction, laser processing can be performed on the wafer unit 30 at any position. For example, laser processing can be performed on areas along the first street region 15 of the first wafer 10 or the second street region 25 of the second wafer 20. 【0026】 The laser irradiation unit 41 includes a laser oscillator 411 that emits a laser beam Lz, which is a pulsed laser with a wavelength that is penetrating to the first base wafer 11 of the first wafer 10 and the second base wafer 21 of the second wafer 20; a mirror 413 that reflects the laser beam Lz from the laser oscillator 411 and guides it to the processing head 412; and a focusing lens 414 provided on the processing head 412 that focuses the laser beam Lz. The processing head 412 can be moved vertically using a lifting mechanism 45. The lifting mechanism 45 is composed of a well-known feed screw mechanism or an air cylinder mechanism. As the processing head 412 moves vertically, the position of the focal point of the laser beam Lz focused by the focusing lens 414 changes vertically (in the thickness direction of the wafer unit 30). Note that the position of the focal point of the laser beam Lz may be changed not by moving the processing head 412, but by the operation or adjustment of the optical elements provided in the laser irradiation unit 41. For example, the position of the focal point can be changed by adjusting a reflective liquid crystal element or the like, located upstream of the focusing lens 414 (on the laser oscillator 411 side) in the optical system of the laser irradiation unit 41, thereby changing the divergence angle of the reflected laser beam Lz. Alternatively, the position of the focal point can be changed by moving the lens, such as the focusing lens 414, along the optical axis. 【0027】 By using the horizontal drive mechanism 44 and the lifting mechanism 45 to move the laser irradiation unit 41 and the holding unit 43 relative to each other in the horizontal and vertical directions, the laser beam Lz can be focused at any position inside the wafer unit 30 held by the holding unit 43 to form a modified layer. 【0028】 The imaging unit 42 can image the first pattern 12 on the first wafer 10 and the second pattern 22 on the second wafer 20 using light (electromagnetic waves) of wavelengths that pass through the first base wafer 11 and the second base wafer 21. In other words, the image sensor in the imaging unit 42 has high light-receiving sensitivity to light of wavelengths that pass through the first base wafer 11 and the second base wafer 21. The light of wavelengths that pass through the first base wafer 11 and the second base wafer 21 is, for example, infrared light. In this case, the imaging unit 42 is an infrared camera equipped with an image sensor that has light-receiving sensitivity in the infrared wavelength band. Note that the wavelength of light that is transmitted will differ depending on the material and structure of the first base wafer 11 and the second base wafer 21. For example, visible light and ultraviolet light may pass through the first base wafer 11 and the second base wafer 21, and in these cases, a camera capable of imaging with light in the wavelength range that has such transmittance (such as a visible light camera or an ultraviolet camera) is used. 【0029】 The imaging unit 42 is capable of changing the focus position (focus position, focal point) of its optical system in the thickness direction of the wafer unit 30. For example, the optical system of the imaging unit 42 is a variable-focus optical system equipped with a focusing lens that moves in the direction of the optical axis, and the focus position is changed by moving the focusing lens. Alternatively, the optical system of the imaging unit 42 is a fixed-focus optical system with a constant focus position, and the focus position may be changed by moving the entire imaging unit 42, including the optical system and image sensor, in the direction of the optical axis. In both the type in which the imaging unit 42 is equipped with a variable-focus optical system and the type in which the imaging unit 42 equipped with a fixed-focus optical system is moved in the direction of the optical axis, the operation and control of changing the focus position of the optical system is defined as focusing, and the operation of performing focusing by automatic control is defined as autofocus. The imaging unit 42 used in this embodiment is capable of autofocus. 【0030】 The imaging unit 42 has an imaging range (field of view) that covers a portion of the wafer unit 30, allowing for precise imaging of fine patterns on the wafer unit 30. The imaging unit 42 is held so that its position relative to the laser irradiation unit 41 in the horizontal direction remains constant, and the horizontal drive mechanism 44 allows it to change its relative horizontal position to the holding unit 43 together with the laser irradiation unit 41. Therefore, in the first and second alignment processes described later, the horizontal position of the imaging unit 42 can be adjusted as needed to image any range of the wafer unit 30. It is also possible to use a wide-angle camera capable of imaging the entire wafer unit 30 as the imaging unit 42, and then process the captured image of the entire wafer unit 30 to enlarge and use a portion of it. 【0031】 The processing method of this disclosure sequentially performs a first alignment step, a first modified layer formation step, a second alignment step, and a second modified layer formation step on a wafer unit 30. First, an outline of the processing method including each of these steps will be described. Each of the steps described below is executed by the control of a control unit 46 provided in the laser processing apparatus 40. The control unit 46 includes an image processing unit 47 that processes images captured by the imaging unit 42. Various processing of the captured images in each of the steps described later is realized by the functions of the image processing unit 47. 【0032】 [First alignment process] In the first alignment step (Figure 3), the imaging unit 42 images the first pattern 12 or the second pattern 22 with light of a wavelength that passes through the first base wafer 11 and the second base wafer 21, and identifies the first planned division line 17 on the first wafer 10. The imaging unit 42 clearly images objects at the focus position of the optical system, and does not form a clear image of objects that are out of focus. Therefore, by setting the focus position to the position of the first pattern 12 in the thickness direction of the wafer unit 30, it is possible to capture an image in which the first pattern 12 is clearly visible, and by setting the focus position to the position of the second pattern 22 in the thickness direction of the wafer unit 30, it is possible to capture an image in which the second pattern 22 is clearly visible. In Figure 3, the focus position Fa of the imaging unit 42 is shown, where the focus is set to the position of the first pattern 12. In this case, an image in which the first pattern 12 is clearly visible can be obtained by imaging with the imaging unit 42. 【0033】 When an image clearly showing the first pattern 12 is acquired, the positional information (coordinate information) of the first street region 15 included in the first pattern 12 can be obtained from the image. Then, using the first street region 15 as a reference, the first planned division line 17, which will be the boundary of the division of the first wafer 10, is identified. For example, the first planned division line 17 is set at the center of the width direction of the first street region 15 in the captured image. When an image clearly showing the second pattern 22 is acquired, the positional information (coordinate information) of the second street region 25 included in the second pattern 22 can be acquired from the image. Then, using the second street region 25 as a reference, the first planned division line 17, which will be the boundary of the division of the first wafer 10, is identified. For example, the first planned division line 17 is set at the center of the width direction of the second street region 25 in the captured image. 【0034】 [First Modified Layer Formation Process] In the first modification layer formation step (Figure 4), a modification layer 18 is formed on the first wafer 10 along the first division line 17 identified in the first alignment step. The laser irradiation unit 41 sets a focal point for the laser beam Lz inside the first wafer 10 (first base wafer 11) in the thickness direction of the wafer unit 30 and irradiates it with the laser beam Lz. A modified layer 18 is formed in the area where the laser beam Lz is focused on the first base wafer 11. As a result of the modification, the modified layer 18 becomes a region with reduced strength compared to the surrounding area. The laser irradiation unit 41 and the holding unit 43 are moved relative to each other in the horizontal direction, and the laser beam Lz is focused at predetermined intervals at multiple positions along each of the first division lines 17 inside the first wafer 10 to form multiple modification layers 18. The focal point of the laser beam Lz is set at a different position in the thickness direction of the wafer unit 30 from the first pattern 12 (closer to the second surface 14 than the first pattern 12). Therefore, the modified layer 18 is formed at a depth position closer to the second surface 14 than to the first pattern 12. The first modified layer formation process is carried out until the modified layer 18 is formed along all of the first division lines 17 identified in the first alignment process. 【0035】 [Second alignment process] In the second alignment step (Figure 5), the imaging unit 42 images the modified layer 18 formed on the first wafer 10 with light of a wavelength that penetrates the first base wafer 11 and the second base wafer 21, and identifies the second planned division line 27 on the second wafer 20 using the modified layer 18 as a reference. By setting the focus position of the imaging unit 42 (shown as focus position Fb in Figure 5) to the position of the modified layer 18 in the thickness direction of the wafer unit 30, it is possible to capture an image in which the modified layer 18 is clearly visible. Since the modified layer 18 is formed at a depth position closer to the second surface 14 than the first pattern 12, when the focus of the imaging unit 42 is set to the modified layer 18, the positions of the first pattern 12 and the second pattern 22 become defocused, and an image in which the modified layer 18 is clearly visible can be obtained without the influence of the first pattern 12 and the second pattern 22. By processing images that clearly show the modified layer 18, accurate positional information (coordinate information) of the modified layer 18 can be obtained. Based on this positional information, the second planned division line 27, which will be the boundary for dividing the second wafer 20, is identified with the modified layer 18 as the reference. Specifically, the second planned division line 27 is set at a position that coincides with each modified layer 18 (a position that appears to overlap with the modified layer 18) in a plan view of the wafer unit 30. In other words, the second planned division line 27 is set by connecting the positions of multiple modified layers 18 formed along the first planned division line 17. 【0036】 [Second Modified Layer Formation Process] In the second modification layer formation process (Figure 7), a modification layer 28 is formed on the second wafer 20 along the second division line 27 identified in the second alignment process. The laser irradiation unit 41 sets a focal point for the laser beam Lz inside the second wafer 20 (second base wafer 21) in the thickness direction of the wafer unit 30 and irradiates it with the laser beam Lz. A modified layer 28 is formed on the second base wafer 21 in the area where the laser beam Lz is focused. As a result of the modification, the modified layer 28 becomes a region with reduced strength compared to the surrounding area. The laser irradiation unit 41 and the holding unit 43 are moved relative to each other in the horizontal direction, and the laser beam Lz is focused at predetermined intervals at multiple positions along each of the second division lines 27 inside the second wafer 20 to form multiple modification layers 28. The focal point of the laser beam Lz is set at a different position in the thickness direction of the wafer unit 30 from the second pattern 22 (closer to the fourth surface 24 than the second pattern 22). Therefore, the modified layer 28 is formed at a depth position closer to the fourth surface 24 than to the second pattern 22. The second modified layer formation process is carried out until the modified layer 28 is formed along all of the second division lines 27 set on the second wafer 20. 【0037】 In the wafer unit 30 processed through the above steps, the modified layer 18 of the first wafer 10 formed in the first modified layer formation step is imaged in the second alignment step, and the second planned division line 27 of the second wafer 20 is identified based on the modified layer 18. As a result, the modified layer 28 formed on the second wafer 20 along the second planned division line 27 in the second modified layer formation step has a high degree of positional agreement with the modified layer 18 on the first wafer 10 side in the width direction of the first street region 15 and the second street region 25, and the positional displacement with respect to the modified layer 18 is suppressed (no positional displacement or extremely small positional displacement). 【0038】 If the modified layer 18 on the first wafer 10 and the modified layer 28 on the second wafer 20 are misaligned by a predetermined amount or more in the width direction of the first street region 15 and the second street region 25, the misalignment between the region of the first wafer 10 that is easily divided by the modified layer 18 and the region of the second wafer 20 that is easily divided by the modified layer 28 will increase. This may result in poor division when applying external force to the wafer unit 30, potentially degrading the quality of the individual semiconductor chips. In contrast, by applying the processing method of this disclosure, the region of the first wafer 10 that is easily divided by the modified layer 18 and the region of the second wafer 20 that is easily divided by the modified layer 28 can be aligned, enabling good division of the wafer unit 30 and the manufacture of high-quality semiconductor chips. 【0039】 Since the wafer unit 30 is joined such that the position of the first street region 15 of the first wafer 10 and the position of the second street region 25 of the second wafer 20 coincide face to face, if the first wafer 10 is aligned with the first street region 15 as a reference to identify the first planned division line 17, and the second wafer 20 is aligned with the second street region 25 as a reference to identify the second planned division line 27, then theoretically, no positional misalignment will occur between the modified layer 18 formed along the first planned division line 17 and the modified layer 28 formed along the second planned division line 27. However, in reality, when joining the first surface 13 of the first wafer 10 and the third surface 23 of the second wafer 20, or when the wafer unit 30 is flipped upside down in a specific process described later, a slight positional misalignment may occur between the first wafer 10 and the second wafer 20. If the relative positions of the first street region 15 and the second street region 25 are significantly misaligned, the bonding of the wafer unit 30 itself will be judged as defective. However, if the misalignment between the first street region 15 and the second street region 25 is within a range that does not hinder the processing or division of the first wafer 10 and the second wafer 20, the wafer unit 30 may not be deemed defective and the modified layer may be formed. In such cases, if the first planned division line 17 and the second planned division line 27 are identified based on the first street region 15 and the second street region 25 of the first wafer 10 and the second wafer 20 respectively, there is a risk that the modified layer 18 and the modified layer 28 will be misaligned as described above. In contrast, by applying the processing method of this disclosure, even if the positional relationship between the opposing first street region 15 and second street region 25 is not perfectly aligned in the wafer unit 30 formed by joining the first wafer 10 and the second wafer 20, it becomes possible to align the positional relationship between the modified layer 18 and the modified layer 28 and perform good division starting from the modified layers 18 and 28, thereby improving the yield in semiconductor chip manufacturing. 【0040】 Next, a specific example of the wafer unit 30, including each of the processes described above, will be explained, mainly with reference to Figures 2 to 7. In this embodiment, the wafer unit 30 has a tape 31 and a ring frame 32 in addition to the bonded first wafer 10 and second wafer 20. The first wafer 10 and second wafer 20 are attached to the ring frame 32 via the tape 31 and then transported to the laser processing apparatus 40. The ring frame 32 is ring-shaped with an opening in the center, and the first wafer 10 and second wafer 20 are placed inside the central opening. At least one of the second surface 14 of the first wafer 10 and the fourth surface 24 of the second wafer 20 is fixed to the tape 31, and the outer peripheral portion of the tape 31 is fixed to the ring frame 32. In this embodiment, the tape 31 is fixed to the fourth surface 24 of the second wafer 20, and the second surface 14 of the first wafer 10 is exposed and not covered by the tape 31. 【0041】 The tape 31 may be of a type that is fixed to the wafer unit 30 and ring frame 32 by the adhesive force of an adhesive layer, or it may be of a type that does not have an adhesive layer and generates a bonding force (tack force) to the wafer unit 30 and ring frame 32 by temperature changes such as heating. The tape 31 is made of a material that transmits light of wavelengths that pass through the first base wafer 11 and the second base wafer 21, and transmits (does not absorb) the laser beam Lz irradiated from the laser irradiation unit 41. Therefore, it is possible to form a modified layer inside the wafer unit 30 (especially the second wafer 20) by injecting the laser beam Lz from the fourth surface 24 side of the second wafer 20 to which the tape 31 is attached, and to image the wafer unit 30 by the imaging unit 42 through the tape 31. 【0042】 [Initial holding process] Figure 2 shows the initial holding process in which the wafer unit 30, which has been transported to the laser processing apparatus 40, is first held in the holding unit 43. In the initial holding process, the wafer unit 30 is placed on the holding surface 431 of the holding unit 43 via the tape 31, with the tape 31 fixed to the fourth surface 24 of the second wafer 20 facing downwards. Negative pressure is applied to the holding surface 431 to hold the wafer unit 30 by suction. The ring frame 32 is placed on the frame holding unit 432 and the ring frame 32 is fixed by a plurality of clamps 433. The wafer unit 30 is held with the second surface 14 of the first wafer 10 facing upwards, and the processing head 412 and imaging unit 42 of the laser irradiation unit 41 are positioned opposite each other above the second surface 14. Following the initial holding process, the first alignment process is performed. 【0043】 [First alignment process] In the first alignment step shown in Figure 3, the imaging unit 42 is instructed to focus on the first pattern 12 of the first wafer 10 at the focus position Fa from the second surface 14 side of the wafer unit 30 and take an image. This acquires an image in which the first pattern 12 is clearly visible. The control unit 46 processes the image captured by the imaging unit 42 using the image processing unit 47 and identifies the first planned division line 17, which will be the boundary of the division of the first wafer 10, based on the first street region 15 included in the first pattern 12. For example, the first planned division line 17 is set as a line passing through the center of the width direction of the first street region 15 included in the first pattern 12. The identified first planned division line 17 is converted into data as horizontal coordinate information and stored in the memory of the control unit 46. 【0044】 Since the imaging unit 42 images with light of wavelengths that transmit through the first base wafer 11 and the second base wafer 21, the objects that the optical system of the imaging unit 42 can focus on (detect clear contrast) for the wafer unit 30 in its initial state without a modified layer formed are either the first pattern 12 formed on the first wafer 10 or the second pattern 22 formed on the second wafer 20. Generally, the autofocus of an imaging unit tends to prioritize focusing on the object closest to the imaging unit within the imaging area. Therefore, in the first alignment process, when the imaging unit 42 performs autofocus on the wafer unit 30 from the side of the second surface 14 of the first wafer 10, it is easier to focus on the first pattern 12, which is closer to the imaging unit 42 than the first pattern 12. For this reason, it is reasonable to set the focus position Fa of the imaging unit 42 to the first pattern 12 of the first wafer 10 and perform imaging, as shown in the first alignment process in Figure 3. Furthermore, in the subsequent first modified layer formation step, a modified layer 18 is formed inside the first wafer 10 (first base wafer 11) to which the first pattern 12 belongs. Therefore, in the first alignment step, by identifying the first planned division line 17 based on the first street region 15 included in the first pattern 12, it becomes easier to ensure the processing accuracy of the modified layer 18 relative to the first street region 15 in the first modified layer formation step. 【0045】 However, in the first alignment process, a configuration may be selected in which the focus position is set to the second pattern 22 of the second wafer 20 located on the side farther from the imaging unit 42 for imaging. After the imaging unit 42's autofocus initially focuses on the first pattern 12, the focus position is further changed in the thickness direction of the wafer unit 30 (the direction farther away from the imaging unit 42), so that the second pattern 22 is then in focus. By performing control to adopt this state as the focus position, imaging with the focus position set to the second pattern 22 can be performed. The control unit 46 then processes the image captured by the imaging unit 42 using the image processing unit 47 and identifies the first planned division line 17, which will be the boundary of the division of the first wafer 10, based on the second street region 25 included in the second pattern 22. For example, the first planned division line 17 is set as a line passing through the center in the width direction of the second street region 25. The identified first planned division line 17 is converted into data as horizontal coordinate information and stored in the memory of the control unit 46. In the second modification layer formation process, a modification layer 28 is formed inside the second wafer 20 (second base wafer 21) to which the second pattern 22 belongs. Therefore, in the first alignment process, by identifying the first planned division line 17 based on the second street region 25 included in the second pattern 22, it becomes easier to ensure the processing accuracy of the modification layer 28 relative to the second street region 25 in the second modification layer formation process. 【0046】 Thus, the configuration for imaging the first pattern 12 shown in Figure 3 (configuration in which the focus position Fa is set to the first pattern 12) is just one example, and in the first alignment step, either the first pattern 12 or the second pattern 22 may be selected as the target for imaging to adjust the focus position. 【0047】 [First Modified Layer Formation Process] In the first modified layer formation process shown in Figure 4, the lifting mechanism 45 is controlled so that the focal point of the laser beam Lz irradiated from the processing head 412 of the laser irradiation unit 41 is at a predetermined depth inside the first base wafer 11, and the operation of the horizontal drive mechanism 44 is controlled so that the position of the focal point of the laser beam Lz in the horizontal direction matches the first division line 17 identified in the first alignment process, thereby injecting the laser beam Lz from the second surface 14 side of the wafer unit 30 and forming a modified layer 18 inside the first base wafer 11. Multiple modified layers 18 are formed at predetermined intervals along each first division line 17. Once all modified layers 18 have been formed along all first division lines 17 identified in the first alignment process, the first modified layer formation process is completed. 【0048】 [Quality confirmation process] After the first modification layer formation process, a quality confirmation process may be performed to confirm the quality of the modified layer 18 formed on the first wafer 10 by imaging the modified layer 18. As shown in Figure 4, at the stage after the first modification layer formation process, the wafer unit 30 is positioned with the modified layer 18, the first pattern 12, and the second pattern 22 in order from the side closest to the imaging unit 42 in the thickness direction. The imaging unit 42 is in a state where it can easily focus on the modified layer 18, which is closer than the first pattern 12 and the second pattern 22. Therefore, an image in focus on the modified layer 18 can be obtained by imaging with the imaging unit 42. 【0049】 The control unit 46 determines the quality based on the image of the modified layer 18 that has been captured. The modified layer 18 is determined to meet the quality standard if multiple modified layers 18 are formed on their respective first division line 17 at appropriate intervals as set, and the size (width and diameter) of each modified layer 18 is as set. The modified layer 18 is determined to be of poor quality if it is formed at a distance greater than a predetermined distance from the first division line 17, and when a line is set to connect multiple modified layers 18, the line meanders more than a predetermined distance relative to the first division line 17. The modified layer 18 is also determined to be of poor quality if the spacing between the modified layers 18 is excessively wide in the direction along the first division line 17, causing breaks in the traces, or if the dimensions of the modified layer 18 as viewed from the direction of the imaging unit 42 are too large or too small. 【0050】 In the second alignment process, the second planned division line 27 is identified based on the modified layer 18. Therefore, if there are quality defects in the modified layer 18 as described above, there is a risk that quality defects such as meandering or breakage may occur in the modified layer 28 formed along the second planned division line 27. For this reason, if a quality defect in the modified layer 18 is discovered in the quality confirmation process, it is desirable to notify the operator of the error via the notification unit of the laser processing device 40, rather than proceeding directly to the second alignment process, so that necessary countermeasures can be taken. Error notification by the notification unit can be done by displaying an error on a display monitor, lighting or flashing a lamp, or sounding an audible message from a speaker. 【0051】 [Inversion process] Following the first modification layer formation step, an inversion step is performed to reverse the orientation of the wafer unit 30 from its initial holding position. In the inversion step, the wafer unit 30 is placed on the holding surface 431 of the holding part 43 with the second surface 14 of the first wafer 10 facing downwards (see Figure 5). Negative pressure is applied to the holding surface 431 to hold the wafer unit 30 by suction, and the ring frame 32 is fixed by the frame holding part 432. The wafer unit 30 is held with the fourth surface 24 of the second wafer 20 facing upwards, and the processing head 412 of the laser irradiation unit 41 and the imaging unit 42 are positioned opposite each other above the fourth surface 24. More specifically, the tape 31 fixed to the fourth surface 24 faces the laser irradiation unit 41 and the imaging unit 42. As the position of the fourth surface 24 changes due to the inversion of the wafer unit 30, the tape 31 deforms between itself and the frame holding part 432 (clamp 433). Following the inversion process, the second alignment process is performed. 【0052】 [Second alignment process] In the second alignment process shown in Figure 5, the imaging unit 42 is instructed to focus on the modified layer 18 within the first wafer 10 from the fourth surface 24 side of the wafer unit 30 and perform imaging with the focus position Fb aligned. The area where the modified layer 18 is formed has changed optical conditions such as refractive index and reflectivity compared to the first base wafer 11 before modification, and has changed to a state in which an image can be formed by the optical system of the imaging unit 42. Furthermore, since the imaging unit 42 images with light of a wavelength that penetrates all of the first base wafer 11, the second base wafer 21, and the tape 31, the tape 31 attached to the fourth surface 24 and the second base wafer 21 below it do not interfere with imaging of the modified layer 18 from the fourth surface 24 side. Therefore, under imaging conditions as shown in Figure 5, it is possible to perform imaging with the focus position Fb set to the position of the modified layer 18, and an image in which the modified layer 18 is clearly visible is obtained. The control unit 46 processes the image captured by the imaging unit 42 with the modified layer 18 in focus using the image processing unit 47, and identifies the second division line 27, which will be the boundary of the division of the second wafer 20, with the modified layer 18 as the reference (so that it coincides with the position of the modified layer 18 in the horizontal direction). The identified second division line 27 is converted into data as coordinate information in the horizontal direction and stored in the memory of the control unit 46. 【0053】 As shown in Figure 5, in the thickness direction of the wafer unit 30 after the inversion process, the first pattern 12 and the second pattern 22 are located between the imaging unit 42 and the modified layer 18. When imaging from the fourth surface 24 side with the imaging unit 42, with a typical autofocus specification, the focus will be on the first pattern 12 and the second pattern 22, which are located in front of the modified layer 18 from the perspective of the imaging unit 42, making it highly likely that the modified layer 18 cannot be imaged as is. In addition, the visual traces of the modified layer 18 are fainter (low visibility) compared to the first pattern 12 and the second pattern 22, making the modified layer 18 itself a difficult subject to focus on. Therefore, in order to reliably focus on the modified layer 18 at the focus position Fb and perform imaging, the following processing is performed in the second alignment process. 【0054】 [Common elements of the second alignment process: First focus adjustment process] First, a first focusing process is performed to align the focus position of the imaging unit 42 with the second pattern 22 of the second wafer 20. Focusing on the second pattern 22 closer to the imaging unit 42 can be easily done using normal autofocus control. 【0055】 Following the first focusing process, either a first configuration is performed, which involves a focus area setting process and a second focusing process, or a second configuration is performed, which involves a focus movement process. In the first configuration, the focus area setting process is further divided into an automatic setting configuration, in which the laser processing device 40 automatically sets the focus area, and a manual setting configuration, in which the focus area is set by an operator operating the laser processing device 40. 【0056】 [First form of the second alignment process: Focus area setting process] In the first embodiment of the focus area setting process, imaging is performed by the imaging unit 42 with the imaging unit 42 in focus on the second pattern 22 in the first focusing process. This results in the acquisition of an image in which the second pattern 22 is clearly visible. Then, as shown in Figure 6, a focus area Fs smaller than the second street area 25 (narrower than the width of the second street area 25) is set in the second street area 25 within the imaging area of ​​the image captured by the imaging unit 42. Specifically, this means setting the imaging unit 42 to focus only within the area inside the second street area 25, so as not to include the boundary line between the second street area 25 and the device area 26 located outside of it, or characteristic patterns and shapes contained in the device area 26. In other words, the imaging unit 42 performs a process to narrow down the focusing area so as to exclude the area outside the second street area 25 that contains noisy image information from the imaging area it can capture. Figure 6 shows the case where the focus area Fs is set for the second street area 25 extending in the left-right direction in the figure, but the focus area Fs can be set similarly for the second street area 25 extending in the up-down direction in the figure. 【0057】 [Variations in the focus area setting process: Pattern identification process] When the laser processing apparatus 40 automatically sets the focus area Fs, a pattern identification process is performed. In the pattern identification process, the position of a predetermined shape included in the second pattern 22 is identified from the image captured by the imaging unit 42 of the second pattern 22. More specifically, the shape and position of the boundary line between the second street area 25 and the device area 26 are determined by image analysis, including pattern matching. 【0058】 Then, in the focus region setting step, based on the position of a predetermined shape (the shape of the boundary line between the second street region 25 and the device region 26) identified in the pattern identification step, the coordinate position of the second street region 25 demarcated by the boundary line is recognized, and a focus region Fs smaller than the second street region 25 is set within the range of the second street region 25. 【0059】 [Variations in the focus area setting process: Image display process] When the focus area Fs is set by an operator operating the laser processing device 40, an image display process is performed. In the image display process, the image of the second pattern 22 captured by the imaging unit 42 (as shown in Figure 6) is displayed on the display monitor (not shown) of the laser processing device 40 or on the screen of an external device (personal computer, tablet terminal, etc.) that can communicate with the laser processing device 40. 【0060】 In the focus area setting process, the operator performs predetermined operations on the displayed image to set a focus area Fs smaller than the second street area 25 within the range of the displayed second street area 25. For example, if the display device that displays the image is a touch panel, the range of the focus area Fs can be specified by touching the touch panel. Alternatively, the operator can specify the range of the focus area Fs on the image on the screen using an input device such as a keyboard or mouse. 【0061】 [First form of the second alignment process: Second focus adjustment process] Following the focus area setting process, a second focusing process is performed. In the second focusing process, the imaging unit 42 focuses within the focus area Fs set by automatic or manual setting in the previous focus area setting process. At the position of the second pattern 22 in the thickness direction of the wafer unit 30, there are no characteristic objects within the focus area Fs that the imaging unit 42 can focus on. Similarly, at the position of the first pattern 12, which has a pattern shape similar to the second pattern 22 (having a first street area 15 with the same shape as the second street area 25), the focus area Fs is smaller than the first street area 15, so there are no characteristic objects within the focus area Fs that the imaging unit 42 can focus on. Therefore, under the condition that focusing is performed in a narrowed range inside the focus area Fs, the positions of the second pattern 22 and the first pattern 12 are not in focus, and the imaging unit 42 changes the focus position in the thickness direction of the wafer unit 30 in autofocus control until an object that can be focused inside the focus area Fs is detected. Since the modified layer 18 formed in the first modified layer formation process is located inside the focus region Fs, the imaging unit 42 will eventually come into focus on the modified layer 18. 【0062】 [Second form of the second alignment process: Focus shifting process] In the second focus shift step, the focus position of the imaging unit 42, which was aligned with the second pattern 22 in the first focus adjustment step, is moved a predetermined distance in the thickness direction of the wafer unit 30 so that the imaging unit 42 is in focus on the modified layer 18. Specifically, the predetermined distance by which the focus position of the imaging unit 42 is moved in the focus shift step is the distance from the second pattern 22 to the modified layer 18 in the thickness direction of the wafer unit 30. The distance from the second pattern 22 to the modified layer 18 can be obtained from the processing condition data of the wafer unit 30 input to the control unit 46. The processing conditions include the total thickness of the first wafer 10 and the thickness of the first pattern 12, the total thickness of the second wafer 20 and the thickness of the second pattern 22, and the processing depth of the modified layer 18 within the first wafer 10. Therefore, the control unit 46 can calculate the distance from the second pattern 22 to the modified layer 18 based on this data. Then, when the focus position of the imaging unit 42, which is set to match the second pattern 22, is moved by that distance, the imaging unit 42 comes into focus on the modified layer 18. 【0063】 As described above, the second alignment process is performed to align the focus position Fb with the modified layer 18. To summarize the flow of this process, first, a first focusing process is performed to align the focus position of the imaging unit 42 with the second pattern 22. After the first focusing process, either the first or second form is selected. In the first form, a narrowed focus area Fs is set (automatically or manually) based on the image captured with the second pattern 22 in focus, so as not to include any extraneous objects other than the modified layer 18. The imaging unit 42 then performs focusing (autofocus) within the limited range inside the focus area Fs, resulting in a state where the modified layer 18 is in focus. In the second form, the focus position of the imaging unit 42, which is aligned with the second pattern 22, is moved by a predetermined distance determined from processing conditions, etc., resulting in a state where the modified layer 18 is in focus. The settings menu of the laser processing device 40 may include an option to select which of the above methods to use to align the focus position Fb with the modified layer 18, allowing the operator to make a selection. 【0064】 Then, using one of the methods described above, the imaging unit 42 takes an image with the modified layer 18 in focus, and the control unit 46 performs image processing using the image processing unit 47 to identify the second planned division line 27 based on the modified layer 18. 【0065】 It is preferable to perform the second alignment process only if the modified layer 18 formed in the first modified layer formation process meets the predetermined quality standards after performing the quality confirmation process described above. The quality confirmation process ensures that the multiple modified layers 18 formed along the first planned division line 17 are arranged appropriately without meandering or interruptions, and that the size of each modified layer 18 is also appropriate. Therefore, the second planned division line 27 identified in the second alignment process based on the modified layer 18 will have an extremely high degree of agreement with the first planned division line 17 and will be highly accurate. If, within the acceptable quality range, a very small number of modified layers 18 are formed slightly off-center from the first planned division line 17, a straight second planned division line 27 may be set that passes through a position that coincides with the other modified layers 18, excluding the specific modified layer 18. In other words, the identification of the second division line 27 in the second alignment process is not limited to a configuration in which the second division line 27 perfectly coincides with the positions of all the modified layers 18 in a plan view of the wafer unit 30. 【0066】 [Second Modified Layer Formation Process] In the second modification layer formation process shown in Figure 7, the lifting mechanism 45 is controlled so that the focal point of the laser beam Lz irradiated from the processing head 412 of the laser irradiation unit 41 is at a predetermined depth inside the second base wafer 21. Furthermore, the operation of the horizontal drive mechanism 44 is controlled so that the position of the focal point of the laser beam Lz in the horizontal direction matches the second division line 27 identified in the second alignment process. The laser beam Lz is incident from the fourth surface 24 side of the wafer unit 30, forming a modification layer 28 inside the second base wafer 21. Multiple modification layers 28 are formed at predetermined intervals along each second division line 27. The tape 31 fixed to the fourth surface 24 is made of a material that transmits the laser beam Lz (does not absorb the laser beam Lz), allowing the laser beam Lz to be incident from the fourth surface 24 side without being obstructed by the tape 31. Once the modified layer 28 has been formed along all of the second planned splitting lines 27 identified in the second alignment process, the second modified layer formation process is complete. 【0067】 In recent years, due to improvements in semiconductor chip manufacturing technology, the width of the streets between chips on a semiconductor wafer has become narrower. Therefore, if a modified layer 18 and a modified layer 28 are formed inside the first base wafer 11 and the second wafer 20, respectively, by laser irradiation from only one outer surface (second surface 14 or fourth surface 24) in the thickness direction of the wafer unit 30, when forming the modified layer on the side farther from the laser incident surface, a portion of the laser beam Lz, which has a beam diameter larger than the focal point, may not remain within the range of the first street region 15 and the second street region 25, but instead extend into the device region 16 and the device region 26, potentially damaging the device region 16 and the device region 26 due to the power of the laser beam Lz. To prevent such problems, in the processing method of this embodiment, in the first modified layer formation step, a laser beam Lz is incident from the second surface 14 side of the first wafer 10, and after the first modified layer formation step, the wafer unit 30 is inverted upside down, and in the second modified layer formation step, a laser beam Lz is incident from the fourth surface 24 side of the second wafer 20. As a result, the modified layers 18 and 28 can be formed with appropriate processing without damaging the device region 16 included in the first pattern 12 or the device region 26 included in the second pattern 22 due to the influence of the laser beam Lz. 【0068】 Furthermore, if the widths of the first street region 15 and the second street region 25 are sufficiently large, and there is no problem in forming both the modified layer 18 and the modified layer 28 by incident laser beam Lz from one outer surface (second surface 14 or fourth surface 24) in the thickness direction of the wafer unit 30, then, as a modification, it is also possible to perform the first alignment process, the first modified layer formation process, the second alignment process, and the second modified layer formation process without performing the inversion process, while the wafer unit 30 remains in the orientation shown in Figure 2 during the initial holding process. In this modification, in the first modified layer formation process, the modified layer 28 is first formed inside the second wafer 20 located on the side farther from the laser irradiation unit 41, and then in the second modified layer formation process, the modified layer 18 is formed inside the first wafer 10 located on the side closer to the laser irradiation unit 41. Then, in the second alignment process, the imaging unit 42 is focused on the modified layer 28 of the second wafer 20. As a result, when forming the modified layer 28 in the first modified layer formation process, the modified layer 18 is not present in the optical path of the laser beam Lz from the laser irradiation unit 41 to the second base wafer 21 (the modified layer 18 is not yet formed), thus avoiding the modification of the modified layer 28 being hindered by the modified layer 18. 【0069】 After each of the above steps, a modified layer 18 is formed inside the first wafer 10 (first base wafer 11) in the wafer unit 30, and a modified layer 28 is formed inside the second wafer 20 (second base wafer 21), completing the processing in the laser processing apparatus 40. 【0070】 [Dividing process] After the second modification layer formation process, an external force is applied to the wafer unit 30, which has the modification layer 18 and modification layer 28 formed inside it, and a splitting process is performed in which the wafer unit 30 is split using the modification layer 18 and modification layer 28 as splitting points. 【0071】 Figures 8 and 9 show an expander 50 as an example of a device that performs a splitting process. The expander 50 transports the wafer unit 30, which has been processed in the laser processing device 40, to the expander 50 for the splitting process. The expander 50 is equipped with a ring-shaped holder 51 that can hold the ring frame 32 of the wafer unit 30, and the ring frame 32 can be fixed on the holder 51 via a plurality of clamps 52 that can be opened and closed relative to the holder 51. The holder 51 can be moved vertically by a lifting mechanism 53, which is composed of an air cylinder or the like. A central opening 54 that penetrates vertically is formed in the center of the holder 51, and a cylindrical push-up member 55 is arranged inside the central opening 54. 【0072】 The wafer unit 30, transported to the expander 50, is held as shown in Figure 8. The upper surface of the holder 51 and the upper end of the push-up member 55 are set to approximately the same height, and the ring frame 32 is placed on the upper surface of the holder 51 and secured with the clamp 52. At this stage, the tape 31 is supported in a flat shape over the upper surface of the holder 51 and the upper end of the push-up member 55, and the first wafer 10 and the second wafer 20 are positioned above the hollow internal space of the push-up member 55. 【0073】 As shown in Figure 8, once the wafer unit 30 is held, the lifting mechanism 53 is operated to move the holding base 51 downward. Then, as shown in Figure 9, the central portion of the tape 31 to which the first wafer 10 and the second wafer 20 are attached is restricted from moving downward by the upper end of the push-up member 55, while the outer circumference of the tape 31 is pulled downward by the holding base 51 together with the ring frame 32. As a result, the tape 31 is pulled and expands, and an external force acts on the first wafer 10 and the second wafer 20 supported by the tape 31 in a direction that expands their diameter. 【0074】 The application of external force by the expanding device 50 causes the first wafer 10 to split starting from the modified layer 18 (not shown in Figures 8 and 9), and the second wafer 20 to split starting from the modified layer 28 (not shown in Figures 8 and 9), and the device regions 16 and 26, which were separated by the first street region 15 and second street region 25 respectively, are separated into individual pieces. In other words, multiple semiconductor chips formed in multiple device regions 16 of the first wafer 10 and multiple semiconductor chips formed in multiple device regions 26 of the second wafer 20 are manufactured. Furthermore, by expanding the tape 31, the spacing between each semiconductor chip is widened, making it easier to remove the semiconductor chips from the tape 31. 【0075】 Furthermore, depending on the formation status of the modified layers 18 and 28 in the laser processing apparatus 40, partial separation of the first wafer 10 and the second wafer 20 may occur before the tape 31 is expanded in the expander 50. In such cases, the entire first wafer 10 and the second wafer 20 can be reliably separated by expanding the tape 31 in the expander 50. In addition, by expanding the tape 31 and widening the spacing between the separated semiconductor chips, it becomes easier to remove the individual manufactured semiconductor chips from the tape 31. 【0076】 The application of external force to the wafer unit 30 during the splitting process may be performed using a device with a different configuration than the expanded device 50 shown in the figure. For example, instead of the push-up member 55, a pressing roller may be provided above the wafer unit 30. The roller is supported so as to be rotatable about an axis extending horizontally. With the ring frame 32 held by the holding base 51 as shown in Figure 8, the first wafer 10 and the second wafer 20 can be split by applying external force to them by rolling the roller and moving it horizontally while pressing the roller against the first wafer 10 from above by raising the holding base 51 by the lifting mechanism 53 or by lowering the roller. 【0077】 As described above, the processing method of this disclosure, which includes a first alignment step, a first modified layer formation step, a second alignment step, and a second modified layer formation step, can suppress misalignment of the modified layers (modified layer 18, modified layer 28) formed on each wafer (first wafer 10, second wafer 20) constituting the bonded wafer (wafer unit 30). Furthermore, by applying the processing method of this disclosure as a semiconductor chip manufacturing method including a splitting step, it is possible to manufacture high-quality semiconductor chips by precisely splitting a bonded wafer, in which the misalignment of the modified layers of each wafer is suppressed, using the modified layers as a starting point (without causing splitting defects). 【0078】 Furthermore, the embodiments of the present invention are not limited to the embodiments and modifications described above, and may be modified, substituted, or altered in various ways without departing from the spirit of the technical idea of ​​the present invention. Moreover, if the technical idea of ​​the present invention can be realized in a different way by advances in the art or by other derived arts, it may be implemented by that method. Accordingly, the claims cover all embodiments that may fall within the scope of the technical idea of ​​the present invention. [Industrial applicability] 【0079】 According to the technology disclosed herein, misalignment of the modified layer formed on each wafer constituting the bonded wafer can be suppressed, enabling the bonded wafer to be neatly separated starting from the modified layer, thereby improving the quality and productivity of semiconductor chips and the like. [Explanation of symbols] 【0080】 10: First wafer 11: First base wafer 12: Pattern 1 13: 1st page 14:Second side 15: First Street Area 16: Device area 17: Planned first division line 18: Modified layer 20: Second wafer 21: Second base wafer 22: Pattern 2 23: 3rd page 24:Side 4 25: Second Street Area 26: Device area 27: Planned line for the second division 28: Modified layer 30: Wafer Unit 31: Tape 32: Ring Frame 40: Laser processing equipment 41: Laser irradiation unit 42: Imaging Unit 43: Holding part 44: Horizontal drive mechanism 45: Lifting mechanism 46: Control Unit 47: Image Processing Unit 50: Expanding device 51: Holding stand 52: Clamp 53: Lifting mechanism 54: Central opening 55: Push-up member 411: Laser oscillator 412: Machining head 413: Miller 414: Focusing lens 431: Holding surface 432: Frame holding part 433: Clamp Fa: Focus position (first alignment process) Fb: Focus position (second alignment process) Fs: Focus area Lz: Laser beam

Claims

[Claim 1] A first pattern is formed on a first base wafer having a first street region including a first division planned line that serves as the boundary for division during fragmentation, and the first wafer has a first surface including the first pattern and a second surface opposite to the first surface, A wafer unit processing method for processing a wafer unit in which a second pattern having a second street region is formed on a second base wafer, and a second wafer having a third surface containing the second pattern and a fourth surface opposite to the third surface is joined with the first surface and the third surface facing each other, A first alignment step involves using an imaging unit to image the first pattern or the second pattern with light of a wavelength that penetrates the first base wafer and the second base wafer, and to identify the first planned division line. A first modified layer formation step in which a modified layer is formed on the first wafer along the first planned division line identified by the first alignment step, A second alignment step involves using the imaging unit to image the modified layer of the first wafer formed in the first modified layer formation step, and identifying the second planned division line which will be the boundary for the division of the second wafer. A second modified layer formation step, in which a modified layer is formed on the second wafer along the second planned division line identified by the second alignment step, A method for processing a wafer unit equipped with [a specific component]. [Claim 2] The second alignment process is as follows: A first focusing step in which the imaging unit focuses on the second pattern on the second wafer, The first focusing step involves capturing the second pattern with the imaging unit in focus on the second pattern, and the focus area setting step involves setting a focus area smaller than the second street area within the imaging area of ​​the imaging unit. The system includes a second focusing step which performs focusing within the focus area set in the first focusing area setting step, The method for processing a wafer unit according to feature 1. [Claim 3] The focus area setting step further includes, The system includes a pattern identification step that identifies the position of a predetermined shape included in the second pattern from an image of the second pattern, Based on the position of a predetermined shape identified by the pattern identification step, a focus area smaller than the second street area is set in the second street area within the imaging area of ​​the imaging unit. The method for processing a wafer unit according to feature 2. [Claim 4] The second alignment process is as follows: A first focusing step in which the imaging unit focuses on the second pattern on the second wafer, The process includes a focus shifting step which moves the focus position of the imaging unit, which has been aligned to the second pattern in the first focusing step, by a predetermined distance. The method for processing a wafer unit according to feature 1. [Claim 5] In the first alignment step, the imaging unit images the first pattern or the second pattern from the side of the second surface of the wafer unit. In the first modified layer formation step, a laser beam is incident on the wafer unit from the side of the second surface to form the modified layer. In the second alignment step, the imaging unit images the modified layer of the first wafer from the fourth side of the wafer unit. In the second modification layer formation step, a laser beam is incident on the wafer unit from the fourth surface side to form the modification layer. A method for processing a wafer unit according to any one of claims 1 to 4. [Claim 6] The wafer unit is, A ring frame with an opening in the center, The wafer unit further comprises a tape that is fixed to the second and / or fourth surface of the wafer unit and the ring frame, A method for processing a wafer unit according to any one of claims 1 to 4. [Claim 7] The method further includes a quality confirmation step of imaging the modified layer formed on the first wafer to confirm the quality of the modified layer. A method for processing a wafer unit according to any one of claims 1 to 4. [Claim 8] A first pattern having a first street region including a first division planned line that serves as the boundary for division during fragmentation is formed on a first base wafer made of semiconductor, and a first wafer having a first surface including the first pattern and a second surface opposite to the first surface, A method for manufacturing a semiconductor chip, comprising: forming a second pattern having a second street region on a second base wafer made of semiconductor material; joining a second wafer having a third surface containing the second pattern and a fourth surface opposite to the third surface, with the first surface and the third surface facing each other, to a wafer unit, and dividing the wafer unit along the first planned division line, A first alignment step involves using an imaging unit to image the first pattern or the second pattern with light of a wavelength that penetrates the first base wafer and the second base wafer, and to identify the first planned division line. A first modified layer formation step in which a modified layer is formed on the first wafer along the first planned division line identified by the first alignment step, A second alignment step involves using the imaging unit to image the modified layer of the first wafer formed in the first modified layer formation step, and identifying the second planned division line which will be the boundary for the division of the second wafer. A second modified layer formation step, in which a modified layer is formed on the second wafer along the second planned division line identified by the second alignment step, The process includes, after the second modified layer formation step, a splitting step in which an external force is applied to the wafer unit and the wafer unit is split using the modified layer as the splitting starting point. A method for manufacturing a semiconductor chip, characterized by the following:

Images