Manufacturing method for semiconductor devices
The method addresses residual stress in semiconductor wafers by forming a crack below a groove and grinding the first surface to enhance reliability and reduce chipping, facilitating efficient separation of the substrate and electrodes.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- DENSO CORP
- Filing Date
- 2022-12-16
- Publication Date
- 2026-05-15
AI Technical Summary
The scribe-and-break method for semiconductor wafer dicing leaves residual stress near the surface, leading to potential chipping and reduced reliability of the semiconductor device.
A manufacturing method that includes forming a crack below a groove on the first surface of the semiconductor substrate, grinding the first surface to remove residual stress, and then dividing the substrate along the crack using a dividing member on the second surface.
This method effectively reduces residual stress, minimizing chipping and enhancing the reliability of semiconductor devices by allowing for easier and less damaging separation of the substrate and electrodes.
Smart Images
Figure 0007859961000001 
Figure 0007859961000002 
Figure 0007859961000003
Abstract
Description
Technical Field
[0001] The technology disclosed in this specification relates to a method for manufacturing a semiconductor device.
Background Art
[0002] Among the steps of a method for manufacturing a semiconductor device, there is a step of singulating (i.e., dividing and separating) a semiconductor wafer on which a plurality of element structures are formed into chips. Patent Document 1 discloses a technology for singulating a semiconductor wafer into chips by so-called stealth dicing.
[0003] In recent years, with regard to dicing of a semiconductor wafer, a method called scribe and break has begun to be adopted. In this method, first, a scribing wheel is pressed against the boundary between adjacent element structures, and a crack is formed in the semiconductor wafer along the boundary. Next, a dividing member is pressed along the boundary to divide the semiconductor wafer along the boundary.
Prior Art Documents
Patent Documents
[0004]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0005] In the scribe-and-break method, a scribing wheel is pressed against a semiconductor wafer to create stress within the wafer, thereby forming cracks. This stress remains as residual stress even after the semiconductor wafer has been fragmented. In particular, this residual stress is significantly present near the surface of the semiconductor wafer against which the scribing wheel was pressed. Therefore, repeated operation of the manufactured semiconductor device can cause chipping or other damage near the areas where residual stress exists. When chipping occurs, the reliability of the semiconductor device decreases. This specification provides a technology that enables the manufacture of highly reliable semiconductor devices using the scribe-and-break method. [Means for solving the problem]
[0006] A method for manufacturing a semiconductor device disclosed herein comprises the steps of forming a groove (G) on a first surface (2a) of a semiconductor substrate (2) having a plurality of element regions (3) by pressing a scribing wheel (32) along the boundary (4) of the element regions to plastically deform the semiconductor substrate, wherein a crack (5) is formed below the groove, along the groove and extending in the thickness direction of the semiconductor substrate; grinding the first surface after the step of forming the groove; and dividing the semiconductor substrate along the boundary by pressing a dividing member (33) against the semiconductor substrate on a second surface (2b) of the semiconductor substrate located on the opposite side of the first surface, along the boundary, after the step of grinding the first surface.
[0007] In this manufacturing method, first, a scribing wheel is pressed against the first surface of the semiconductor substrate to form a crack in the semiconductor substrate. The crack is formed from the first surface side. Therefore, residual stress is significantly present in the vicinity of the first surface of the semiconductor substrate. Subsequently, the first surface is ground. This removes the region where residual stress is significantly present (i.e., the vicinity of the first surface). Then, the semiconductor substrate is divided by pressing a dividing member against the second surface side. In this way, the above manufacturing method can remove the region where residual stress is significantly present, thus suppressing chipping of the semiconductor substrate. Therefore, a semiconductor device with high reliability can be manufactured using the above manufacturing method. [Brief explanation of the drawing]
[0008] [Figure 1] This is a plan view of a semiconductor substrate. [Figure 2] This is a diagram illustrating the process of attaching the support plate. [Figure 3] This diagram illustrates the rough grinding process before crack formation. [Figure 4] This is a diagram illustrating the precision grinding process before cracks are formed. [Figure 5] This is a diagram illustrating the crack formation process. [Figure 6] This is a diagram illustrating how cracks are formed. [Figure 7] This is a manipulated electron microscope image of a cross-section of a semiconductor substrate in which cracks have formed. [Figure 8] This graph shows the relationship between the distance from the surface of the semiconductor substrate and residual stress. [Figure 9] This diagram illustrates the grinding process after crack formation. [Figure 10] This is a diagram illustrating the electrode formation process. [Figure 11] This is a diagram illustrating the process of applying dicing tape. [Figure 12] This is a diagram illustrating the support plate removal process. [Figure 13] This is a diagram for explaining the protective member coating process. [Figure 14] This is a diagram for explaining the dividing process. [Figure 15] This is a diagram for explaining the pickup process. [Figure 16] This is a diagram for explaining the soldering process.
Embodiments for Carrying Out the Invention
[0009] In an example of the manufacturing method disclosed in this specification, in the step of grinding the first surface, the first surface may be ground until the groove is removed.
[0010] In an example of the manufacturing method disclosed in this specification, in the step of forming the groove, a stress region (R) where stress remains is formed in the surface layer portion near the first surface of the semiconductor substrate, the crack is formed below the stress region, and in the step of grinding the first surface, the first surface may be ground until the stress region is removed.
[0011] In an example of the manufacturing method disclosed in this specification, before the step of forming the groove, the method further includes a step of grinding the first surface, and the step of grinding the first surface before the step of forming the groove may include rough grinding of grinding the first surface with a coarser grinding wheel than the grinding wheel used in the step of grinding the first surface after the step of forming the groove.
[0012] In an example of the manufacturing method disclosed in this specification, the step of grinding the first surface before the step of forming the groove may include fine grinding of grinding the first surface with a more precise grinding wheel than the grinding wheel used in the rough grinding after the rough grinding.
[0013] In an example of the manufacturing method disclosed in this specification, the method may further include a step of forming an electrode (8) on the first surface and a step of soldering the electrode.
[0014] (Example) The manufacturing method of the embodiment will be described with reference to the drawings. FIG. 1 is a plan view of a semiconductor substrate 2 in which a plurality of element regions 3 are formed in a matrix. In FIG. 1, each element region 3 is schematically shown by a solid line. For convenience of explanation, the boundary between adjacent element regions 3, which becomes the edge of each individual element region (semiconductor device) obtained by the division when the semiconductor substrate 2 is later divided into individual element regions 3, is called a planned division line 4. The planned division line 4 is a virtual line, not a line actually marked on the semiconductor substrate 2. The planned division line 4 may be a line or groove actually drawn on the semiconductor substrate 2 so as to be visible. In each element region 3, a semiconductor element having functions such as a transistor or a diode is formed.
[0015] The semiconductor substrate 2 is made of silicon carbide (SiC). Note that the semiconductor substrate 2 may be made of other semiconductor materials such as silicon (Si) or gallium nitride (GaN). As shown in FIG. 2 and the like, the semiconductor substrate 2 has a first surface 2a and a second surface 2b located on the back side of the first surface 2a. On the second surface 2b of the semiconductor substrate 2, a main electrode 6 (source electrode, anode electrode, etc.) is formed.
[0016] The manufacturing method of the embodiment includes a support plate attachment step, a rough grinding step, a first fine grinding step, a crack formation step, a second fine grinding step, a metal film formation step, a dicing tape attachment step, a support plate peeling step, a protective member coating step, a division step, and a soldering step.
[0017] (Support plate attachment step) In the support plate attachment step, as shown in FIG. 2, a support plate 12 is attached to the second surface 2b of the semiconductor substrate 2. The support plate 12 is attached to the second surface 2b via an adhesive 11. The support plate 12 is made of, for example, glass. The adhesive 11 is, for example, a silicon-based adhesive, and in addition to the function of bonding the semiconductor substrate 2 to the support plate 12, it also has the function of protecting the second surface 2b of the semiconductor substrate 2. Therefore, here, the adhesive 11 is applied so that the thickness of the adhesive 11 is greater than the thickness of the main electrode 6.
[0018] (Rough grinding process) Next, as shown in Figure 3, the first surface 2a of the semiconductor substrate 2 is ground with a grinding wheel 31. The grinding wheel 31 used in the rough grinding process is a relatively coarse grinding wheel (i.e., with a relatively small grain size). This thins the semiconductor substrate 2.
[0019] (1st fine grinding process) Next, as shown in Figure 4, the first surface 2a of the semiconductor substrate 2 is ground with a grinding wheel 131. The grinding wheel 131 used in the first fine grinding step is a finer grinding wheel (i.e., has a larger grain size) than the grinding wheel 31 used in the rough grinding step. This further thins the semiconductor substrate 2. In the modified example, neither the rough grinding step nor the first fine grinding step may be performed, or only one of them may be performed.
[0020] (Crack formation process) Next, the crack formation process shown in Figure 5 is carried out. In the crack formation process, a scribing wheel 32 is pressed against the first surface 2a of the semiconductor substrate 2 attached to the support plate 12, thereby forming a groove G with a crack 5 in the semiconductor substrate 2. The scribing wheel 32 is a disc-shaped member and is rotatably supported by a support device (not shown). Here, the outer circumferential surface of the scribing wheel 32 is pressed against the first surface 2a of the semiconductor substrate 2 and moved (scanned) along the planned division line 4. The outer circumferential surface of the scribing wheel 32 has a sharp convex shape. As the scribing wheel 32 moves along the planned division line 4, it rolls (moves) on the first surface 2a of the semiconductor substrate 2 like a tire rolling on a road surface. A groove G (see Figure 7) is formed on the first surface 2a of the semiconductor substrate 2 by plastic deformation of the semiconductor substrate 2 along the planned division line 4. That is, the groove G is not formed by cutting the first surface 2a of the semiconductor substrate 2 with the scribing wheel 32. As shown in Figures 6 and 7, when the first surface 2a is pressed by the scribing wheel 32, compressive stress is generated in the surface region R near the first surface 2a inside the semiconductor substrate 2. The compressive stress is generated isotropically, starting from the point of pressure by the scribing wheel 32 (the point of contact between the periphery of the scribing wheel 32 and the first surface 2a), as shown by arrow 20 in Figure 6. A groove G is formed at the point of pressure by the scribing wheel 32, while tensile stress is generated inside the semiconductor substrate 2 directly below the region R where compressive stress is generated. The tensile stress is generated directly below the region R where compressive stress is generated, along the first surface 2a of the semiconductor substrate 2, in a direction away from the planned division line 4, as shown by arrow 22. This tensile stress causes a crack 5 to form inside the semiconductor substrate 2, extending in the thickness direction of the semiconductor substrate 2. Here, by pressing the scribing wheel 32 against the first surface 2a and moving it along the planned division line 4, a crack 5 is formed along the boundary of adjacent element regions 3 and extending in the thickness direction of the semiconductor substrate 2. The crack 5 is formed near the surface of the first surface 2a of the semiconductor substrate 2.Generally, compressive stress suppresses crack formation and extension. Therefore, crack 5 is formed so that it extends from outside the region R where compressive stress occurs at the point of pressure by the scribing wheel 32 on the first surface 2a of the semiconductor substrate 2 to the region where tensile stress occurs directly below the region R where compressive stress occurs. As described above, a rough grinding process and a first fine grinding process are performed before the crack formation process. As a result, a fractured layer is formed on the first surface 2a of the semiconductor substrate 2. Because the fractured layer is brittle, crack 5 can be formed in the semiconductor substrate 2 with a small load during the crack formation process.
[0021] Figure 7 is a scanning electron microscope image of a cross-section of the semiconductor substrate 2 after crack 5 has been formed by the scribing wheel 32. As shown in Figure 7, it can be seen that a groove G is formed on the first surface 2a of the semiconductor substrate 2 by plastic deformation when the scribing wheel 32 is pressed against the first surface 2a of the semiconductor substrate 2. Furthermore, the compressive stress generated in the surface region R near the first surface 2a of the semiconductor substrate 2 remains in region R. In addition, it can be seen that a crack 5 has been formed directly below region R, along the boundary of the element region 3.
[0022] Figure 8 is a graph showing the residual stress in the semiconductor substrate 2 after crack 5 is formed by the scribing wheel 32. The distance on the horizontal axis represents the distance from the second surface 2b of the semiconductor substrate 2, and the position of the right end of the graph represents the position of the first surface 2a of the semiconductor substrate 2. As shown in Figure 8, the residual stress is locally present in the surface layer (i.e., region R) near the first surface 2a of the semiconductor substrate 2. There is almost no residual stress in areas of the semiconductor substrate 2 other than region R.
[0023] (Second fine grinding process) Next, as shown in Figure 9, the first surface 2a of the semiconductor substrate 2 is ground with a grinding wheel 131. The grinding wheel 131 used in the second fine grinding step is a grinding wheel with the same grit size as the grinding wheel 131 used in the first fine grinding step. This removes the region R in which residual stress exists from the first surface 2a of the semiconductor substrate 2. As a result, there is almost no residual stress in the semiconductor substrate 2 after grinding.
[0024] (Metal film formation process) Next, the metal film formation process shown in Figure 10 is carried out. In the metal film formation process, a metal film 8 is formed on the first surface 2a of the semiconductor substrate 2. The materials constituting the metal film 8 are not particularly limited, but for example, it is a multilayer film made of titanium, nickel, and gold. The metal film 8 is formed to cover substantially the entire area of the first surface 2a. That is, the metal film 8 is formed on the first surface 2a so as to span multiple element regions 3. The metal film 8 functions as an electrode in the completed semiconductor device.
[0025] (Dicing tape application process) Next, the dicing tape application process shown in Figure 11 is carried out. In the dicing tape application process, the dicing tape 13 is applied to the surface of the metal film 8. The dicing tape 13 is applied so as to cover almost the entire surface of the metal film 8. The dicing tape 13 is fixed to a dicing frame (not shown). Note that in Figure 11 and subsequent figures, the semiconductor substrate 2 is depicted with the second surface 2b facing upwards.
[0026] (Support plate removal process) Next, the support plate peeling process shown in Figure 12 is performed. In the support plate peeling process, the support plate 12 and adhesive 11 are peeled off from the second surface 2b of the semiconductor substrate 2. Here, for example, the adhesive 11 is dissolved with a solvent, thereby peeling the support plate 12 off from the second surface 2b together with the adhesive 11. As a result, the semiconductor substrate 2 is supported by the dicing tape 13.
[0027] (Protective material coating process) Next, the protective member coating process shown in Figure 13 is carried out. In the protective member coating process, the protective member 15 is attached so as to span the surface of each main electrode 6 in each element region 3 of the semiconductor substrate 2, thereby covering the second surface 2b of the semiconductor substrate 2 with the protective member 15. The material of the protective member 15 is not particularly limited, but for example, resin can be used. By coating with the protective member 15, the second surface 2b of the semiconductor substrate 2 is protected in subsequent processes such as the splitting process.
[0028] (splitting process) Next, the splitting process shown in Figure 14 is carried out. In the splitting process, the break plate 33 is pressed along the planned splitting line 4 (crack 5 formed in the crack formation process), and the semiconductor substrate 2 is split along the planned splitting line 4 (along the boundary of the element region 3). Here, first, the semiconductor substrate 2 is placed on two support bases 34. The two support bases 34 are spaced apart. When the semiconductor substrate 2 is placed on the support bases 34, the semiconductor substrate 2 is placed so that the spacing is located below the position to be split (the position where the break plate 33 is pressed). After that, the break plate 33 is pressed against the second surface 2b of the semiconductor substrate 2 via the protective member 15. The break plate 33 is a plate-shaped member, and its lower end (the edge pressed against the second surface 2b) is ridge-shaped (sharp blade-shaped), but it is pressed against the semiconductor substrate 2 without cutting it.
[0029] Since there is no support base 34 below the break plate 33 (there is space between the two support bases 34), when the break plate 33 is pressed against the second surface 2b, the semiconductor substrate 2 bends so that it fits into the space between the two support bases 34. Here, the crack 5 is formed on the first surface 2a side of the semiconductor substrate 2. Therefore, when the break plate 33 is pressed against the semiconductor substrate 2 from the second surface 2b side, the semiconductor substrate 2 bends around the pressed portion (line) as an axis, and on the first surface 2a side, a force is applied in a direction that pulls the two element regions 3 adjacent to the division position away from the crack 5. Also, as described above, tensile stress is applied around the crack 5. Therefore, when the break plate 33 is pressed against the second surface 2b, the crack 5 extends in the thickness direction of the semiconductor substrate 2, and the semiconductor substrate 2 is divided along the planned division line 4. Furthermore, since the metal film 8 is formed on the first surface 2a of the semiconductor substrate 2, a force is applied to the metal film 8 in a direction that pulls the two element regions 3 adjacent to the division position apart, causing the metal film 8 to deform and be divided. Alternatively, instead of the two support bases 34, the entire first surface 2a of the semiconductor substrate 2 may be supported by a single elastic support plate (or one or more support bases via a single elastic support plate). In this case, although the elastic support plate is located below the break plate 33, when the semiconductor substrate 2 flexes, the elastic support plate deforms in accordance with the flexing of the semiconductor substrate 2. Therefore, when the break plate 33 is pressed against the second surface 2b, a force is applied to the crack 5 in a direction that pulls the two element regions 3 adjacent to the division position apart, similar to the case where it is supported by two support bases 34 (when there are no support bases 34 located below the break plate 33). The break plate 33 is an example of a "dividing member".
[0030] In the splitting process, the process of pressing the break plate 33 described above against the second surface 2b is repeatedly performed along each planned splitting line 4. This allows the semiconductor substrate 2 and the metal film 8 to be split along the boundary of each element region 3. Subsequently, as shown in Figure 15, the individualized element regions 3 and metal film 8 are peeled off from the dicing tape 13. When peeling the individualized element regions 3 and metal film 8 from the dicing tape 13, the dicing tape 13 can be expanded to separate the individualized element regions 3 and metal film 8 from each other before peeling.
[0031] (Soldering process) Next, the soldering process shown in Figure 16 is performed. In the soldering process, the metal film 8 of the individual semiconductor device and the conductive plate 10 on which the semiconductor device is placed are joined with solder 9. The conductive plate 10 constitutes part of the circuit electrically connected to the semiconductor element. This completes multiple semiconductor devices.
[0032] As described above, in this embodiment, first, a crack 5 is formed on the first surface 2a side of the semiconductor substrate 2 by pressing the scribing wheel 32 against the first surface 2a of the semiconductor substrate 2. Since the crack 5 is formed on the first surface 2a side of the semiconductor substrate 2, when the break plate 33 is pressed against it from the second surface 2b side, the semiconductor substrate 2 bends along the crack 5. As a result, a force is applied along the crack 5 from the first surface 2a side in a direction that folds the semiconductor substrate 2 apart. Consequently, the crack 5 extends in the thickness direction of the semiconductor substrate 2, and the semiconductor substrate 2 can be easily divided along the boundary of the element region 3. Furthermore, since the metal film 8 is formed on the first surface 2a of the semiconductor substrate 2, a force is also applied to the metal film 8 on both sides of the crack 5 in a direction that pulls the metal film 8 apart, causing the metal film 8 to deform and be easily divided. Thus, in this embodiment, the metal film 8 can be divided together with the semiconductor substrate 2 by a simple process of pressing the scribing wheel 32 and the break plate 33 against the semiconductor substrate 2.
[0033] Furthermore, in this embodiment, a crack 5 is formed in advance on the first surface 2a side inside the semiconductor substrate 2 before forming the metal film 8 on the first surface 2a. Therefore, both the semiconductor substrate 2 and the metal film 8 can be separated in a single step of pressing the break plate 33 from the second surface 2b side. In this embodiment, the crack 5 is formed on the first surface 2a side of the semiconductor substrate 2 before forming the metal film 8 on the first surface 2a. Therefore, compared to the case where the scribing wheel 32 is pressed against the first surface 2a via the metal film 8 to form the crack 5, the crack 5 can be formed with a lower load, thus reducing damage to the semiconductor substrate 2. Also, compared to the case where the scribing wheel 32 is pressed against the second surface 2b of the semiconductor substrate 2 to form the crack, damage to the boundary portion between element regions on the second surface 2b (the peripheral portion of the resulting semiconductor device) can be reduced.
[0034] Furthermore, in this embodiment, the scribing wheel 32 is pressed against the first surface 2a of the semiconductor substrate 2, resulting in a significant presence of residual stress in the vicinity of the first surface 2a of the semiconductor substrate 2. Subsequently, the first surface 2a is ground in the second precision grinding step. This removes the region R in which residual stress is significantly present (i.e., the vicinity of the first surface 2a). Thus, in this embodiment, since the region R in which residual stress is significantly present can be removed, chipping of the semiconductor substrate 2 can be suppressed. Therefore, the manufacturing method of this embodiment can produce a semiconductor device with high reliability.
[0035] In particular, with respect to the solder 9, if residual stress exists near the first surface 2a, the void ratio of the solder 9 is relatively high when the semiconductor device is repeatedly used. In this regard, in this embodiment, the region R in which residual stress is significantly present can be removed, so the void ratio of the solder 9 can be lowered.
[0036] In this embodiment, a support plate 12 made of glass is attached to the semiconductor substrate 2, and a crack 5 is formed on the first surface side of the semiconductor substrate 2. Since the support plate 12 is made of a relatively hard material, a crack 5 can be formed on the first surface side of the semiconductor substrate 2 with a relatively low load when the scribing wheel 32 is pressed against the semiconductor substrate 2.
[0037] In this embodiment, the semiconductor substrate 2 and the metal film 8 are separated while the dicing tape 13 is attached. Since the semiconductor substrate 2 (metal film 8) is fixed to the dicing tape 13, when the break plate 33 is pressed against the semiconductor substrate 2, displacement of the semiconductor substrate 2 can be suppressed, and scattering of the resulting semiconductor device can be prevented.
[0038] In this embodiment, the break plate 33 is pressed against the semiconductor substrate 2 while the second surface 2b is covered by the protective member 15. Since the second surface 2b is protected by the protective member 15, it is possible to prevent the break plate 33 from damaging the second surface 2b.
[0039] In the above-described embodiment, the support plate attachment step, the dicing tape attachment step, and the protective member covering step may be omitted.
[0040] Furthermore, in the second precision grinding step of the above-described embodiment, the first surface 2a of the semiconductor substrate 2 was ground until region R was removed. However, in the modified example, in the second precision grinding step, the first surface 2a of the semiconductor substrate 2 may be ground until groove G is removed. In this modified example, it is not possible to remove all of the region where residual stress is significantly present (i.e., region R), but the residual stress present in the semiconductor substrate 2 can be reduced.
[0041] The configurations of the manufacturing methods disclosed herein are listed below. (Composition 1) A method for manufacturing a semiconductor device, A step of forming grooves (G) on a first surface (2a) of a semiconductor substrate (2) having a plurality of element regions (3) by pressing a scribing wheel (32) along the boundary (4) of the element regions to plastically deform the semiconductor substrate, wherein a crack (5) is formed below the groove, along the groove and extending in the thickness direction of the semiconductor substrate, After the step of forming the groove, the first surface is ground, After the step of grinding the first surface, a splitting step is performed in which the semiconductor substrate is split along the boundary by pressing a splitting member (33) against the second surface (2b) of the semiconductor substrate located on the opposite side of the first surface along the boundary, A manufacturing method that includes the following features. (Configuration 2) The manufacturing method according to configuration 1, wherein the step of grinding the first surface is to grind the first surface until the groove is removed. (Composition 3) In the step of forming the groove, a stress region (R) is formed in the surface layer near the first surface of the semiconductor substrate, in which stress remains. The crack is formed in the lower part of the stress region, The manufacturing method according to configuration 1 or 2, wherein the step of grinding the first surface is to grind the first surface until the stress region is removed. (Composition 4) Prior to the step of forming the groove, the process further includes grinding the first surface, The manufacturing method according to any one of the configurations 1 to 3, wherein the step of grinding the first surface before the step of forming the grooves includes rough grinding, in which the first surface is ground with a grinding wheel coarser than the step of grinding the first surface after the step of forming the grooves. (Composition 5) The manufacturing method according to configuration 4, wherein the step of grinding the first surface before the step of forming the groove includes, after the rough grinding, fine grinding in which the first surface is ground with a grinding wheel that is more precise than the grinding wheel used in the rough grinding. (Composition 6) The process involves forming an electrode (8) on the aforementioned first surface, A step of joining solder to the electrode, A manufacturing method according to any one of the configurations 1 to 5, further comprising:
[0042] The specific examples of the technology disclosed in this specification have been described in detail above, but these are merely illustrative and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes to the specific examples described above. The technical elements described in this specification or drawings exhibit technical usefulness individually or in various combinations, and are not limited to the combinations described in the claims at the time of filing. Furthermore, the technology illustrated in this specification or drawings can achieve multiple objectives simultaneously, and achieving even one of these objectives itself constitutes technical usefulness. [Explanation of Symbols]
[0043] 2: Semiconductor substrate, 2a: First surface, 2b: Second surface, 3: Device region, 4: Planned division line, 5 : Crack, 6: Main electrode, 8: Metal film, 11: Adhesive, 12: Support plate, 13: Daishin G: Groove, R: Area
Claims
1. A method for manufacturing a semiconductor device, A step of forming grooves (G) on a first surface (2a) of a semiconductor substrate (2) having a plurality of element regions (3) by pressing a scribing wheel (32) along the boundary (4) of the element regions to plastically deform the semiconductor substrate, wherein a crack (5) is formed below the groove, along the groove and extending in the thickness direction of the semiconductor substrate, After the step of forming the groove, the first surface is ground until the groove is removed, After the step of grinding the first surface, a dividing step is performed in which the semiconductor substrate is divided along the boundary by pressing a dividing member (33) against the semiconductor substrate along the boundary against the second surface (2b) of the semiconductor substrate located on the opposite side of the first surface, A manufacturing method that includes the following features.
2. A method for manufacturing a semiconductor device, A step of forming a groove (G) on a first surface (2a) of a semiconductor substrate (2) having a plurality of element regions (3) by pressing a scribing wheel (32) along the boundary (4) of the element regions to plastically deform the semiconductor substrate, wherein a crack (5) is formed below the groove, along the groove and extending in the thickness direction of the semiconductor substrate, and a stress region (R) is formed in the surface layer near the first surface of the semiconductor substrate where stress remains, and the crack is formed below the stress region. After the step of forming the groove, the first surface is ground down to remove the stress region and to leave the crack intact. After the step of grinding the first surface, a dividing step is performed in which the semiconductor substrate is divided along the boundary by pressing a dividing member (33) against the semiconductor substrate along the boundary against the second surface (2b) of the semiconductor substrate located on the opposite side of the first surface, A manufacturing method that includes the following features.
3. Prior to the step of forming the groove, the process further includes grinding the first surface, The manufacturing method according to claim 1 or 2, wherein the step of grinding the first surface before the step of forming the grooves includes rough grinding, in which the first surface is ground with a grinding wheel coarser than the step of grinding the first surface after the step of forming the grooves.
4. The manufacturing method according to claim 3, wherein the step of grinding the first surface before the step of forming the groove includes fine grinding, in which the first surface is ground with a grinding wheel that is more precise than the grinding wheel used in the rough grinding, after the rough grinding.
5. The steps include forming an electrode (8) on the first surface, A step of joining solder to the electrode, The manufacturing method according to claim 1 or 2, further comprising: