JP2013062372A - Device wafer and method for cutting device wafer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device wafer from which each individual chip can be precisely divided.SOLUTION: A device wafer 100 includes a groove 10 formed by etching a substrate 1 in a wafer processing step and arranged on a street 3 along the extension direction of the street 3. The groove 10 is continuous with a crack 20 in a planar view of the device wafer 100, the crack being formed by being pressed relative to the device wafer 100 by a rotary knife 200 of a scriber.

Description

  The present invention relates to a device wafer having a chip region in which a plurality of elements are formed in a matrix on a substrate and a street between the adjacent chip regions, and a device wafer cutting that divides the device wafer into individual chips. Regarding the method.

Semiconductor manufacturing processes are roughly divided into a wafer manufacturing process, a wafer processing process, a chip forming process, and an assembly process.
Of these, the chip forming process is a process of cutting each device into individual chips from the wafer on which the electronic circuit is formed in the wafer processing process. As a method of cutting individual chips from the wafer, a thin blade outer peripheral blade is used. There are dicing method for dicing and scribing / braking method.

  In this dicing method, in order to cut hard silicon, a diamond blade rotating at high speed is heated, and ground conductive silicon waste is generated. For this reason, in the dicing method, a large amount of pure water for cooling the blade and discharging silicon waste from the cutting site is flowed to the wafer, and a MEMS (Micro Electro Mechanical Systems) device is applied to the patterned wafer. It cannot be applied. In particular, the low-k material used as the interlayer insulating film in the wiring is porous and lacks mechanical strength. Therefore, if the blade-dicing is performed as it is, the material is destroyed and the dicing method cannot be applied.

  In contrast, the scribing and braking method does not require pure water for cooling the blade and discharging silicon scraps, and after applying a shallow scratch on the wafer surface to the wafer surface with a diamond tool, the pressure is applied to the wafer. In this method, the wafer is divided along the scratches into chips, and compound semiconductors are mainly applied to the manufacture of small chips (5 mm square or less).

  For example, a conventional method for manufacturing a semiconductor laser includes a step of forming a semiconductor layer provided with a plurality of ridges on a GaN substrate, and a plurality of scratches at a predetermined interval from the semiconductor layer side to the GaN substrate. And a step of cleaving the GaN substrate along the scratch, wherein the semiconductor layer includes a plurality of ridge portions and a plurality of scratches before the cleavage step. At least one groove is provided on the same straight line as a plurality of scratches (for example, see Patent Document 1).

  Further, a conventional method for dividing a crystal wafer includes a step of forming a wafer dividing groove only at and near the intersection of a grid line parallel to the intersection of the cleavage plane of the crystal wafer and the surface of the crystal wafer, and the wafer dividing groove. A step of applying an external force to the grid line formed with (see, for example, Patent Document 2).

JP 2008-227461 A JP-A-8-213348

In the conventional semiconductor laser manufacturing method, the groove and the scribe mark are formed along one direction (cleavage direction, cleavage line), but the groove and the scribe mark are formed along a direction perpendicular to this direction (resonator direction). A scribe mark is not formed. For this reason, in the conventional semiconductor laser manufacturing method, since there is no groove or scribe mark in the bar, the bar cannot be accurately divided into chips, and chipping may occur at the corners of the chip.
In the conventional semiconductor laser manufacturing method, the groove is used to relieve the stress concentrated in the cleaving direction when cleaving the GaN wafer and cleave the GaN wafer in a desired direction. Without the depth, it is not a depth that allows the GaN wafer to be cleaved alone because the thin film on the GaN wafer is formed by etching.

  Further, the conventional method for dividing a crystal wafer does not form a wafer dividing groove except at the intersection of the grid lines and the vicinity thereof, and by simply pressing the wafer dividing jig from the lower surface side of the dicing tape, The chip cannot be divided with high accuracy, and chipping may occur at the edge portion of the chip.

  SUMMARY An advantage of some aspects of the invention is that it provides a device wafer and a device wafer cutting method capable of accurately dividing individual chips from a device wafer.

  The device wafer according to the present invention is formed by etching a substrate in a wafer processing step, and includes a groove portion disposed in the street along the street extending direction, and the groove portion is formed by a rotating blade of a scriber. It is continuous or overlaps with a crack formed by being pressed against the wafer in a plan view of the device wafer.

  The disclosed device wafer has an effect that the grooves disposed on the streets assist the progress of cracks formed by the rotating blades of the scriber, and the individual chips can be accurately divided from the device wafer.

(A) is a top view which shows schematic structure of the device wafer which concerns on 1st Embodiment, (b) is the elements on larger scale of the device wafer shown to Fig.1 (a). (A) is sectional drawing of the arrow AA 'line of the device wafer shown in Drawing 1 (b), and (b) is a sectional view of the arrow BB' line of the device wafer shown in Drawing 1 (b). It is a figure, (c) is sectional drawing of the arrow CC 'line of the device wafer shown in FIG.1 (b), (d) is arrow DD of the device wafer shown in FIG.1 (b). It is sectional drawing of a line, (e) is explanatory drawing for demonstrating the advancing direction of the rotary blade of a scriber. (A) is explanatory drawing for demonstrating an example with which the one end part of a groove part and a crack continue in planar view in sectional drawing shown in FIG.2 (d), (b) is a cross section shown in FIG.2 (d). It is explanatory drawing for demonstrating an example with which the other end part of a groove part and a crack in a figure continue in planar view, (c) is one end part of the groove part in the sectional view shown in FIG.2 (d), and the starting point of a crack. FIG. 4D is an explanatory diagram for explaining an example in which the groove is continuous in plan view, and FIG. 2D is a diagram illustrating an example in which the other end of the groove and the end point of the crack are continuous in plan view in the cross-sectional view shown in FIG. It is explanatory drawing for doing. (A) is a top view which shows schematic structure of the other device wafer which concerns on 1st Embodiment, (b) is the elements on larger scale of the device wafer shown to Fig.4 (a). (A) is sectional drawing of the arrow AA 'line of the device wafer shown in Drawing 4 (b), and (b) is a sectional view of the arrow BB' line of the device wafer shown in Drawing 4 (b). It is a figure, (c) is sectional drawing of the arrow CC 'line of the device wafer shown in FIG.4 (b). (A) is explanatory drawing for demonstrating an example with which the groove part and crack in the street shown in FIG.4 (b) overlap in planar view, (b) is the groove part and crack in the street shown in FIG.4 (b). And FIG. 4C is an explanatory diagram for explaining another example in which the grooves and cracks in the street shown in FIG. 4B are not superimposed in a plan view. It is a figure, (d) is explanatory drawing for demonstrating the other example in which the groove part and crack in the street shown in FIG.4 (b) do not overlap by planar view. (A) is a sectional view showing a state in which a silicon oxide film, a silicon nitride film, and a photoresist film are formed on a substrate, and (b) is a sectional view showing a state in which an opening is formed in the photoresist film, (C) is a cross-sectional view showing a state in which an opening is formed in the silicon nitride film and the photoresist film, and (d) is a cross-sectional view showing a state in which the photoresist pattern on the silicon nitride film is removed. ) Is a cross-sectional view showing a state in which a groove is formed on the substrate. (A) is explanatory drawing for demonstrating an example of a tape sticking process, (b) is explanatory drawing for demonstrating an example of a scribe process, (c) is for demonstrating an example of a breaking process. It is explanatory drawing, (d) is explanatory drawing for demonstrating an example of a bar-shaped device wafer. (A) is explanatory drawing for demonstrating the other example of a tape sticking process, (b) is explanatory drawing for demonstrating the other example of a scribe process, (c) is another figure of a break process. It is explanatory drawing for demonstrating an example, (d) is explanatory drawing for demonstrating the other example of a bar-shaped device wafer.

(First embodiment of the present invention)
As shown in FIG. 1 and FIG. 2 or FIG. 4 and FIG. 5, the device wafer 100 is a semiconductor wafer (hereinafter referred to as “substrate 1”) such as silicon (Si) or gallium nitride (GaN) manufactured through the wafer manufacturing process. In addition, a chip region 2 in which a plurality of elements are respectively formed in a matrix by a wafer processing step and a street 3 between the adjacent chip regions 2 are provided.

Further, the device wafer 100 is formed by etching the substrate 1 in a wafer processing step, and includes a groove portion 10 disposed on the street 3 along the extending direction of the street 3.
The groove 10 is formed in the device wafer 100 in a crack 20 formed by being pressed against the device wafer 100 by a rotary blade (disc-shaped cutter wheel) 200 of a scriber shown in FIG. 8B or FIG. It will be continuous or superposed in plan view.

  2, 3 and 5, the thickness of the substrate 1, the length, width and depth of the groove 10, the length, width and depth of the crack 20, and the thin film formed in the wafer processing step. Since the dimensional difference with the film thickness (silicon oxide film, silicon nitride film, oxide film, protective film, etc.) is remarkable, the illustration of the thin film is omitted, and the scale of the substrate 1, the groove 10 and the crack 20 is changed. It is shown.

  Here, the groove 10 is continuous with the crack 20 in a plan view of the device wafer 100, as shown in FIG. 3A or FIG. 3B, scribing the surface 101 side of the device wafer 100 where the elements are formed. When the crack 20 is made in the surface 101, the groove 10 and the crack 20 are integrated.

  Further, the groove 10 is continuous with the crack 20 in a plan view of the device wafer 100, as shown in FIG. 3C or FIG. 3D, the back surface 102 side facing the front surface 101 of the device wafer 100 is scribed. In other words, when the crack 20 is made in the back surface 102, the groove 10 and the crack 20 overlap with each other through the substrate 1.

  In FIG. 3C, the end portion of the groove portion 10 on the traveling direction side of the rotary blade 200 (hereinafter referred to as “one end portion 10 a”) and the starting point 20 a of the crack 20 are the substrate 1. However, if the groove 10 and the crack 20 overlap with each other via the substrate 1, the one end 10a of the groove 10 and the start point 20a of the crack 20 do not need to match. .

  Similarly, in FIG. 3D, the end portion of the groove portion 10 on the side opposite to the traveling direction of the rotary blade 200 (hereinafter referred to as “other end portion 10 b”), the end point 20 b of the crack 20, However, if the groove portion 10 and the crack 20 overlap with each other through the substrate 1, the other end portion 10b of the groove portion 10 and the end point 20b of the crack 20 coincide with each other. do not have to.

  Further, the groove 10 overlaps with the crack 20 in a plan view of the device wafer 100 when the crack 20 is formed in the surface 101 of the device wafer 100 as shown in FIG. This does not include the case where the crack 20 is formed outside the groove 10 as shown in FIG.

  Also, the groove 10 is superimposed on the crack 20 in a plan view of the device wafer 100 when the crack 20 is made on the back surface 102 of the device wafer 100 as shown in FIG. 6B. 10 and the crack 20 overlap, and does not include the case where the groove 10 and the crack 20 do not overlap via the substrate 1 as shown in FIG.

  In the device wafer 100 according to the present embodiment, the groove 10 is continuous or overlapped with the crack 20 in a plan view, whereby the vertical crack due to the crack 20 formed in the scribing process is connected to the groove 10 and is discontinuous or Compared to the case of non-overlapping, the effect of facilitating the progress of vertical cracks in the breaking process is achieved.

  The peripheral portion 201 of the rotary blade 200 of the scriber depends on the structure of the element formed in the chip region 2 (the thickness of the substrate 1), and the device wafer 100 can be divided into individual chips in the breaking process. The cutting edge is selected so as to be the length, width and depth of the crack 20. For example, in the peripheral edge 201 of the rotary blade 200, when the thickness of the substrate 1 is about 150 μm, the length of the crack 20 is about 40 μm to 50 μm, the width of the crack 20 is about 10 μm, and the depth of the crack 20 is A cutting edge formed by forming protrusions such as a V shape, a U shape, a saw shape, or a rectangular shape that are provided at predetermined intervals along the peripheral edge 201 of the rotary blade 200, which is approximately 5 μm.

In addition, the groove 10 is set to have a length equal to or greater than the length of the crack 20, a width equal to or greater than the width of the crack 20, and a depth equal to or greater than the depth of the crack 20. .
In the present embodiment, the rotary blade 200 of the scriber has a plurality of protrusions on the peripheral edge portion 201, and the groove portion 10 is formed on the crack 20 formed on the device wafer 100 by the impact of the impact of the rotary blade 200. A plurality of streets 3 are arranged along the extending direction of the streets 3 so as to be continuous in a plan view of 100 (see FIG. 1). For example, the groove 10 has a length of about 120 μm, a width of about 30 μm, and a depth of about 20 μm when the crack 20 has a length of about 40 μm, a width of about 10 μm and a depth of about 5 μm. 50 μm is set.

  However, if the groove portion 10 has a groove depth that allows the device wafer 100 to hold strength that is not divided into individual chips (bars) during the transfer of the device wafer 100 from after the scribing process to before the breaking process, one groove The groove 3 may be one continuous groove from one end to the other end of the street 3 (see FIG. 4).

  In particular, as shown in FIG. 1B, the groove portion 10 according to the present embodiment is a portion where a plurality of streets 3 intersect (hereinafter referred to as “intersection region 3 a”) in the extending direction of the streets 3. A cross-shaped cross groove portion 11 that intersects with each other, and a linear straight groove portion 12 that extends along the extending direction of the street 3 between the adjacent cross-groove portions 11 (hereinafter referred to as “straight region 3b”). .

  As described above, the device wafer 100 is provided with the cross groove portion 11 in the intersecting region 3a of the street 3 so that the vertical crack caused by the crack 20 formed in the scribing process is extended to the cross groove portion 11 (street 3) in the breaking process. It progresses in the present direction, and does not progress to the chip region 2 at the intersection region 3a of the street 3, so that it is possible to prevent the occurrence of chipping at the corner of the chip.

  Further, in the device wafer 100, by arranging the straight groove portion 12 in the straight region 3b of the street 3, the vertical crack caused by the crack 20 formed in the scribing process is extended in the extending direction of the straight groove portion 12 (street 3) in the breaking process. Thus, the straight region 3b of the street 3 does not progress to the chip region 2, and the chipping at the edge portion of the chip can be prevented.

  The planar shape of the groove 10 in a plan view of the device wafer 100 is such that the traveling direction side (one end 10a) of the rotary blade 200 in the end of the groove 10 is the rotary blade as shown in FIG. 200 is a taper shape in which the tip in the length direction of the groove portion 10 is located on the center line of the groove portion 10 along the extending direction of the street 3 in a plan view of the device wafer 100. It is preferable.

  Thus, the device wafer 100 can guide the rotary blade 200 on the center line of the street 3 because the planar shape of the groove portion 10 is tapered at the one end portion 10 a of the groove portion 10. The vertical cracks caused by the cracks 20 inserted into the wall 3 can be accurately advanced in the direction in which the streets 3 extend.

  In the breaking process, vertical cracks are propagated from one end 10a of the groove 10, but the other end 10b of the groove 10 does not contribute much to the progress of the vertical crack. Therefore, in the present embodiment, only one end 10a of the groove 10 is tapered, but if at least only one end 10a of the groove 10 is tapered, the one end 10a and the other end 10b of the groove 10 are tapered. Both end portions may be tapered.

  Moreover, in this embodiment, as shown in FIG.2 (e), since the reciprocating motion of the rotary blade 200 of a scriber reverses the advancing direction of the rotary blade 200 in the adjacent streets 3 parallel to each other, the groove portion As shown in FIG. 1B, the planar shape 10 is tapered in the opposite directions on the adjacent streets 3 parallel to each other.

  However, when the planar shape of the groove portion 10 is tapered in the opposite streets 3 parallel to each other, the direction of travel of the rotary blade 200 of the scriber and the one end portion 10a of the other end portion 10b of the groove portion 10 are It is necessary to match the direction. For this reason, it is preferable that both ends of the one end portion 10a and the other end portion 10b of the groove portion 10 are tapered so that the planar shape of the groove portion 10 can correspond to the advancing direction and the counter-advancing direction of the rotary blade 200. .

  Further, the cross-sectional shape of the groove 10 in a cross-sectional view perpendicular to the extending direction of the street 3 is tapered toward the depth direction of the device wafer 100 as shown in FIGS. 2 (b) and 2 (c). In particular, it is preferable to have a tapered shape in which the tip in the depth direction of the groove 10 is located immediately below the center line of the groove 10 along the extending direction of the street 3 in plan view of the device wafer 100.

  As described above, the device wafer 100 has a taper shape in the cross section of the groove portion 10, so that the vertical crack caused by the crack 20 formed in the scribing process can be accurately advanced in the depth direction of the device wafer 100 in the breaking process. There is an effect of being able to.

  The number and shape of the grooves 10 extending on one street 3 are appropriately set according to the dimensions of the substrate 1 (wafer diameter and thickness) and the shape of the peripheral edge 201 of the rotary blade 200. In order to protect the corners of the chip that are prone to occur, it is preferable to dispose the cross groove 11 at least in the intersection region 3 a of the street 3.

Next, a method for manufacturing the groove 10 according to the present embodiment will be described with reference to FIG.
In the following description, the case where the groove portion 10 is simultaneously formed when the silicon oxide film 4 and the silicon nitride film 5 formed on the substrate 1 are etched will be described. However, the groove portion 10 is formed by etching in the wafer processing step. If it forms, it will not be restricted to this process.

  First, a silicon oxide film 4 is formed on the substrate 1 by a thermal oxidation method or the like, a silicon nitride film 5 is deposited on the silicon oxide film 4 by a CVD (chemical vapor deposition) method or the like, and a spin coating method or the like is performed. Then, a photoresist film 6 is applied on the silicon nitride film 5 (FIG. 7A).

  The silicon oxide film 4 relieves stress generated at the interface between the silicon nitride film 5 and the substrate 1 and prevents the occurrence of defects such as dislocations on the surface of the substrate 1 due to the stress. In addition, the silicon nitride film 5 is used as a mask for preventing the surface of the substrate 1 from being oxidized when the substrate 1 in the street 3 is etched to form the groove 10.

  Then, a photoresist pattern 7 having an opening at a desired position in the intersection region 3a and the straight region 3b of the street 3 is formed on the silicon nitride film 5 by a lithography process (FIG. 7B).

  Thereafter, by selectively etching the silicon nitride film 5 on the street 3 and the silicon oxide film 4 therebelow by dry etching using the photoresist pattern 7 as a mask, the intersection region 3a and the straight region 3b of the street 3 are obtained. The opening 8 is formed at a desired position in FIG. 7 (c).

  Then, the photoresist pattern 7 is removed from the silicon nitride film 5 by ashing or the like (FIG. 7D), and the groove 10 is formed in the substrate 1 by dry etching using the silicon nitride film 5 as a mask (FIG. 7). (E)).

  As described above, the device wafer 100 is formed by etching the substrate 1 in the wafer processing process, so that the number of manufacturing steps of the device wafer 100 can be reduced without adding a special process for forming the groove 10. And the effect that the increase in manufacturing cost can be suppressed is produced.

Next, a cutting method for dividing the device wafer 100 into individual chips will be described with reference to FIGS.
Note that a process (wafer processing step) in which elements are formed on the substrate 1 and a plurality of grooves 10 are formed in the streets 3 along the extending direction of the streets 3 (wafer processing step) except for the manufacturing method of the grooves 10 described above. Since this is a normal wafer processing step, description thereof is omitted.

First, the case where the surface 101 of the device wafer 100 is scribed will be described with reference to FIG.
After the wafer processing step, a wafer mounter (not shown) aligns the wafer frame 300 and the device wafer 100, attaches a protective tape 400, which is an adhesive resin tape, to the back surface 102 of the device wafer 100, and attaches the device wafer 100 to the device wafer 100. It fixes to the wafer frame 300 (tape pasting process, Fig.8 (a)).

  Then, a scriber (not shown) rolls the rotary blade 200 against the surface 101 of the device wafer 100 in a state of being pressed against the groove portion 10 (FIG. 8B), and moves to the groove portion 10 in a plan view of the device wafer 100. A scribe line composed of continuous or overlapping cracks 20 is formed on the surface 101 of the device wafer 100 (scribing step, FIG. 3A, FIG. 3B, or FIG. 6A).

  A breaking device (not shown) places a blade 500 on the back surface 102 side of the device wafer 100 along the groove 10 and crack 20 (scribe line) extending to the street 3, and stress is applied directly below the scribe line by the blade 500. Is added (FIG. 8C). Then, the break device propagates vertical cracks toward the front surface 101 and the back surface 102 of the device wafer 100 to form the device wafer 100 in a bar shape (FIG. 8D).

  Similarly, the break device arranges a blade 500 on the back surface 102 side of the bar-shaped device wafer 100 along the groove 10 and the crack 20 (scribe line) extending to the street 3, and the blade 500 directly below the scribe line. Stress is applied to (Fig. 8 (c)). Then, the break device develops vertical cracks on the front surface 101 and back surface 102 sides of the bar-shaped device wafer 100, and divides the bar-shaped device wafer 100 into individual chips (break process).

Next, a case where the back surface 102 of the device wafer 100 is scribed will be described.
After the wafer processing step, a wafer mounter (not shown) aligns the wafer frame 300 and the device wafer 100, attaches a protective tape 400 to the surface 101 of the device wafer 100, and fixes the device wafer 100 to the wafer frame 300. (Tape applying process, FIG. 9A).

A scriber (not shown) scans laser light from the back surface 102 side of the device wafer 100 using a transparent infrared laser on the semiconductor that is the material of the substrate 1, and the chip region 2 and street 3 on the front surface 101 of the device wafer 100. Is recognized, and the rotary blade 200 is aligned with the back surface 102 of the device wafer 100.
Further, the scriber rolls with the rotary blade 200 in contact with the back surface 102 of the device wafer 100 so as to correspond to the groove portion 10 (FIG. 9B), and continues to the groove portion 10 in a plan view of the device wafer 100 or A scribe line composed of overlapping cracks 20 is formed on the back surface 102 of the device wafer 100 (scribing step, FIG. 3C, FIG. 3D, or FIG. 6B).

  A breaking device (not shown) places a blade 500 on the surface 101 side of the device wafer 100 along the groove 10 and the crack 20 (scribe line), and applies stress directly below the scribe line by the blade 500 (FIG. 9 ( c)). Then, the breaking device develops vertical cracks on the back surface 102 and the front surface 101 side of the device wafer 100, thereby forming the device wafer 100 in a bar shape (FIG. 9D).

  Similarly, the breaking device arranges a blade 500 on the surface 101 side of the bar-shaped device wafer 100 along the groove 10 and the crack 20 (scribe line), and applies stress directly below the scribe line by the blade 500 (FIG. 9 (c)). Then, the breaking device develops vertical cracks on the back surface 102 and the front surface 101 side of the bar-shaped device wafer 100, and divides the bar-shaped device wafer 100 into individual chips (breaking process).

  As described above, when scribing the back surface 102 of the device wafer 100, the groove 10 formed on the front surface 101 of the device wafer 100 is connected to the vertical crack that has developed from the crack 20 formed on the back surface 102 of the device wafer 100. There is an effect that the device wafer 100 is subdivided into individual chips.

  Further, when scribing the back surface 102 of the device wafer 100, the protective tape 400 attached to the front surface 101 of the device wafer 100 serves as a buffer material against an external force applied to the elements in the chip region 2 and is generated in the scribing process. There is an effect that the particles can be prevented from adhering to the elements due to chipping of the substrate 1 or peeling of the thin film, and the elements in the chip region 2 can be protected.

DESCRIPTION OF SYMBOLS 1 Substrate 2 Chip area 3 Street 3a Crossing area 3b Linear area 4 Silicon oxide film 5 Silicon nitride film 6 Photoresist film 7 Photoresist pattern 8 Opening 10 Groove 10a One end 10b Other end 11 Cross groove 12 Linear groove 20 Crack 20a Start point 20b End point 100 Device wafer 101 Front surface 102 Back surface 200 Rotary blade 201 Peripheral portion 300 Wafer frame 400 Protective tape 500 Blade

Claims (7)

In a device wafer comprising a chip region in which a plurality of elements are respectively formed in a matrix on a substrate, and a street between the adjacent chip regions,
Formed by etching the substrate in a wafer processing step, comprising a groove portion disposed in the street along the extending direction of the street,
The device wafer, wherein the groove portion is continuous or overlapped with a crack formed by being pressed against the device wafer by a rotary blade of a scriber in a plan view of the device wafer. The device wafer according to claim 1,
The device wafer, wherein the groove portion includes a cross-shaped cross groove portion that intersects each other along the extending direction of each street at a portion where the plurality of streets intersect. The device wafer according to claim 2, wherein
The device wafer, wherein the groove part includes a linear groove part having a linear shape extending along the street extending direction between the adjacent cross groove parts. In the device wafer according to any one of claims 1 to 3,
The planar shape of the groove portion in a plan view of the device wafer is a taper shape in which the traveling direction side of the rotary blade in the end portion of the groove portion faces the traveling direction of the rotary blade. In the device wafer according to any one of claims 1 to 4,
A device wafer, wherein a cross-sectional shape of the groove portion in a cross-sectional view of the device wafer in a direction perpendicular to an extending direction of the street is a tapered shape toward a depth direction of the device wafer. In a device wafer cutting method for dividing a device wafer comprising chip regions each having a plurality of elements formed in a matrix on a substrate and streets between the adjacent chip regions into individual chips,
A wafer processing step of forming the element on the substrate and forming a groove in the street along the extending direction of the street;
A scribing step of rolling a rotating blade of a scriber against the device wafer in a pressure contact state to form a crack that is continuous or overlapped with the groove portion in a plan view of the device wafer;
A breaking step of dividing the device wafer along the scribe line consisting of the cracks and dividing it into individual chips;
A method for cutting a device wafer, comprising: The device wafer cutting method according to claim 6, wherein:
Between the wafer processing step and the scribe step, including a tape attaching step of attaching a protective tape to the surface of the device wafer on which the element is formed,
The method for cutting a device wafer, wherein the scribing step forms the crack on the back surface facing the front surface of the device wafer.

Images