CHIP MANUFACTURING PROCESSES

Abstract

A manufacturing method for chips (23), wherein a workpiece (11) having a plurality of planned dividing lines (15) on a front surface (11a) is divided along each planned dividing line (15) to produce the chips (23) from the workpiece (11), the manufacturing method for chips (23) comprising: a cutting step of causing a cutting blade (48a) to cut into the workpiece (11), for which one side of the front surface (11a) of the workpiece (11) is held by a holding table (26, 98) in such a manner that one side of a rear surface (11b) of the workpiece (11) is exposed, and forming a cutting groove, which does not reach the front surface (11a) of the workpiece (11), on the side of the rear surface (11b) of the workpiece (11) along each planned parting line (15); an adhesion step of adhering an expansion film (17) to the workpiece (11); and a dividing step of dividing the workpiece (11) along each planned dividing line (15) by expanding the expanding film (17) after the adhering step and the cutting step to form the chips (23) from the workpiece (11), wherein the workpiece (11) comprises a silicon carbide substrate and a metal layer (27) arranged on a surface of the silicon carbide substrate and arranged on the side of the rear surface (11b) of the workpiece (11), and the cutting blade (48a) is caused to cut into the workpiece (11) from the side of the rear surface (11b) of the workpiece (11) to form the cutting grooves in the cutting step.

Description

TECHNICAL BACKGROUNDTechnical field

[0001] The present invention relates to a chip manufacturing method in which a workpiece having a plurality of planned dividing lines on a front surface is divided along each planned dividing line to manufacture chips from the workpiece. Description of related technology

[0002] In electronic equipment such as mobile phones and personal computers, a component chip is used with a component such as an electronic circuit. The component chip is manufactured, for example, by dividing the front surface of a wafer made of a semiconductor material such as silicon into multiple regions along planned dividing lines (streets), forming a component in each region, and then dividing the wafer along these planned dividing lines. When a plate-shaped workpiece such as a wafer is divided into small parts such as component chips, a laser processing device is used, for example.In particular, a method has been proposed in which a modified layer is formed within a workpiece by irradiating a laser beam along planned dividing lines of the workpiece in such a manner that the focal point of the laser beam having a wavelength such that it is transmitted through the workpiece is positioned within the workpiece and thereafter an external force is applied to the workpiece to divide the wafer (see, for example, Japanese Patent No. 3408805).

[0003] However, if the workpiece is relatively thick, multiple modified layers must be formed in the thickness direction of the workpiece, and therefore, external force must be applied to each layer using a breaking device. Therefore, in some cases, it takes a long time to break the workpiece. Moreover, in some cases, if the size of the designed component chip is relatively small, it also takes a long time to divide the workpiece. Meanwhile, in some cases, the workpiece is divided using a cutting device instead of the laser processing device. The cutting device includes a spindle serving as a rotation axis, and an annular abrasive tool called a cutting blade is attached to this spindle.By causing the cutting blade to cut into the workpiece along the planned dividing lines while rotating the cutting blade at a high speed, the workpiece is cut from the back surface to the front surface to be divided into multiple component chips. PRESENTATION OF THE INVENTION

[0004] However, in the case of dividing the workpiece using the cutting blade, the width of the planned parting line must be determined in advance, taking into account the cutting groove (i.e., a notch width) formed in the workpiece by cutting and the amount of chipping that occurs due to cutting. In particular, when components are formed on the front surface of the workpiece, the width of the planned parting lines, an arrangement of the components, etc., must be designed taking into account the notch width formed on the front surface. The width of the planned parting lines affects the number of chips obtained per one workpiece, for example. If the width of the planned parting lines is larger, the number of chips obtained per one workpiece becomes smaller.Therefore, the width of the planned dividing lines must be reduced to increase the number of chips obtained per workpiece. The present invention was conceived in view of this problem and aims to provide a chip manufacturing method that allows the notch width to be made small even when using a cutting device.

[0005] According to one aspect of the present invention, there is provided a chip manufacturing method in which a workpiece having a plurality of planned dividing lines on a front surface is divided along each planned dividing line to manufacture chips from the workpiece.The chip manufacturing method includes a cutting step of causing a cutting blade to cut into the workpiece, for which the front surface side of the workpiece is held by a holding table in such a manner as to expose the rear surface side of the workpiece, and forming a cutting groove, which does not reach the front surface of the workpiece, on the rear surface side of the workpiece along each scheduled dividing line, an adhering step of adhering an expanding film to the workpiece, and a dividing step of dividing the workpiece along each scheduled dividing line by expanding the expanding film after the adhering step and the cutting step, to form the chips from the workpiece.

[0006] Preferably, the chip manufacturing method further includes a cutting position detection step of imaging the front surface of the workpiece with an infrared camera from the back surface of the workpiece and detecting at least one planned dividing line before the cutting step. Furthermore, the holding table preferably includes a holding member that is at least partially transparent from a front surface of the holding member to a back surface of the holding member, and the chip manufacturing method further includes a cutting position detection step of imaging the front surface of the workpiece with a visible light camera through the holding member and detecting at least one planned dividing line before the cutting step.

[0007] Furthermore, the workpiece includes a silicon carbide substrate and a metal layer disposed on a surface of the silicon carbide substrate and disposed on the back surface side of the workpiece. Preferably, the cutting blade is made to cut into the workpiece from the back surface side of the workpiece to form the cutting grooves in the cutting step. Preferably, in addition, the adhesion step is performed before the cutting step, the expanding film is adhered to the front surface side of the workpiece in the adhesion step, and the cutting grooves are formed in the workpiece in the cutting step in a state where the front surface side of the workpiece is formed by the holding table with the expanding film interposed.

[0008] Furthermore, it is preferable that the adhesion step is performed after the cutting step, and the expansion film is adhered to the back surface side of the workpiece in the adhesion step. Furthermore, it is preferable that the cutting blade has a pair of inclined surfaces or a curved surface on an outer peripheral part, and the workpiece is cut by the cutting blade in the cutting step. In addition, it is preferable that the workpiece has a circuit layer with insulation layers and interconnection layers alternately stacked on the front surface side, and the cutting grooves that do not reach the circuit layer are formed in the workpiece in the cutting step.

[0009] In the cutting step according to one aspect of the present invention, the cutting groove, which does not reach the front surface of the workpiece, is formed along each planned parting line. Furthermore, in the adhering step, the expanding film is adhered to the workpiece. Then, in the dividing step after the adhering step and the cutting step, the workpiece is divided along each planned parting line by expanding the expanding film, and chips are formed from the workpiece. As above, the workpiece is divided in the cutting step without cutting the front surface of the workpiece. Therefore, in the case of breakage of the workpiece from the front surface to the back surface in the cutting step, the notch width of the front surface side can be made smaller. For this reason, the width of the planned parting lines can be made small.

[0010] The above and other objects, features and advantages of the present invention and the mode for carrying them out will become more apparent and the invention itself will be best understood by studying the following description and the appended claims with reference to the attached drawings which show preferred embodiments of the invention. SHORT DESCRIPTION OF THE FIGURES Fig. 1A is a perspective view of a workpiece; Fig. 1B is a perspective view of a workpiece unit; Fig. 2 is a perspective view of a cutting device; Fig. 3A is a partial sectional side view of a workpiece, etc., in a holding step; Fig. 3B is a partial sectional side view of the workpiece, etc., in a cutting position detecting step; Fig. 4A is a partial sectional side view of the workpiece, etc., in a cutting step; Fig. 4B is a sectional view of a cutting blade before use; Fig. 4C is a sectional view of the cutting blade after use for a certain period of time; Fig. 5A is a sectional view of a portion of the workpiece having cutting grooves having a first depth; Fig. 5B is a sectional view of a portion of the workpiece having the cut grooves having a second depth; Fig. 6A is a partially sectional side view of a band expanding device; Fig. 6B is a partial sectional side view of the workpiece, etc., in a dividing step; Fig. 7 is a sectional view of several chips; Fig. 8 is a flowchart of a manufacturing method of chips in a first embodiment; Fig. 9 is a flowchart of a manufacturing method of chips in a second embodiment; Fig. 10A is a perspective view of a stacked laminated body with the workpiece; Fig. 10B is a perspective view illustrating how a front surface protective tape is removed from the workpiece; Fig. 11 is a perspective view of a workpiece unit according to a third embodiment; Fig. 12 is a perspective view of a cutting device in the third embodiment; Fig. 13 is a perspective view of a Y-axis moving table, etc.; Fig. 14A is a sectional view of a part of the cutting device; Fig. 14B is a sectional view in which a region A in Fig. 14A is enlarged; Fig. 15 is a sectional view of a portion of a workpiece having the cutting grooves having a third depth; Fig. 16A is a perspective view of a stacked laminated body with the workpiece; and Fig. Figure 16B is a perspective view showing how the front surface protective tape is removed from the workpiece. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Embodiments according to one aspect of the present invention will be described with reference to the accompanying drawings. First, an embodiment example of a workpiece 11 will be described. Fig. 1A is a perspective view of the workpiece 11. The workpiece 11 has a substantially circular disk shape and has a front surface 11a and a back surface 11b. The workpiece 11 has a substrate 11c formed of silicon (Si) (see Fig. 5A etc.) On the side of the front surface 11a of the workpiece 11, a circuit layer 11d (see Fig. 5A, etc.) is formed in contact with the substrate 11c. The circuit layer 11d comprises insulator layers with a low dielectric constant (essentially referred to as low-k material layers) and interconnect layers that are alternately layered.

[0012] The front surface 11a side of the workpiece 11 is divided into a plurality of regions by a plurality of planned dividing lines (streets) 15 arranged in a lattice pattern to intersect each other. A component 13 such as an integrated circuit (IC), a large-scale integration (LSI), and a light-emitting diode (LED) is formed in each of the plurality of regions. Each component 13 is formed from the above-described substrate 11c and a circuit layer 11d. There are no restrictions on the material, shape, structure, size, etc. of the substrate 11c. For example, the substrate 11c may be formed of a material such as another semiconductor (GaAs, InP, GaN, or the like), sapphire, glass, ceramic, plastic, or multiple oxides (LiNbO3, LiTaO3). Furthermore, there are no restrictions on the type, number, shape, structure, size, arrangement, etc. of the components 13.

[0013] When the workpiece 11 is machined, first, the workpiece 11 and an annular frame 19 made of a metal are integrated by a circular division band 17 made of a material such as plastic to form a workpiece unit 21. Fig. 1B is a perspective view of the workpiece unit 21. The annular frame 19 has a circular opening portion larger than the diameter of the workpiece 11, and the workpiece 11 is arranged in this opening portion. The dividing band 17 connects the front surface 11a side of the workpiece 11 and one surface of the annular frame 19.

[0014] The dividing band 17 has a diameter larger than the opening of the annular frame 19. The dividing band 17 is an expansion film that has expandability and can be easily deformed in response to tensile stress. The dividing band 17 is a film made of a plastic and is obtained by laminating an adhesive layer (bonding agent layer) on a base layer formed of, for example, the plastic. The base layer is formed of a plastic such as polyolefin (PO), polyvinyl chloride (PVC), or polyethylene terephthalate (PET), and the bonding agent layer is formed of a plastic such as a rubber-based plastic, an acrylic plastic, or a silicone-based plastic.

[0015] The adhesive layer is, for example, an ultraviolet-curable plastic and is arranged on the entire surface of the base layer. The ultraviolet-curable plastic has a strong adhesive force before being irradiated with ultraviolet light, but the adhesive force decreases when the plastic is irradiated with ultraviolet light (UV). For example, if the adhesive layer is irradiated with ultraviolet light through a base layer through which ultraviolet light can be transmitted, the workpiece 11 can be easily removed from the dividing band 17. By machining and dividing the workpiece 11 along each planned dividing line 15, a plurality of chips 23 (see Fig. 7), i.e., component chips, each having the component 13, are manufactured. Different methods can be used for processing the workpiece 11. For example, a cutting device 2 is used, which cuts the workpiece 11 along the predetermined parting lines 15 with a cutting blade.

[0016] Fig. 2 is a perspective view of the cutting device. An X-axis direction (machining feed direction), a Y-axis direction (index feed direction), and a Z-axis direction (height direction, cut-in feed direction), which will be used in the following description, are perpendicular to each other. The cutting device 2 includes a base 4 supporting the respective structures. An X-axis moving unit 10 is arranged on a part of the upper surface of the base 4. The X-axis moving unit 10 includes a pair of X-axis guide rails 12 parallel to the X-axis direction. Each X-axis guide rail 12 is fixed to the upper surface of the base 4.

[0017] An X-axis motion table 14 is slidably mounted on each X-axis guide rail 12. A nut member (not shown) is disposed on the rear surface (bottom surface) of the X-axis motion table 14, and an X-axis ball screw 16 arranged parallel to the X-axis guide rails 12 is rotatably connected to this nut member. An X-axis pulse motor 18 is coupled to one end of the ball screw 16. When the X-axis ball screw 16 is rotated by the X-axis pulse motor 18, the X-axis motion table 14 moves along the X-axis guide rails 12 in the X-axis direction.

[0018] A support base 20 having a cylindrical shape is arranged substantially at the center of the upper surface of the X-axis moving table 14. Furthermore, four support columns 22 having a prismatic column shape are fixed to four corners of the upper surface of the X-axis moving table 14, and a table cover 24 having a rectangular shape is fixed to the upper side of the support columns 22. A chuck table (holding table) 26 for sucking and holding the workpiece 11 is arranged on the upper surface of the table cover 24. Here, the structure of the chuck table 26 will be described with reference to Fig. 3A are described.

[0019] The chuck table 26 includes a frame body 26b made of a metal material such as stainless steel and having a cylindrical shape. A recessed portion having a substantially circular disc shape is formed at the upper portion of the frame body 26b. At the lower portion of this recessed portion, a plurality of bottom flow paths 26c1 are formed, each formed in a straight line fashion along the radial direction of the recessed portion. Furthermore, at the central portion of the bottom portion of the recessed portion, a center flow path 26c2 is formed, penetrating the frame body 26b from the bottom portion of the recessed portion to the rear surface of the frame body 26b. A holding member 26d having a circular disc shape is fitted into the recessed portion of the frame body 26b, and the holding member 26d is fixed to the recessed portion by an adhesive or the like.The holding member 26d is a porous plate formed of a porous member such as a porous ceramic, and is connected to a suction source 28a having an ejector or the like through the bottom flow paths 26c1 and the central flow path 26c2.

[0020] A valve 28b is arranged between the central flow path 26c2 and the suction source 28a. By placing the valve 28b in the open state, a suction force is generated on the front surface of the holding member 26d (i.e., the holding surface 26a). Four clamp units 26e for fixing the annular frame 19 from four sides are arranged around the chuck table 26. The chuck table 26 is connected to a rotation drive source (not shown), such as a motor, arranged in the support base 20, and rotates, for example, about a rotation axis substantially parallel to the Z-axis direction.

[0021] Here the description of the cutting device is repeated with reference to Fig. 2. A Y-axis moving unit 30 is arranged on a different part of the upper surface of the base 4 from the X-axis moving unit 10. The Y-axis moving unit 30 includes a pair of Y-axis guide rails 32 parallel to the Y-axis direction. Each Y-axis guide rail 32 is fixed to the upper surface of the base 4. A Y-axis moving block 34 is arranged on the pair of Y-axis guide rails 32. The Y-axis moving block 34 includes a plate-shaped horizontal plate 34a parallel to the X-axis and Y-axis directions. The bottom surface side of the horizontal plate 34a is slidably attached to each Y-axis guide rail 32.

[0022] A nut member (not shown) is disposed on the lower surface side of the horizontal plate 34a, and a Y-axis ball screw 36 arranged parallel to the Y-axis guide rails 32 is rotatably connected to this nut member. A Y-axis pulse motor 38 is coupled to one end of the Y-axis ball screw 36. When the Y-axis ball screw 36 is rotated by the Y-axis pulse motor 38, the horizontal plate 34a moves along the Y-axis guide rails 32 in the Y-axis direction.

[0023] A plate-shaped vertical plate 34b parallel to the Y-axis and Z-axis directions is fixed to one end of the horizontal plate 34a in the X-axis direction. A Z-axis moving unit 40 is arranged on the surface of the vertical plate 34b on the other end side, located on the opposite side in the X-axis direction from one end of the horizontal plate 34a. The Z-axis moving unit 40 includes a pair of Z-axis guide rails (not shown) parallel to the Z-axis direction. Each Z-axis guide rail is fixed to the surface of the vertical plate 34b on the other end side in the X-axis direction. A holder 44 that fixes a cutting unit 48 is slidably mounted on the pair of Z-axis guide rails.

[0024] A nut portion (not shown) is disposed on the rear surface side (i.e., one side of the vertical plate 34b) of the holder 44, and a Z-axis ball screw (not shown) arranged parallel to the Z-axis guide rails is rotatably connected to this nut portion. A Z-axis pulse motor 42 is coupled to one end of the Z-axis ball screw. When the Z-axis ball screw is rotated by the Z-axis pulse motor 42, the holder 44 moves along the Z-axis guide rails in the Z-axis direction. The holder 44 has a recessed portion having a circular column shape. The holder 44 is arranged in such a manner that the height direction of the recessed portion having the circular column shape is parallel to the Y-axis direction. A spindle housing 46, which forms the cutting unit 48, is attached to this recess part.The spindle housing 46 has a cylindrical shape and is fixed to the recessed part of the holder 44.

[0025] The cutting unit 48 includes a spindle (not shown) serving as a rotational axis parallel to the Y-axis direction. The spindle is partially rotatably housed in the spindle housing 46. However, one end of the spindle is exposed from the spindle housing 46. A cutting blade 48a, which is obtained by fixing abrasive grains with a bonding agent and has a circular ring shape, is attached to one end of the spindle (see Fig. 4A). Meanwhile, a rotary drive source (not shown) such as a motor is coupled to the other end of the spindle. When the rotary drive source is driven, the cutting blade 48a rotates, with the spindle as the rotation axis.

[0026] A nozzle for supplying a cutting fluid such as clean water to the workpiece 11 and the cutting blade 48a is arranged next to the cutting blade 48a. As the cutting fluid is supplied from the nozzle, the rotating cutting blade 48a is caused to cut into the workpiece 11 held by the chuck table 26. This allows the workpiece 11 to be cut. A camera unit 50 for photographing the workpiece 11 or the like held by the chuck table 26 is arranged at a position next to the cutting unit 48. The camera unit 50 is attached to a lateral side of the spindle housing 46 and moves together with the cutting unit 48 in the Y-axis and Z-axis directions by the Y-axis moving unit 30 and the Z-axis moving unit 40.

[0027] The camera unit 50, in a first embodiment, includes an infrared (IR) camera. The infrared camera includes an objective lens (not shown) disposed on the side of the support surface 26a and an imaging element (not shown) that images a subject through the objective lens. A control unit (not shown) is disposed in the cutting device 2, which controls the operation of the X-axis movement unit 10, the chuck table 26, the Y-axis movement unit 30, the Z-axis movement unit 40, the cutting unit 48, the camera unit 50, and so on. The control unit is configured by a computer including a processing device such as a central processing unit (CPU) and a storage device such as a flash memory.The processing device is made to operate according to software such as a program stored in the storage device, and thereby the control unit acts as a special means in which the software and the processing device (hardware resources) cooperate.

[0028] Next, a machining process of the workpiece 11, etc. according to the first embodiment will be described. In the first embodiment, first, as shown in Fig. 1B, the workpiece 11 is placed in the opening portion of the annular frame 19, and the dividing band 17 is adhered to the front surface 11a side of the workpiece 11 and one surface of the annular frame 19 (adhering step (S10)). Thus, the workpiece unit 21 is formed in which the rear surface 11b of the workpiece 11 is exposed. After the adhering step (S10), the dividing band 17 side in the workpiece unit 21 is sucked and held by the chuck table 26 (holding step (S20)). Fig. 3A is a partial sectional side view of the workpiece 11, etc. in the holding step (S20).

[0029] In the holding step (S20), the valve 28b is first closed, and the workpiece 11 is placed on the chuck table 26 in such a manner that the rear surface 11b side of the workpiece 11 is exposed upward. Then, the annular frame 19 is fixed by the clamp units 26e. In addition, the suction source 28a is actuated, and the valve 28b is opened. Thus, the front surface 11a side of the workpiece 11 is held by the chuck table 26 with the dividing tape (expanding film) 17 interposed. After the holding step (S20), the front surface 11a of the workpiece 11 is imaged by the camera unit 50 from the rear surface 11b side of the workpiece 11, and at least one planned dividing line 15 is detected (cutting position detection step (S30)). Fig. 3B is a partial sectional side view of the workpiece 11, etc. in the cutting position detection step (S30).

[0030] The camera unit 50 in the present embodiment is an infrared camera and can therefore image the front surface 11a side with a transmission through the substrate 11c. The image obtained by this image is stored in the memory device of the control unit. This image includes an alignment mark existing on the front surface 11a side, which is referred to as a key pattern or the like. The coordinates of at least one planned parting line 15 are detected based on the coordinates of this alignment mark. In this way, the image obtained by the camera unit 50 is used for alignment, etc., of the workpiece 11.

[0031] After the cutting position detection step (S30), the cutting blade 48a is caused to cut into the workpiece 11 in the state where the front surface 11a side is held by the chuck table 26, and a cutting groove is formed along each planned parting line 15 (cutting step (S40)). Fig. 4A is a partial sectional side view of the workpiece 11, etc., in the cutting step (S40). In the cutting step (S40), first, the chuck table 26 is rotated based on an image obtained by imaging with the camera unit 50 using the rotation drive source in such a manner that the planned parting line 15 becomes parallel to the X-axis direction. Then, the position of the cutting blade 48a is aligned with the upper side of an extension line of the planned parting line 15, which is the target of machining, using the Y-axis moving unit 30.

[0032] Then, the position of the cutting unit 48 in the Z-axis direction is adjusted using the Z-axis moving unit 40 in such a way that the height of the lower end of the cutting blade 48a is adjusted between the front surface 11a and the rear surface 11b of the workpiece 11. Thereby, the height of the lower end of the cutting blade 48a is adjusted in such a way that the cutting depth of the cutting blade 48a does not reach the front surface 11a of the workpiece 11. Thereafter, the chuck table 26 is moved relative to the cutting blade 48a using the X-axis moving unit 10 in parallel to the X-axis direction. As a result, the workpiece 11 is cut along the one planned parting line 15, and a cutting groove is formed on the rear surface 11b side of the workpiece 11.

[0033] After the cutting groove is formed along one planned parting line 15, the workpiece 11 is indexed by a predetermined index amount using the Y-axis moving unit 30. Then, a cutting groove similarly formed along another planned parting line 15 adjacent to the cut planned parting line 15 in the Y-axis direction. This process is repeated until the workpiece 11 is machined along all the planned parting lines 15. As a result, cutting grooves 15a are formed along all the planned parting lines 15 on the rear surface 11b side, without reaching the front surface 11a side.

[0034] The cutting blade 48 is formed into a circular disc shape (circular ring shape) by mixing, for example, abrasive grains of diamond, cubic boron nitride (cBN), or the like with a binder made of a metal, ceramic, plastic, or the like. However, there is no limitation on the binder and abrasive grains constituting the cutting blade 48a, and the binder and abrasive grains are selected and changed according to the specifications, etc., of the cutting blade 48a. Fig. 4B is a sectional view of the cutting blade 48a before use. As shown in Fig. 4B, annular inclined surfaces 48c are formed on an outer peripheral part 48b from the outer peripheries of the circular disc-shaped (circular annular) end surfaces of the cutting blade 48a to the outer periphery located substantially at the center of the cutting blade 48a in the thickness direction. As above, at the outer peripheral part 48b, the pair of inclined surfaces 48c are formed substantially line-symmetrically with respect to a plane passing substantially through the center of the cutting blade 48a in the thickness direction and parallel to the circular disc-shaped (circular annular) end surfaces of the cutting blade 48a.

[0035] When the cutting blade 48a is viewed in a section passing through the diameter of the cutting blade 48a and parallel to the thickness direction of the cutting blade 48a, a projection portion having a substantially V-shape is formed by the pair of inclined surfaces 48c on the outer peripheral portion 48b. When the workpiece 11 is cut by the cutting blade 48a, the cutting groove 15a having a substantially V-shape is formed in a sectional view of cutting the workpiece 11 with a plane orthogonal to a straight line parallel to the intended parting line 15. In the cutting blade 48a, the corner portion has been rounded due to wear after use for a certain period of time. Therefore, the outer peripheral portion 48b may not have the pair of inclined surfaces 48c but a curved surface 48d formed therein, as shown in FIG. Fig. 4C is shown. Fig. Figure 4C is a sectional view of the cutting blade 48a after use over a certain period of time. Fig. 4C, the outline of the outer peripheral part 48b before use is shown by a dashed line.

[0036] As in Fig. 4C, the annular curved surface 48d, which forms a protrusion shape from the center of the cutting blade 48a toward the outer peripheral side, is formed on the outer peripheral part 48b. That is, the outline of the outer peripheral part 48b has a substantially U-shape when the cutting blade 48a is viewed in a section passing through the diameter of the cutting blade 48a and parallel to the thickness direction of the cutting blade 48a. When the workpiece 11 is cut by the cutting blade 48a, the cutting groove 15a having a substantially U-shape is formed in a sectional view of cutting the workpiece 11 with a plane perpendicular to a straight line parallel to the projected parting line 15.

[0037] In Fig. 5A and Fig. 5B shows the cutting grooves 15a formed in the cutting step (S40). Fig. Figure 5A is a sectional view of a portion of the workpiece 11 having the cutting grooves 15a having a first depth and Fig. 5B is a sectional view of a portion of the workpiece 11 having the cutting grooves 15a with a second depth. In Fig. 5A and Fig. 5B shows the case where the cutting grooves 15a have a substantially U-shape, and the dividing band 17 on the side of the front surface 11a of the workpiece 11 is omitted. The bottom parts of the Fig. The cutting grooves 15a having the first depth shown in FIG. 5A are arranged at a boundary 11e between the substrate 11c and the circuit layer 11d. By preventing the circuit layer 11d from being cut by the cutting blade 48a as above, a peeling phenomenon in which the cutting blade 48a cuts the circuit layer 11d and the circuit layer 11d is removed from the substrate 11c can be prevented.

[0038] The bottom part of the cutting groove 15a can be separated from the boundary 11e by a predetermined distance. The bottom parts of the Fig. The cutting grooves 15a having the second depth shown in FIG. 5B do not reach the circuit layer 11d and are separated from the boundary 11e by a predetermined distance of several micrometers up to and including 10 µm. By separating the bottom portions of the cutting grooves 15a from the boundary 11e by the predetermined distance to form uncut portions of the substrate 11c, as above, the possibility of cutting the circuit layer 11d by the cutting blade 48a can be further reduced compared to the case where the bottom portions of the cutting grooves 15a are arranged at the boundary 11e. Therefore, the peeling phenomenon can be more reliably prevented.

[0039] Furthermore, by forming the uncut portions of the substrate 11c at the planned parting lines 15, the mechanical strength of the workpiece 11 can be improved compared to the case where the bottom portions of the cutting grooves 15a are located at the boundary 11e. Therefore, it is possible to reduce the possibility of a crack occurring at an unexpected position or in an unexpected direction in the substrate 11c during machining, transportation, etc. of the workpiece unit 21. When the substrate 11c is formed of silicon, if the distance from the bottom portion of the cutting groove 15a to the boundary 11e is set longer than 10 µm, there is a possibility of a crack occurring at an unexpected position or in an unexpected direction in the substrate 11c. Thus, it is preferable to set this distance equal to or less than 10 µm.

[0040] After the cutting step (S40), the workpiece 11 is divided along each planned dividing line 15 (dividing step (S50)). In the dividing step (S50), for example, a strip expanding device 52 is used to expand the dividing strip (expanding film) 17. Fig. 6A is a partial sectional side view of the belt expanding device 52. The belt expanding device 52 includes a drum 54 having a larger diameter than the diameter of the workpiece 11 and a cylindrical shape, and a frame holding table 58 disposed on the lateral side of the drum 54 and having a circular ring shape.

[0041] The upper surface of the frame holding table 58 is a substantially flat placement surface 58a on which the annular frame 19 of the workpiece unit 21 is placed. As shown in Fig. As shown in Fig. 6A, when the workpiece unit 21 is placed on the belt expanding device 52, the height positions of the upper surface 54a of the drum 54 and the placement surface 58a of the frame holding table 58 have been adjusted to be substantially the same. On the outer peripheral side surface of the frame holding table 58, a plurality of clamp units 60 are fixed at different positions in the circumferential direction. Moreover, the upper end portions of a plurality of piston rods 62, which can move along the height direction of the drum 54, are fixed to the bottom surface of the frame holding table 58. A part of the lower end portion side of the piston rod 62 is disposed in an air cylinder 64. When the piston rod 62 is pulled into each air cylinder 64, the upper surface of the frame holding table 58 is lowered relative to the upper surface 54a of the drum 54.

[0042] Next, the dividing step (S50) using the band expanding device 52 will be described. Fig. 6B is a partial sectional side view of the workpiece 11, etc., in the dividing step (S50). In the dividing step (S50), first, the workpiece unit 21 is placed on the upper surface 54a of the drum 54 and the placement surface 58a of the frame holding table 58, which is set at substantially the same height position. At this time, the part corresponding to the workpiece 11 in the workpiece unit 21 is placed on the upper surface 54a, and the part corresponding to the annular frame 19 in the workpiece unit 21 is placed on the placement surface 58a. Then, the position of the annular frame 19 is fixed by the clamp units 60.

[0043] Thereafter, the air cylinders 64 are actuated and the frame holding table 58 is lowered relative to the upper surface 54. As a result, as in Fig. 6B, the dividing band (expanding film) 17 is expanded in a radial manner from the center of the circular dividing band 17, and the workpiece 11 is divided along each planned dividing line 15. Specifically, in the dividing step (S50), cracks extend between the bottom parts of the cutting grooves 15a and the front surface 11a, with the bottom parts of the cutting grooves 15a having a V-shape or a U-shape as the origin points, and the uncut portions of the substrate 11c and the circuit layer 11d directly under the uncut portions are fractured. As a result, a plurality of chips 23 are formed from the workpiece 11. The width of the cracks formed in the front surface 11a at the time of fracture is in a range of less than 1 μm. For this reason, the notch width of the front surface 11a can be made smaller than in the case of cutting the front surface 11a by the cutting blade 48a.Therefore, in the manufacturing process of the chips 23 in the present embodiment, the width of the planned dividing lines 15 can be set smaller than in the case of cutting the front surface 11a by the cutting blade 48a.

[0044] Fig. 7 is a sectional view of several chips 23. In Fig. Fig. 7 is a sectional view of a plurality of chips 23 when the cutting grooves 15a with the second depth are formed in the workpiece 11 (see Fig. 5B), and the division band 17 on the front surface 11a side of the workpiece 11 is omitted. After the division step (S50), the front surface 11a side of each chip 23 is irradiated with ultraviolet light, and the adhesive force of the adhesive layer of the division band 17 is reduced. Thereafter, the rear surface 11b side of the chip 23 is sucked by a suction head of an inversion pickup device (not shown), and the chip 23 is picked up.

[0045] Next, the suction head is reversed so that the front surface 11a side of the chip 23 faces downward. At this time, the chip 23 assumes the state where the lower side is supported by the suction head of the inversion support device. Next, for example, the front surface 11a side of the chip 23 is sucked onto a suction head of a bonding device (not shown). This transfers the chip 23 from the inversion support device to the bonding device. In the case of conveying the chip 23, the inversion support device does not necessarily have to be used. For example, after the dividing step (S50), an adhesive tape (not shown) for transferring the chip 23 is adhered to the rear surface 11b side of each chip 23, and the front surface 11a side is irradiated with ultraviolet light to remove the dividing tape 17 from the workpiece unit 21.As a result, each chip 23 is transferred to the adhesive tape. In this case, for example, the suction head of the bonding device (not shown) can suction the front surface 11a side of the chip 23 without using the inversion pickup device to pick up and convey the chip 23.

[0046] Fig. 8 is a flowchart of the manufacturing method of the chips 23 in the first embodiment. The workpiece 11 is divided without cutting the front surface 11a of the workpiece in the cutting step (S40) in the first embodiment. After that, the notch width of the front surface 11a side can be made smaller than in the case of cutting the workpiece 11 from the front surface 11a to the back surface 11b to break the workpiece 11. For this reason, the width of the planned parting lines 15 can be made small. Meanwhile, in the above-described parting step (S50), instead of the tape expanding device 52 that expands the parting tape 17 in a radial manner, another type of tape expanding device (not shown) that pulls the parting tape (expanding film) 17 outward from four sides may be used.

[0047] Next, using Fig. 9, Fig. 10A and Fig. 10B a second embodiment will be described. Fig. 9 is a flowchart of a manufacturing method of the chip 23 in the second embodiment. In the second embodiment, first, a front surface protective tape 17a as a front surface protective member is adhered to the front surface 11a side of the workpiece 11 (front surface protecting step (S15)). Instead of the front surface protective tape 17a, a substrate formed of a component having rigidity may be used as the front surface protective member. Thus, a stacked-layer body 25 configured from the front surface protective tape 17a and the workpiece 11 is formed. Fig. 10A is a perspective view of the stacked laminated body 25 with the workpiece 11. The annular frame 19 is not adhered to the front surface protective tape 17a in the present embodiment. However, if the front surface protective tape 17a has a diameter larger than the diameter of the workpiece 11, the annular frame 19 may be adhered to the outer peripheral part of the front surface protective tape 17a.

[0048] The front surface protective tape 17a is a sheet made of a resin and can be obtained, for example, by laminating a base layer made of the ultraviolet-transmitting resin and an ultraviolet-curable adhesive layer. The front surface protective tape 17a in the second embodiment has such extensibility that it can be easily deformed in response to a tensile stress. However, the front surface protective tape 17a does not necessarily have to have extensibility. After the front surface protecting step (S15), the holding step (S20), the cutting position detecting step (S30), and the cutting step (S40) are performed similarly to the first embodiment. In the cutting step (S40), the cutting grooves 15a, which do not reach the front surface 11a side, are formed on the rear surface 11b side.After the front surface protection step (S15), processing for grinding the back surface or the like may be performed on the workpiece 11.

[0049] Then, after the cutting step (S40), an adhesion step (S45) is performed. In the adhesion step (S45), first, the stacked laminated body 25 is placed in the opening part of the annular frame 19, and an expansion film 17b is adhered to the rear surface 11b side of the workpiece 11 and a surface of the annular frame 19. The expansion film 17b is a layer obtained by laminating a base layer and an adhesive layer. The expansion film 17b is made of a plastic and has expandability. Moreover, the adhesive layer for the expansion film 17b is, for example, an ultraviolet-curing plastic. Next, in the adhesion step (S45), the front surface 11a side of the workpiece 11 is irradiated with ultraviolet light. The ultraviolet light is transmitted through the base layer of the front surface protective tape 17a and reaches the adhesive layer.This reduces the adhesive force of the protective tape 17a of the front surface.

[0050] Thereafter, the front surface protective tape 17a is removed by using a removing device (not shown). Fig. 10B is a perspective view illustrating how the front surface protective tape 17a is removed from the workpiece 11. By removing the front surface protective tape 17a, a workpiece unit configured from the workpiece 11, the expanding film 17b, and the annular frame 19 is formed. After the adhesion step (S45), the workpiece 11 is divided into a plurality of chips 23 along the planned division lines 15 using the above-described tape expanding device 52 (division step (S50)). Thereafter, the rear surface 11b side of the chip 23 is irradiated with ultraviolet light, and the adhesive force of the adhesive layer of the expanding film 17b is reduced.

[0051] In the second embodiment, without using the inversion pickup device (not shown), the adhesive tape for transfer (not shown), etc. as in the first embodiment, a suction head of a connecting device (not shown) or the like can suck the front surface 11a side of each chip 23 and pick up each chip 23 to convey each chip 23. Of course, in the second embodiment, the workpiece 11 is also divided without cutting the front surface 11a of the workpiece 11. Therefore, the notch width of the front surface 11a side can be made smaller than in the case of cutting the workpiece 11 from the front surface 11a to the back surface 11b to break the workpiece 11. For this reason, the width of the planned parting lines 15 can be made small.

[0052] Next, a third embodiment will be described. In the third embodiment, a workpiece 31, on which a metal layer 27 is formed on the rear surface 31b side, is divided along the planned dividing lines 15 to form a plurality of chips 23. Fig. Fig. 11 is a perspective view of a workpiece unit 41 according to the third embodiment. The workpiece 31 is a wafer having a substantially circular disk shape and a front surface 31a and a back surface 31b. The workpiece 31 has a substrate 31c formed of silicon carbide (SiC) (see Fig. 15). On the side of the front surface 31a of the workpiece 31, the above-described circuit layer 11d (see Fig. 15) in contact with the substrate 31c.

[0053] The front surface 31a side of the workpiece 31 is divided into a plurality of regions by a plurality of planned dividing lines (streets) 15 arranged in a lattice shape to intersect each other. A device 13 such as a power semiconductor device exists in each of the plurality of regions, and each device 13 is composed of the substrate 31c and the circuit layer 11d. The metal layer 27 is disposed on one surface of the substrate 31c disposed on the rear surface 31b side of the workpiece 31. The metal layer 27 is, for example, a stacked-layer body obtained by laminating a titanium (Ti) layer in contact with the substrate 31c and a nickel (Ni) layer disposed on the opposite side of the substrate 31c on another side of the titanium layer.

[0054] However, the material constituting the stacked laminate is not limited to the titanium layer and the nickel layer. Furthermore, the metal layer 27 may be a layer of a single metal material or may be a stacked laminate obtained by laminating metal layers of three or more different materials. The front surface 31a side of the workpiece 31 is fixed to the opening part of the annular frame 19 by the dividing band (expansion film) 17. The dividing band 17 is a sheet obtained by laminating a base layer and an adhesive layer. The dividing band 17 is made of a plastic and has expandability.Since the metal layer 27 is formed on the entire rear surface 31b side of the workpiece 31, it is impossible to image the intended parting line 15 on the front surface 31a side when imaging the rear surface 31b side with an infrared camera. Thus, in the third embodiment, a cutting device 72 capable of imaging the front surface 31a side from the lower side through the parting line 17 is used.

[0055] Fig. 12 is a perspective view of the cutting device 72 in the third embodiment. The cutting device 72 includes a base 74 supporting the respective components. An X-axis-Y-axis moving mechanism (machine feed and index feed mechanism) 78 is disposed on the upper surface of the base 74. The X-axis-Y-axis moving mechanism 78 includes a pair of X-axis guide rails 80 substantially parallel to the X-axis direction. The X-axis guide rails 80 are fixed to the upper surface of the base 74.

[0056] An X-axis motion table 82 is slidably mounted on the X-axis guide rails 80. A nut member (not shown) is disposed on the lower surface side of the X-axis motion table 82, and an X-axis ball screw 84 substantially parallel to the X-axis guide rails 80 is rotatably connected to this nut member. An X-axis pulse motor 86 is coupled to one end portion of the X-axis ball screw 84. When the X-axis ball screw 84 is rotated by the X-axis pulse motor 86, the X-axis motion table 82 moves along the X-axis guide rails 80 in the X-axis direction. An X-axis scale 80a, which is used when the position of the X-axis moving table 82 in the X-axis direction is detected, is arranged at a position adjacent to the X-axis guide rail 80.

[0057] A pair of Y-axis guide rails 90 substantially parallel to the Y-axis direction are fixed to the upper surface of the X-axis motion table 82. A Y-axis motion table 92 is slidably mounted on the Y-axis guide rails 90. Fig. 13 is a perspective view of the Y-axis moving table 92, etc. Fig. 14A is a sectional view of a portion of the cutting device 72. Fig. 14B is a view obtained by enlarging an area A in Fig. 14A obtained sectional view. In Fig. 14A and Fig. 14B, the workpiece 31 machined by the cutting device 72, etc. are shown together. In addition, hatching of sections in Fig. 14A has been waived. As in Fig. 12, Fig. 13 and Fig. 14A, the X-axis moving table 92 has a bottom plate part 92a whose shape is a rectangle when viewed in the Z-axis direction.

[0058] The lower end of a side plate part 92b, whose shape in the Y-axis direction is a rectangle when viewed in the Y-axis direction, is connected to one end of the bottom plate part 92a. The upper end of the side plate part 92b is connected to one end of an upper plate part 92c in the Y-axis direction, whose shape in the Z-axis direction is a rectangle similar to that of the bottom plate part 92a. This means that a space 92d, which leads to the outside at the other end in the Y-axis direction and both ends in the X-axis direction, is formed between the bottom plate part 92a and the upper plate part 92c. Nut parts 92e ( Fig. 14A) are arranged on the lower surface side of the base plate portion 92a of the Y-axis motion table 92, and a Y-axis ball screw 94 substantially parallel to the Y-axis guide rails 90 is rotatably connected to the nut portions 92e. A Y-axis pulse motor 96 is coupled to one end portion of the Y-axis ball screw 94.

[0059] When the Y-axis ball screw 94 is rotated by the Y-axis pulse motor 96, the Y-axis motion table 92 moves along the Y-axis guide rails 90 in the Y-axis direction. A Y-axis scale 90a (see Fig. 12), which is used when the position of the Y-axis moving table 92 in the Y-axis direction is detected, is arranged at a position adjacent to the Y-axis guide rail 90. A clamping table (holding table) 98, which is used when the plate-shaped workpiece 31 is held (see Fig. 14A and Fig. 14B) is disposed on the side of the upper surface of the upper plate portion 92c of the Y-axis moving table 92. The chuck table 98 is supported by the upper plate portion 92c in such a manner as to be rotatable about a rotation axis substantially parallel to the Z-axis direction.

[0060] The chuck table 98 includes a frame body 100 formed, for example, from a metal material such as stainless steel and having a cylindrical shape. A holding member 102 having a circular disc shape is disposed at the upper portion of the frame body 100 to close an opening portion 100a on the side of the upper portion of the frame body 100. The holding member 102 has a substantially flat front surface (holding surface) 102a and a rear surface 102b on the side opposite the front surface 102a, and is constructed of a transparent material such as sodium glass, borosilicate glass, or quartz glass through which visible light is transmitted. As shown in Fig. As shown in Fig. 14B, a suction path 100c used for sucking the workpiece 31 is formed within the holding member 102. Furthermore, a plurality of openings 102d corresponding to one end of the suction path 102 are formed in the front surface 102a of the holding member 102. The plurality of openings 102d are arranged at substantially equal intervals in the front surface 102a. A suction source 104 including an ejector or the like is connected to the other side of the suction path 102.

[0061] The holding member 102 is transparent to visible light from the front surface 102a to the rear surface 102b at least in a portion excluding the suction path 102c and the opening 102d. Therefore, the workpiece 31 or the like arranged on the front surface 102a side of the holding member 102 can be imaged from the rear surface 102b side of the holding member 102 by a visible-light camera 150. In the present embodiment, the holding member 102 is shown entirely formed of a visible-light transparent material. However, it is sufficient if at least a portion of the holding member 102 is transparent from the front surface 102a to the rear surface 102b.

[0062] A rotary drive source 108, such as a motor, is arranged on the side plate portion 92b of the Y-axis moving table 92. A belt 110 for transmitting power to the rotary drive source 108 is stretched on a pulley portion 100b arranged on the outer periphery of the frame body 100 and a pulley 108a coupled to a rotation shaft of the rotary drive source 108. Therefore, the chuck table 98 can rotate about a rotation axis substantially parallel to the Z-axis direction by power transmitted from the rotary drive source 108 via the belt 110. On the outer periphery of the frame body 100, a plurality of clamps 100c, which are used when fastening the annular frame 19, are arranged next to the pulley portion 100b (see Fig. 14A). The plurality of clamps 100c are fixed to the frame body 100 in such a manner that they do not interfere with rotation of the chuck table 98. Furthermore, the chuck table 98, together with the X-axis moving table 82 and the Y-axis moving table 92, can move in the X-axis direction and the Y-axis direction by the X-axis-Y-axis moving mechanism 78 described above.

[0063] As in Fig. As shown in Fig. 12, a support structure 112 having a columnar shape or a wall shape is arranged in an area that does not overlap with the X-axis-Y-axis moving mechanism 78 on the upper surface of the base 74. A Z-axis moving mechanism 114 is arranged on a side surface of the support structure 112. The Z-axis moving mechanism 114 includes a pair of Z-axis guide rails 116 fixed to the side surface of the support structure 112 and substantially parallel to the Z-axis direction. A spindle housing 120, which forms a cutting unit 118, is slidably mounted on the Z-axis guide rails 116. A nut portion (not shown) is arranged on the side surface of the spindle housing 120 on the side of the support structure 112, and a Z-axis ball screw 122 substantially parallel to the Z-axis guide rails 116 is rotatably connected to this nut portion.

[0064] A Z-axis pulse motor 124 is coupled to one end portion of the Z-axis ball screw 122. When the Z-axis ball screw 122 is rotated by the Z-axis pulse motor 124, the spindle housing 120 moves along the Z-axis guide rails 116 in the Z-axis direction. A Z-axis scale (not shown), which is used when detecting the position of the spindle housing 120 in the Z-axis direction, is arranged at a position adjacent to the Z-axis guide rail 116. The cutting unit 118 includes a spindle (not shown) serving as a rotation axis parallel to the Y-axis direction. The spindle is rotatably supported in the above-described spindle housing 120. One end portion of the spindle is exposed from the spindle housing 120. A cutting blade 118a having a circular ring shape is attached to this end part of the spindle.Meanwhile, a rotary drive source (not shown) such as a motor is coupled to the other end side of the spindle.

[0065] An upper imaging unit 128 for imaging the workpiece 31 or the like held by the chuck table 98 from the upper side is fixed to the spindle housing 120 of the cutting unit 118. Thus, the upper imaging unit 128 is fixed to the spindle housing 120 and moves in the Z-axis direction together with the cutting unit 118 by the Z-axis moving mechanism 114. An imaging unit support structure 132 having a columnar shape or a wall shape is arranged in a region separated in the Y-axis direction from the X-axis-Y-axis moving mechanism 78 and located below the spindle housing 120 in the upper surface of the base 74. An imaging unit moving mechanism 134 is arranged on a side surface of the imaging unit support structure 132.

[0066] The imaging unit movement mechanism 134 includes a pair of Z-axis guide rails 136 fixed to the side surface of the imaging unit support structure 132 and substantially parallel to the Z-axis direction. A Z-axis movement plate 138 is slidably mounted on the Z-axis guide rails 136. A nut member (not shown) is disposed on the side surface of the Z-axis movement plate 138 on the side of the imaging unit support structure 132, and a Z-axis ball screw 140 substantially parallel to the Z-axis guide rails 136 is rotatably connected to this nut member (see Fig. 14A).

[0067] A Z-axis pulse motor 142 is coupled to one end portion of the Z-axis ball screw 140. When the Z-axis ball screw 140 is rotated by the Z-axis pulse motor 142, the Z-axis motion plate 138 moves along the Z-axis guide rails 136 in the Z-axis direction. A Z-axis scale (not shown), which is used when detecting the position of the Z-axis motion plate 138 in the Z-axis direction, is arranged at a position adjacent to the Z-axis guide rail 136. A lower imaging unit 146 is attached to the Z-axis motion plate 138 via a support arm 144 extending in the Y-axis direction.The lower imaging unit 146 includes an illumination device 148 that irradiates a subject (workpiece 31 in the present embodiment) with visible light, and a visible light camera 150 that includes an imaging element that receives light reflected from the object and forms an image.

[0068] A control unit (not shown) is connected to components such as the X-axis-Y-axis moving mechanism 78, the rotary drive source 108, the Z-axis moving mechanism 114, the cutting unit 118, the upper imaging unit 128, the imaging unit moving mechanism 134, and the lower imaging unit 146. The control unit controls the operation of each component so that the workpiece 31 can be properly processed. The control unit is configured by a computer including a processing device such as a CPU and a storage device such as a flash memory. The processing device is made to operate according to software such as a program stored in the storage device, and thus the control unit acts as a specific means by which the software and the processing device (hardware resources) cooperate.

[0069] Next, the machining process of the workpiece 31 according to a third embodiment will be described. In the third embodiment, the workpiece 31 is machined in the same manner as in the Fig. 8 shown flowchart of the first embodiment. First, the Fig. 11 is formed (adhering step (S10)). After the adhering step (S10), the holding step (S20) is performed similarly to the first embodiment. In the holding step (S20), first, the workpiece unit 41 is placed on the chuck table 98 in such a manner that the dividing band 17 comes into contact with the front surface 102a of the holding member 102. Then, the annular frame 19 is fixed by the clamps 100c. In addition, a negative pressure of the suction source 104 is caused to act from the plurality of openings 102d associated with one end of the suction path 102c. Thereby, the workpiece 31 is held by the chuck table 98 in the state in which the rear surface 11b side is exposed upward.

[0070] After the holding step (S20), the cutting position detection step (S30) is performed. In the cutting position detection step (S30), the operation of the X-axis-Y-axis moving mechanism 78 and the imaging unit moving mechanism 134 is controlled, and the lower imaging unit 146 is placed below the transparent area of ​​the holding member 102. That is, the lower imaging unit 146 is inserted into the space 92d between the bottom plate part 92a and the upper plate part 92c of the Y-axis moving table 92. The position of the lower imaging unit 146 relative to the holding member 102 is selectably set within a range suitable for imaging the workpiece 31.

[0071] As described above, a part of the holding member 102 is transparent. Therefore, the front surface 31a side of the workpiece 31 can be imaged when visible light is emitted from the illuminator 148 of the lower imaging unit 146 toward the workpiece 31 on the upper side, and light reflected from the front surface 31a side of the workpiece 31 by the holding member 102 and the dividing band 17 is received by the visible light imaging element of the camera 150. The image obtained by this imaging is stored in the storage device of the control unit. This image includes an alignment mark such as a key pattern existing on the front surface 31a side of the workpiece 31. The coordinates of at least one planned dividing line 15 are detected based on the coordinates of the alignment mark.In this way, the image obtained by the visible light camera 150 is used for alignment and so on of the workpiece 31.

[0072] After the cutting position detection step (S30), the cutting step (S40) is executed. In the cutting step (S40), an operation of the X-axis-Y-axis moving mechanism 78 is controlled, and the position of the cutting blade 118a is aligned with the upper side of an extension line of the planned parting line 15, which is the target of machining. Then, an operation of the Z-axis moving mechanism 114 is controlled, and the position of the cutting unit 118 in the Z-axis direction is adjusted in such a way that the height of the lower end of the cutting blade 118a is adjusted between the front surface 31a and the rear surface 31b of the workpiece 31.

[0073] Next, the chuck table 98 is moved in the X-axis direction by the X-axis-Y-axis moving mechanism 78 while rotating the cutting blade 118a. This causes the cutting blade 118a to cut from the rear surface 31b side of the workpiece 31 along the target design parting line 15, and the workpiece 31 is machined along the design parting line 15. After the cutting groove 15a is formed along one design parting line 15, the workpiece 31 is indexed by a predetermined index amount using the X-axis-Y-axis moving mechanism 78. Then, the cutting groove 15a is similarly formed along another design parting line 15 adjacent to the one cut design parting line 15 in the X-axis direction. Such a process is repeated until the workpiece 31 is machined along all planned parting lines 15.Thereby, the cutting grooves 15a, which do not reach the front surface 31a side, are formed on the rear surface 31b side along all the planned parting lines 15.

[0074] Fig. 15 is a sectional view of a portion of the workpiece 31 having the cutting grooves 15a with a third depth. In Fig. 15 shows the case where the cutting grooves 15a have a substantially U-shape, and the dividing band 17 on the side of the front surface 31a of the workpiece 31 is omitted. The bottom parts of the cutting grooves 15a with the Fig. 15 do not reach the circuit layer 11d and are separated from the boundary 11e between the substrate 31c and the circuit layer 11d by a predetermined distance. The bottom portions of the cutting grooves 15a with the third depth are further separated from the boundary 11e than the bottom portions of the cutting grooves 15a with the second depth described above. The bottom portions of the cutting grooves 15a with the third depth are separated from the boundary 11e by the predetermined distance of several micrometers up to and including 30 µm, preferably by the predetermined distance that is greater than 10 µm and equal to or shorter than 30 µm.

[0075] By separating the bottom portions of the cutting grooves 15a from the boundary 11e by the predetermined distance to form uncut regions in the substrate 31c, the possibility of the circuit layer 11d being cut by the cutting blade 118a can be further reduced compared to the case where the bottom portions of the cutting grooves 15a are arranged at the boundary 11e. Therefore, the peeling phenomenon can be more reliably prevented. Furthermore, by forming the uncut regions, the mechanical strength of the workpiece 31 can be improved at the planned parting line 15 compared to the case where the bottom portions of the cutting grooves 15a are arranged at the boundary 11e. Therefore, it is possible to reduce the possibility of a crack occurring at an unexpected position or in an unexpected direction in the substrate 31c during processing, conveyance, etc. of the workpiece unit 41.

[0076] After the cutting step (S40), the negative pressure of the suction source 104 is shut off, and the clamps 100c are released from securing the annular frame 19. Then, the workpiece unit 41 is transported out from the chuck table 98 to the above-described tape expanding device 52. Thereafter, the dividing step (S50) is performed. In the dividing step (S50), the workpiece 31 is divided into a plurality of chips 23 by expanding the dividing tape (expanding film) 17 using the above-described tape expanding device 52. The substrate 31c in the third embodiment is formed of SiC, which has a higher hardness than silicon. However, the present inventor has confirmed that there is no obstacle to dividing the workpiece 31 in the dividing step (S50), even if the amount of an uncut portion is set larger than in the first embodiment.

[0077] After the dividing step (S50), the front surface 31a side of each chip 23 is irradiated with ultraviolet light, and the adhesive force of the adhesive layer of the dividing tape 17 is reduced. Then, a suction head of an inversion pickup device (not shown) suctions the rear surface 31b side of the chip 23 and transfers the front surface 31a side of the chip 23 to a suction head of a bonding device (not shown).

[0078] Instead, after the dividing step (S50), an adhesive tape (not shown) for transferring the chip 23 may be adhered to the rear surface 31b side of each chip 23, and the front surface 31a side may be irradiated with ultraviolet light to remove the dividing tape 17. This transfers each chip 23 to the adhesive tape. In this case, the suction head of the connecting device (not shown) may suction the front surface 31a side of the chip 23 to pick up and convey the chip 23. Also, in the cutting step (S40) in the third embodiment, the workpiece 31 is divided without cutting the front surface 31a of the workpiece 31. Therefore, the notch width of the front surface 31a side can be made smaller than in the case of cutting the workpiece 31 from the front surface 31a to the rear surface 31b to break the workpiece 31.For this reason, the width of the planned dividing lines 15 can be made small.

[0079] Next, a fourth embodiment will be described. In the fourth embodiment, the workpiece 31 with the metal layer 27 is processed in the same manner as in the flowchart of the Fig. 9. First, the front surface protective tape 17a is adhered to the front surface 31a side of the workpiece 31 (front surface protecting step (S15)). Thus, a stacked laminated body 35 configured from the front surface protective tape 17a and the workpiece 31 is formed. Fig. 16A is a perspective view of the stacked laminated body 35 with the workpiece 31. The annular frame 19 is not adhered to the front surface protective tape 17a in the present embodiment. However, if the front surface protective tape 17a has a diameter larger than the diameter of the workpiece 31, the annular frame 19 may be adhered to the outer peripheral part of the front surface protective tape 17a.

[0080] After the front surface protection step (S15), the holding step (S20), the cutting position detection step (S30), and the cutting step (S40) are performed. In the cutting step (S40), the cutting grooves 15a, which do not reach the rear surface 31b side, are formed on the rear surface 31b side. Then, after the cutting step (S40), the adhesion step (S45) is performed. In the adhesion step (S45), the stacked laminated body 35 is arranged in the opening part of the annular frame 19, and the expansion film 17b is adhered to the rear surface 31b side of the workpiece 31 and a surface of the annular frame 19.

[0081] In addition, after the front surface side 31a of the workpiece 31 is irradiated with ultraviolet light, the front surface protective tape 17a is removed using a removing device (not shown). Fig.16B is a perspective view illustrating how the front surface protective tape 17a is removed from the workpiece 31. By removing the front surface protective tape 17a, a workpiece unit composed of the workpiece 31, the expanding film 17b, and the annular frame 19 is formed. After the adhesion step (S45), the workpiece 31 is divided into a plurality of chips 23 along the planned division lines 15 using the above-described tape expanding device 52 (division step (S50)). Thereafter, the rear surface 31b side of each chip 23 is irradiated with ultraviolet light, and the adhesive force of the adhesive layer of the expanding film 17b is reduced.

[0082] In the fourth embodiment, without using the inversion pickup device (not shown) of the adhesive tape (not shown) for transfer, etc., as in the third embodiment, suction can be effected on the front surface 31a side of each chip 23, and each chip 23 can be picked up by a pickup device (not shown) so that each chip 23 can be conveyed. Of course, in the cutting step (S40) in the fourth embodiment, the workpiece 31 is divided without cutting the front surface 31a of the workpiece 31. Therefore, the notch width of the front surface 31a side can be made smaller than in the case of cutting the workpiece 31 from the front surface 31a to the back surface 31b to break the workpiece 31. For this reason, the width of the planned parting lines 15 can be made small.

[0083] In addition, structures, methods, etc., according to the above-described embodiments can be implemented with appropriate modifications without departing from the scope and purpose of the present invention. For example, in each dividing step (S50) in the second embodiment, the third embodiment, and the fourth embodiment, a different type of band expanding device (not shown) that pulls the dividing band 17 outward from four sides may also be used.

[0084] The present invention is not limited to the details of the preferred embodiments described above. The scope of the invention is defined by the appended claims, and all changes and modifications that fall within the equivalent scope of the claims are therefore encompassed by the invention.

Claims

[1] A manufacturing method for chips (23), wherein a workpiece (11) having a plurality of planned dividing lines (15) on a front surface (11a) is divided along each planned dividing line (15) to manufacture the chips (23) from the workpiece (11), the manufacturing method for chips (23) comprising: a cutting step of causing a cutting blade (48a) to cut into the workpiece (11), for which one side of the front surface (11a) of the workpiece (11) is held by a holding table (26, 98) in such a manner that one side of a rear surface (11b) of the workpiece (11) is exposed, and forming a cutting groove, which does not reach the front surface (11a) of the workpiece (11), on the side of the rear surface (11b) of the workpiece (11) along each planned parting line (15); an adhesion step of adhering an expansion film (17) to the workpiece (11); and a dividing step of dividing the workpiece (11) along each planned dividing line (15) by expanding the expanding film (17) after the adhering step and the cutting step to form the chips (23) from the workpiece (11), wherein the workpiece (11) comprises a silicon carbide substrate and a metal layer (27) arranged on a surface of the silicon carbide substrate and arranged on the side of the rear surface (11b) of the workpiece (11), and the cutting blade (48a) is caused to cut into the workpiece (11) from the side of the rear surface (11b) of the workpiece (11) to form the cutting grooves in the cutting step. [2] A manufacturing method for chips (23) according to claim 1, further comprising: a cutting position detection step of imaging the front surface (11a) of the workpiece (11) by an infrared camera from the side of the rear surface (11b) of the workpiece (11) and detecting at least one planned parting line (15) before the cutting step. [3] A manufacturing method for chips (23) according to claim 1 or 2, wherein the holding table (26, 98) comprises a holding member (102) that is at least partially transparent from a front surface (11a) of the holding member (102) to a rear surface (11b) of the holding member (102), the manufacturing method for chips (23) further comprising: a cutting position detection step of imaging the front surface (11a) of the workpiece (11) by a visible light camera (150) through the holding member (102) and detecting at least one planned parting line (15) before the cutting step. [4] Manufacturing method for chips (23) according to one of the preceding claims, wherein an adhesion step is carried out before the cutting step, the expansion film (17) is adhered to the side of the front surface (11a) of the workpiece (11) in the adhesion step, and the cutting grooves are formed in the workpiece (11) in the cutting step in a state in which the side of the front surface (11a) of the workpiece (11) is held by the holding table (26, 98) with the interposition of the expanding film (17). [5] Manufacturing method for chips (23) according to one of claims 1 to 3, wherein the adhesion step is carried out after the cutting step, and the expansion film (17) is adhered to the side of the rear surface (11b) of the workpiece (11) in the adhesion step. [6] Manufacturing method for chips (23) according to one of the preceding claims, wherein the cutting blade (48a) has a pair of inclined surfaces (48c) or a curved surface (48d) on an outer peripheral part (48b), and the workpiece (11) is cut by the cutting blade (48a) in the cutting step. [7] Manufacturing method for chips (23) according to one of the preceding claims, wherein the workpiece (11) has a circuit layer (11d) with insulation layers and connection layers alternately layered on the side of the front surface (11a), and the cutting grooves which do not reach the circuit layer (11d) are formed in the workpiece (11) in the cutting step.

Images