Wafer processing methods

Abstract

Wafer processing method for dividing a wafer (2) into a plurality of individual components (22) along a plurality of parting lines (23) formed on a front face of the wafer (2), wherein the wafer (2) is composed of a substrate (20) and a functional layer (21) formed on a front face (20a) of the substrate (20), and individual components (22) are formed by the functional layer (21) and are divided by the parting lines (23), wherein the wafer processing method comprises: a protective tape application step for applying a protective tape (3) having an adhesive layer (31) to a front face (21a) of the functional layer (21) of the wafer (2) in a state in which the adhesive layer (31) of the protective tape (3) is in contact with the front face (21a) of the functional layer (21); and a wafer splitting step to hold the wafer (2) processed by the protective tape application step on a holding surface of a clamping table (51) in a state in which the protective tape (3) is in contact with the holding surface, and then application of a laser beam (LB) having an absorption wavelength for the substrate (20) and the functional layer (21) of the wafer (2) from a rear side (20b) of the substrate (20) along each parting line (23) to form a laser-machined groove (24) with a depth that reaches the protective tape (3) along each parting line (23), thereby dividing the wafer (2) into individual component chips corresponding to the individual components (22), wherein the protective tape (3) is applied in the protective tape application step in such a way as to be close to the front face (21a) of the functional layer (21) that the adhesive layer (31) of the protective tape (3) comes into close contact with the components (22) formed by the functional layer (21) in order to prevent the adhesion of residual particles to a front face of each component (22), wherein the residual particles are generated from the wafer (2) by the application of the laser beam (LB) in the wafer splitting step along each dividing line (23), wherein, after the protective tape (3) is applied to the front (21a) of the functional layer (21) of the semiconductor wafer (2), pressure is exerted on the protective tape (3) in the protective tape application step, thereby bringing the adhesive layer (31) into close contact with the components (22) formed by the functional layer (21), and wherein a pressure roller (4) is rolled on the protective belt (3) to exert pressure on the protective belt (3).

Description

BACKGROUND OF THE INVENTION FIELD OF THE INVENTION

[0001] The present invention relates to a wafer processing method for dividing a wafer into a plurality of individual components along a plurality of separation lines formed on the front side of the wafer, wherein the wafer is constructed from a substrate and a functional layer formed on the front side of the substrate, and the individual components are formed by the functional layer and are divided by the separation lines. DESCRIPTION OF THE RELATED PRIOR ART

[0002] As is well known in the art, in a semiconductor device manufacturing method, a functional layer composed of an insulating film and a functional film is formed on the front surface of a substrate, such as a silicon substrate, and a plurality of devices, such as ICs and LSIs, are formed like a matrix from this functional layer, thereby obtaining a semiconductor wafer having the plurality of devices. The plurality of devices are divided by a plurality of dividing lines formed on the front surface of the semiconductor wafer. The semiconductor wafer is divided along these dividing lines to obtain individual chip devices corresponding to the plurality of devices.

[0003] In recent years, a semiconductor wafer intended for improving the processing performance of semiconductor chips (devices) such as ICs and LSIs has been put into practice. This semiconductor wafer is composed of a substrate such as a silicon substrate and a functional layer formed on the front surface of the substrate. The functional layer is composed of a low-permeability insulating film (low-k film) and a functional film formed on the low-k film, the functional film forming a variety of circuits. Thus, semiconductor devices are formed from the functional layer. The low-k film is formed of an inorganic film such as SiOF, BSG (SiOB), etc., or an organic film such as a polymer film such as polyimide, parylene, etc.

[0004] The division of such a semiconductor wafer along the dividing lines is usually carried out by using a cutting device called a dicing saw. This cutting device includes a chuck table as workpiece holding means for holding the semiconductor wafer as a workpiece, a cutting means for cutting the semiconductor wafer held on the chuck table, and a moving means for relatively moving the chuck table and the cutting means. The cutting means includes a rotary spindle adapted to be rotated at high speed and a cutting blade mounted on the rotary spindle. The cutting blade is formed of a disk-shaped base and an annular cutting edge mounted on a side surface of the base along its outer periphery. The annular cutting edge is a hot-electroformed diamond blade formed, for example, by bonding.Adhesion of abrasive diamond grains with a grain size of approximately 3 µm.

[0005] However, the above-mentioned low-k film differs in material from the substrate of the semiconductor wafer, making it difficult to cut the substrate along with the low-k film using a cutting blade. This means that the low-k film, like mica, is very brittle. Accordingly, when the semiconductor wafer containing the low-k film is cut along the dividing lines using a cutting blade, a problem arises that the low-k film may be separated, and this separation (delamination) may reach the device (circuits), causing devastating damage to the device.

[0006] To solve this problem, Japanese Patent Application Laid-Open No. 2005-64231 discloses a wafer dividing method including the steps of applying a laser beam along both sides of each dividing line on a semiconductor wafer to form two laser-machined grooves along each dividing line and thereby divide the functional layer, and next positioning a cutting blade between the two laser-machined grooves along each dividing line to move the cutting blade and the semiconductor wafer relative to each other and thereby cut the semiconductor wafer along each dividing line.

[0007] JP S61-222703 A discloses a method for dividing semiconductor wafers to form multiple semiconductor elements. A semiconductor wafer provided with electrodes of semiconductor elements on one side is bonded with an adhesive to a side opposite the common electrode. The semiconductor wafer is then cut by a laser beam from the electrode side to divide it into individual semiconductor elements. JP 2014-165246 A also discloses such a method. SUMMARY OF THE INVENTION

[0008] In the above-mentioned wafer processing method disclosed in JP 2005-64231 A, a laser beam is applied along both sides of each dividing line formed on the semiconductor wafer, thereby forming two laser-machined grooves that divide the functional layer along each dividing line. Thereafter, the cutting blade is positioned between the two laser-machined grooves along each dividing line and then actuated to cut the semiconductor wafer along each dividing line. Accordingly, this conventional method has the following problems. (1) At least two laser-machined grooves must be formed along each parting line to divide the functional layer. Consequently, productivity is reduced. (2) If the pitch of the functional layer is insufficient when forming the laser-machined grooves, the cutting blade may be deflected or tilted, or uneven wear of the cutting blade may occur. (3) When the laser beam is applied to the wafer from its front side to form the laser-machined grooves, debris or residual particles will disperse and adhere to the front side of the wafer. Therefore, a protective film must be formed on the front side of the wafer. (4) To form the two laser-machined grooves along each dividing line, the laser beam is applied in two passes along each dividing line. As a result, thermal stress remains in the wafer, causing a reduction in the strength of each component. (5) Since the distance between the laser-machined grooves along each dividing line is larger than the width of the cutting blade, the width of each dividing line must be increased, which causes a reduction in the number of devices that can be formed on the wafer. (6) On some wafers, an activation film made of SiO2, SiO, SiN, SiNO, etc. is formed on the front side of the functional layer. Accordingly, when a laser beam is applied to the wafer from its front side, the laser beam passes through the passivation film to reach the inner side of the functional layer. As a result, the functional layer is processed by the laser beam, and the energy of the laser beam is confined by the passivation film, allowing a so-called undercut, or undercut, to be formed, allowing the laser beam processing to be spread laterally to each device.

[0009] It is therefore an object of the present invention to provide a wafer processing method that can divide a wafer into a plurality of individual components along a plurality of separation lines without the above problems, wherein the wafer includes a substrate and a functional layer formed on the front side of the substrate, and the individual components are formed by the functional layer and divided by the separation lines.

[0010] In accordance with one aspect of the present invention, there is provided a wafer processing method for dividing a wafer into a plurality of individual components along a plurality of dividing lines formed on the front side of the wafer, wherein the wafer is composed of a substrate and a functional layer formed on the front side of the substrate, the individual components are formed from the functional layer and divided by the dividing lines, and the wafer processing method includes a protective tape applying step for applying a protective tape having an adhesive layer to the front side of the functional layer of the wafer in the state where the adhesive layer of the protective tape is in contact with the front side of the functional layer, and a wafer dividing step for holding the wafer processed by the protective tape applying step on a holding surface of a chuck in the statein which the protective tape is in contact with the holding surface, and next, applying a laser beam having an absorption wavelength for the substrate and the functional layer of the wafer from the back of the substrate along each dividing line to form a laser-machined groove with a depth that reaches the protective tape along each dividing line, thereby dividing the wafer into individual component chips corresponding to the individual components; wherein the protective tape is closely attached to the front side of the functional layer in such a way in the protective tape attachment step that the adhesive layer of the protective tape comes into close contact with the components formed by the functional layer to prevent the adhesion of residual particles to the front side of each component,wherein the residual particles are generated from the wafer along each dividing line by applying the laser beam in the wafer dividing step; wherein, after applying the protective tape to the front side of the functional layer of the semiconductor wafer in the protective tape application step, pressure is applied to the protective tape, thereby bringing the adhesive layer into close contact with the components formed by the functional layer, and wherein a pressure roller is rolled on the protective tape to apply the pressure to the protective tape.

[0011] According to the wafer processing method of the present invention, in the protective tape application step, the protective tape is tightly attached to the front surface of the functional layer in such a manner that the adhesive layer of the protective tape comes into close contact with the components formed by the functional layer to prevent the adhesion of residual particles to the front surface of each component. The residual particles are generated from the wafer along each separation line by applying the laser beam in the wafer dividing step. As mentioned above, in the wafer dividing step, residual particles can be generated from the wafer by applying the laser beam having an absorption wavelength suitable for the substrate and the functional layer.However, since the protective tape is applied close to the functional layer in such a way that the adhesive layer of the protective tape is in close contact with the components formed by the functional layer, adhesion of the residual particles to the components can be prevented.

[0012] The wafer processing method according to the present invention can also exhibit the following effects. (1) It is not necessary to form a plurality of laser-machined grooves for dividing the functional layer along each parting line, so that productivity can be improved. (2) No laser-machined groove is formed in the functional layer before cutting the substrate with the cutting blade. Accordingly, the cutting blade cannot be deflected or tilted, and it is impossible for the cutting blade to wear unevenly. (3) Since the laser beam is not applied from the front of the wafer, no protective film is necessary to cover the front of the wafer. (4) Since the cutting groove is formed on the back side of the substrate along each dividing line, it is not necessary to increase the width of each dividing line, so that the number of devices that can be formed on the wafer can be increased. (5) Since the laser beam is not applied from the front side of the wafer, it is not possible for the laser beam to pass through the activation film of SiO2, SiO, SiN, SiNO, etc. to process the functional layer, and the heat generated in the functional layer can be temporarily contained by the activation film to cause the lateral propagation of the processing through each device.

[0013] The above and other objects, features and advantages of the present invention and the manner of carrying them into effect will become more apparent and the invention will be best understood from a study of the following description and appended claims with reference to the accompanying drawings which show some preferred embodiments of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1A is a perspective view of a semiconductor wafer; Fig. 1B is an enlarged sectional view of an essential part of the Fig. 1A shown semiconductor wafer; The Fig. 2A to 2C are perspective views illustrating a protective tape attaching step Fig. 3 is a perspective view of an essential part of a laser processing apparatus for carrying out a first preferred embodiment of a wafer dividing step The Fig. 4A to 4C are views for illustrating the first preferred embodiment of the wafer dividing step Fig. 5 is a perspective view of an essential part of a cutting apparatus for performing a cutting groove forming step in a second preferred embodiment of the wafer dividing step; The Fig. 6A to 6D are views for illustrating the cutting groove forming step; Fig. 7 is a perspective view of an essential part of a laser processing apparatus for performing a laser processing step in the second preferred embodiment of the wafer dividing step; The Fig. 8A to 8D are views for illustrating the laser processing step; The Fig. 9A and Fig. 9B are perspective views illustrating a wafer supporting step; The Fig. 10A and Fig. 10B are perspective views illustrating a protective tape peeling step; Fig. 11 is a perspective view of a belt expanding device for performing a belt expanding step; and The Fig. 12A to 12C are views illustrating the band expansion step. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0014] The wafer processing method in accordance with the present invention will now be described in more detail with reference to the accompanying drawings. Fig. 1A is a perspective view of a semiconductor wafer 2 to be divided into individual components by the wafer processing method in accordance with the invention, and Fig. 1B is an enlarged sectional view of an essential part of the Fig. 1A shown semiconductor wafer 2. As shown in the Fig. 1A and Fig. As shown in FIG. 1B, the semiconductor wafer 2 is composed of a substrate 20, such as a silicon substrate, and a functional layer 21 formed on the front surface 20a of the substrate 20. For example, the substrate 20 has a thickness of 200 μm. The functional layer 21 is composed of an insulating film and a functional film formed on the insulating film, the functional film forming a plurality of circuits. A plurality of devices 22, such as ICs and LSIs, are formed from the functional layer 21 like a matrix. These devices 22 are separated by a plurality of intersecting division lines 23 formed on the functional layer 21. In this preferred embodiment, the insulating film forming the functional layer 21 is provided by an SiO2 film or a low-transmittance insulating film (low-k film).Examples of low-k films include an inorganic film of SiOF, BSG (SiOB), etc., and an organic film such as a polymer film of polyimide, parylene, etc. For example, the thickness of the insulating film is set to 10 µm. Furthermore, a passivation film of SiO2, SiO, SiN, SiNO, etc. is formed on the front side 21a of the functional layer 21.

[0015] The wafer processing method for dividing the semiconductor wafer 2 along the dividing line 23 will now be described. First, a protective tape application step is carried out in such a manner that a protective tape having an adhesive layer is applied to the front surface 21a of the functional layer 21 forming the semiconductor wafer 2 in the state where the adhesive layer of the protective tape is in contact with the front surface 21a of the functional layer 21. More specifically, as shown in FIGS. Fig. 2A and Fig. As shown in Fig. 2B, a protective tape 3 having an adhesive layer 31 is attached to the front surface 21a of the functional layer 21 of the semiconductor wafer 2 in the state where the adhesive layer 31 of the protective tape 3 is in contact with the front surface 21a of the functional layer 21. That is, the protective tape 3 is composed of a base layer and the adhesive layer 31 formed on the front surface (one side) of the base layer. The base layer is formed of polyvinyl chloride (PVC) and has a thickness of 100 µm. The adhesive layer 31 is formed of an acrylic resin and has a thickness of approximately 5 µm.In this protective tape application step, it is important to apply the protective tape 3 close to the front surface 21a of the functional layer 21 in such a manner that the adhesive layer 31 of the protective tape 3 comes into close contact with the devices 22 formed by the functional layer 21, in order to prevent the adhesion of residual particles to the front surface of each device 22, which residual particles can be generated by applying a laser beam along each dividing line 23 in a wafer dividing step to be performed later. After applying the protective tape 3 to the front surface 21a of the functional layer 21 of the semiconductor wafer 2, in this preferred embodiment, as shown in FIGS. Fig. 2A and Fig. 2B, a pressure roller 4 is rolled on the protective tape 3 to, as shown in Fig. 2C, to exert pressure on the protective tape 3, whereby the adhesive layer 31 is brought into close contact with the components 22 formed by the functional layer 21.

[0016] After performing the above-mentioned protective tape attachment step, a first preferred embodiment of a wafer dividing step is carried out in such a manner that the semiconductor wafer 2 with the protective tape 3 is held on a holding surface of a workpiece holding means in the state where the protective tape 3 is in contact with the holding surface, and a laser beam having an absorption wavelength for the substrate 20 and the functional layer 21 is next applied to the semiconductor wafer 2 from the back side 20b of the substrate 20 along each dividing line 23 to form a laser-machined groove with a depth that reaches the protective tape 3 along each dividing line 23, thereby dividing the semiconductor wafer 2 into individual device chips. The first preferred embodiment of this wafer dividing step is carried out using a method described in Fig. 3 shown laser processing device 5. The Fig. The laser processing apparatus 5 shown in Figure 3 includes a chuck table 51 as a workpiece holding means for holding a workpiece, a laser beam application means 52 for applying a laser beam to the workpiece held on the upper surface of the chuck table 51 as a holding surface, and an imaging means 53 for imaging the workpiece held on the chuck table 51. The chuck table 51 is configured to hold the workpiece to the holding surface with suction force. The chuck table 51 is configured both in the direction indicated by an arrow X in Fig. 3 by a feeding means not shown and in the direction indicated by an arrow Y in Fig. 3 shown dividing direction by a dividing means not shown.

[0017] The laser beam application means 52 includes a cylindrical housing 521 extending in a substantially horizontal direction, a laser beam oscillation means (not shown) provided in the housing 521 for oscillating a pulsed laser beam, and a focusing means 522 mounted at the front end of the housing 521 for applying the pulsed laser beam to the workpiece. The imaging means 53 is mounted at a front end portion of the housing 521, which forms the laser beam application means 52.Although not shown, the imaging means 53 includes a conventional imaging device (CCD) for imaging the workpiece using visible light, an infrared light application means for applying infrared light to the workpiece, an optical system for capturing the infrared light applied to the workpiece by the infrared light application means, and an imaging device (infrared CCD) for outputting an electrical signal corresponding to the infrared light captured by the optical system. An image output signal from the imaging means 53 is transmitted to the control means (not shown).

[0018] The first preferred embodiment of the wafer dividing step to be carried out using the laser processing apparatus 5 will now be described with reference to Fig. 3 and the Fig. 4A to 4C. First, the semiconductor wafer 2 processed by the protective tape application step is placed on the chuck table 51 of the Fig. 3 is placed in the state in which the protective tape 3 attached to the semiconductor wafer 2 is in contact with the upper surface (holding surface) of the chuck table 51. Thereafter, a suction means (not shown) is actuated to hold the semiconductor wafer 2 by the protective tape 3 on the chuck table 51 under suction (wafer holding step). Accordingly, the back side 20b of the substrate 20 of the semiconductor wafer 2 held on the chuck table 51 is oriented upward. Although an annular frame supporting the protective tape 3 is not provided in Fig. 3, this annular frame is held by a suitable frame holding means provided on the chuck table 51. Thereafter, the chuck table 51 holding the semiconductor wafer 2 is moved to a position directly below the imaging means 53 by actuating the feeding means not shown.

[0019] In the state where the chuck table 51 is positioned directly below the imaging means 53, an alignment process is performed by the imaging means 53 and the control means (not shown) to detect a target area of ​​the semiconductor wafer 2 to be laser-processed. Specifically, the imaging means 53 and the control means perform image processing such as pattern recognition to align the dividing lines 23 extending in a first direction on the functional layer 21 of the semiconductor wafer 2 and the focusing means 522 of the laser beam application means 52 for applying the laser beam to the wafer 2 along the dividing lines 23, thereby performing the alignment of a laser beam application position (alignment step).Similarly, the alignment of a laser beam application position is performed for the other dividing lines 23 extending in a second direction perpendicular to the first direction on the functional layer 21. Although the front side 21a of the functional layer 21, on which the dividing lines 23 are formed, faces downward, the dividing lines 23 can be imaged from the back side 20B of the substrate 20 because the imaging means 53 includes the infrared light application means for applying infrared light, the optical system for receiving the infrared light, and the imaging device (infrared CCD) for outputting an electrical signal corresponding to the infrared light.

[0020] After performing the above-mentioned alignment step, the chuck table 51 is moved to a laser beam application area where, as shown in Fig. 4A, the focusing means 522 of the laser beam applying means 52 is arranged, whereby one end (in Fig. 4A) of a predetermined one of the dividing lines 23 extending in the first direction is positioned directly under the focusing means 522. Furthermore, the focal point P of the pulsed laser beam to be applied by the focusing means 522 is set near the back side 20B of the substrate 20. Thereafter, the pulsed laser beam having an absorption wavelength for the substrate 20 and the functional layer 21 is applied by the focusing means 522 onto the wafer 2, and the chuck table 51 is moved at a predetermined feed speed in the direction indicated by an arrow X1 in Fig. 4A. When the other end (in Fig. 4B right end) of the predetermined dividing line 23, the position directly below the focusing means 522, as in Fig. 4B, the application of the pulsed laser beam is stopped and the movement of the clamping table 51 is also stopped.

[0021] For example, the above-mentioned wafer dividing step can be carried out under the following processing conditions. Light source: pulsed YAG reader Wavelength: 355 nm Repetition frequency: 200 kHz Average power: 3 W Focus point diameter: 10 µm Working feed speed: 300 mm / s

[0022] After the above-mentioned execution of the wafer dividing step along the predetermined dividing line 23, the chuck table 51 is moved in the dividing direction of the arrow Y by an amount corresponding to the pitch of the dividing lines 23 (dividing step), and the wafer dividing step is similarly performed along the next dividing line 23 extending in the first direction. After the wafer dividing step has been performed along all the dividing lines 23 extending in the first direction, the chuck table 51 is rotated by 90° to similarly execute the wafer dividing step along all the remaining dividing lines 23 extending in the second direction perpendicular to the first direction. As a result, a laser-machined groove 24 with a depth reaching the protective tape 3 is formed in the substrate 20 and the functional layer 21 of the semiconductor wafer 2, as shown in Fig. 4C, along each dividing line 23. Accordingly, the semiconductor wafer 2 is divided by the laser-machined grooves 24, so that the individual device chips corresponding to the devices 22 are obtained. In this wafer dividing step, by applying the pulsed laser beam having an absorption wavelength for the substrate 20 and the functional layer 21, residual particles may be generated from the semiconductor wafer 2. However, since the protective tape 3 is attached close to the functional layer 21 in such a manner that the adhesive layer 31 of the protective tape 3 is in close contact with the devices 22 formed by the functional layer 21, adhesion of the residual particles to the devices 22 can be prevented.The residual particles generated by the application of the pulsed laser beam adhere to the adhesive layer 31 of the protective tape 3 in an area corresponding to each separating line 23.

[0023] A second preferred embodiment of the wafer dividing step will now be described with reference to the Fig. 5 to 8D. In the second preferred embodiment of the wafer dividing step, first, a cutting groove forming step is carried out in such a manner that the substrate 20 of the semiconductor wafer 2 is cut from the back side 20b in an area corresponding to each dividing line 23 using a cutting blade, to thereby form a cutting groove with a depth that does not reach the functional layer 21, so that a part of the substrate 20 remains in this area. This cutting groove forming step is carried out by using a Fig. 5 shown cutting device 6. As shown in Fig. 5, the cutting device 6 includes a clamping table 61 as a workpiece holding means for holding a workpiece, a cutting means 62 for cutting the workpiece held on the clamping table 61, and an imaging means 63 for imaging the workpiece held on the clamping table 61. The clamping table 61 has an upper surface as a holding surface for holding the workpiece thereon under suction. The clamping table 61 is arranged in both the direction indicated by an arrow X in Fig. 5 shown feeding or feed direction by a feeding means not shown as well as in the direction indicated by an arrow Y in Fig. 5 by a dividing means not shown.

[0024] The cutting means 62 includes a spindle housing 621 extending in a substantially horizontal direction, a rotating spindle 622 rotatably supported by the spindle housing 621, and a cutting blade 623 mounted on the rotating spindle 622 at its front end portion. The rotating spindle 622 is adapted to be rotated in the direction indicated by an arrow 623a by a servo motor (not shown) provided in the spindle housing 621. The cutting blade 623 is composed of a disc-shaped base 624 formed of aluminum and an annular cutting edge 625 mounted on a side surface of the base 624 along its outer periphery. The annular cutting edge 625 is an electroformed diamond blade made by adhering abrasive diamond particles having a grain size of 3 to 4 µm with a nickel coating on the side surface of the outer peripheral portion of the base 624.For example, the cutting edge 625 has a thickness of 40 µm and an outer diameter of 52 mm.

[0025] The imaging means 63 is mounted on a front end portion of the spindle housing 621. Although not shown, the imaging means 63 includes a conventional imaging device (CCD) for imaging the workpiece using visible light, an infrared light application means for applying infrared light to the workpiece, an optical system for receiving the infrared light applied to the workpiece by the infrared light application means, and an imaging device (infrared CCD) for outputting an electrical signal corresponding to the infrared light captured by the optical system. An image signal output from the imaging means 63 is transmitted to a control means (not shown).

[0026] When carrying out the cutting groove forming step using the above-mentioned cutting device 6, the semiconductor wafer 2 processed by the protective tape attaching step is placed on the chuck table 61 in the state in which the protective tape 3 on the semiconductor wafer 2 is in contact with the upper surface (holding surface) of the chuck table 61, as shown in Fig. 5. Thereafter, a suction means (not shown) is actuated to hold the semiconductor wafer 2 through the protective tape 3 to the chuck table 61 with a suction force (wafer holding step). Accordingly, the back side 20b of the substrate 20 of the semiconductor wafer 2 held on the chuck table 61 is directed upward. Although in Fig. 5, no annular frame supporting the protective tape 3 is shown, this annular frame is held by a frame holding means provided on the chuck table 61. Thereafter, the chuck table 61 holding the semiconductor wafer 2 is moved to a position directly below the imaging means 63 by actuating the feed means not shown.

[0027] In the state where the chuck table 61 is positioned directly below the imaging means 63, an alignment process is performed by the imaging means 63 and the control means (not shown) to detect a target area of ​​the semiconductor wafer 2 to be cut. Specifically, the imaging means 63 and the control means perform image processing such as pattern recognition to cause the alignment of the cutting blade 623 and the area of ​​the substrate 20 corresponding to each division line 23 extending in a first direction of the semiconductor wafer 2 on the functional layer 21, thereby performing the alignment of the cut area by the cutting blade 623 (alignment step).Similarly, the alignment of another cutting area is performed by the cutting blade 623 for the remaining parting lines 23 extending in a second direction perpendicular to the first direction on the functional layer 21. Although the front side 21a of the functional layer 21, on which the parting lines 23 are formed, faces downward, the parting lines 23 can be imaged from the back side 20b of the substrate 20 because the imaging means 63 includes the infrared light applying means for applying infrared light, the optical system for receiving the infrared light, and the imaging device (infrared CCD) for outputting an electrical signal corresponding to the infrared light.

[0028] After performing the above-mentioned alignment step to detect the area of ​​the substrate 20 corresponding to each dividing line 23 of the semiconductor wafer 2 held on the chuck table 61, the chuck table 61 is moved to a cutting start position in the cutting area by the cutting blade 623, whereby one end (which is shown in Fig. 6A) of a predetermined one of the dividing lines 23 on the right side of a position with a predetermined extent, as shown in Fig. 6A, positioned directly below the cutting blade 623.

[0029] In the state where the semiconductor wafer 2 held on the chuck table 61 is set to the cutting start position in the cutting area as described above, the cutting blade 623 is moved by a line shown by a dotted line in Fig. 6A shown standby position to a position indicated by a Fig. 6A with a continuous line shown working position, as indicated by an arrow Z1 in Fig. 6A. As shown in the Fig. 6A and Fig. 6C, the working position is set so that the lower end of the cutting blade 623 does not reach the functional layer 21 of the semiconductor wafer 2 (for example, the lower end of the cutting blade 623 is set to a vertical position raised from the front surface 20a of the substrate 20 toward the back surface 20b thereof with a distance of 5 to 10 µm).

[0030] Thereafter, the cutting blade 623 is moved in the direction indicated by an arrow 623a in Fig. 6A at a predetermined rotation speed, and the clamping table 61 is rotated in the direction indicated by an arrow X1 in Fig. 6A at a predetermined feed rate. When the other end, ie the one shown in Fig. 6B, the predetermined separation line 23 reaches a position on the left side of the position directly under the cutting blade 623 at a predetermined distance, as shown in Fig. 6D, the movement of the chuck table 61 is stopped. As a result, a cutting groove 25 is formed on the back side 20B of the substrate 20 of the semiconductor wafer 2 along the predetermined separation line 23, so that the depth of the cutting groove 25 does not reach the functional layer 21, that is, a part 201 of the substrate 20 is formed, as shown in Fig. 6D, left standing on the front side 20a (cut groove forming step).

[0031] Thereafter, the cutting blade 623 is raised from the working position to the standby position as indicated by an arrow Z2 in Fig. 6B, and the clamping table 61 is next moved in the direction indicated by an arrow X2 in Fig. 6B to the direction shown in Fig. 6A. Thereafter, the clamping table 61 is moved by an amount in the direction perpendicular to the sheet plane of the Fig. 6A (division direction) corresponding to the pitch of the dividing lines 23, thereby aligning the cutting blade 623 with the area corresponding to the next dividing line 23 extending in the first direction. In the state where the cutting blade 623 is aligned with the area corresponding to the next dividing line 23 as mentioned above, the cut groove forming step is carried out in a similar manner.

[0032] For example, the above-mentioned cutting groove forming step is carried out under the following machining conditions. Outer diameter of the cutting blade: 52 mm Cutting blade thickness: 40 µm Cutting blade rotation speed: 30,000 rpm Working feed speed: 50 mm / s

[0033] In this way, the cutting groove forming step is carried out in a similar manner along all the parting lines 23 extending in the first direction, the chuck table 61 is rotated by 90° to similarly carry out the cutting groove forming step along the remaining parting lines 23 extending in the second direction perpendicular to the first direction.

[0034] After performing the above-mentioned cut groove forming step, a laser processing step is performed in such a manner that the semiconductor wafer 2 with the protective tape 3 is held on a holding surface of the workpiece holding means in the state where the protective tape 3 is in contact with the holding surface, and a laser beam having an absorption wavelength for the substrate 20 and the functional layer 21 is next applied to the semiconductor wafer 2 from the back surface 20b of the substrate 20 along the bottom of each cut groove 25 to form a laser-machined groove with a depth that reaches the protective tape 3 along each dividing line 23, thereby dividing the semiconductor wafer 2 into individual device chips. This laser processing step is performed using a laser processing apparatus 5 shown in Fig. 7 is shown. The Fig. 7 is the same as the laser processing device 5 shown in Fig. 3 shown.

[0035] By performing the laser processing step using the Fig. In the laser processing apparatus 5 shown in Fig. 7, the semiconductor wafer 2 processed by the cut groove forming step is placed on the chuck table 51 in the state where the protective tape 3 attached to the semiconductor wafer 2 is in contact with the upper surface (holding surface) of the chuck table 51. Thereafter, the suction means (not shown) is actuated to hold the semiconductor wafer 2 to the chuck table 51 by the protective tape 3 (wafer holding step). Accordingly, the back side 20b of the substrate 20 of the semiconductor wafer 2 held on the chuck table 51 is directed upward. Although in Fig. 7, no annular frame supporting the protective tape 3 is shown, this annular frame is held by a suitable frame holding means provided on the chuck table 51. Thereafter, the chuck table 51 holding the semiconductor wafer 2 is moved to a position directly below the imaging means 53 by actuating the feed means not shown.

[0036] In the state where the chuck table 51 is positioned directly below the imaging means 53, an alignment process is performed by the imaging means 53 and the control means (not shown) to detect a target area of ​​the semiconductor wafer 2 to be laser processed. Specifically, the imaging means 53 and the control means perform image processing such as pattern recognition to cause the alignment of the cutting groove 25 extending in the first direction on the back surface 20b of the substrate 20 from the semiconductor wafer 2 and the focusing means 522 of the laser beam application means 52 for applying the laser beam to the wafer 2 along the cutting grooves 25, thereby performing the alignment of a laser beam application position (alignment step).In a similar manner, the alignment of a laser beam application position is carried out for the other cutting grooves 25 extending in the second direction perpendicular to the first direction.

[0037] After performing the above-mentioned alignment step, the chuck table 51 is moved to a laser beam application area, where the focusing means 522 of the laser beam application means 52, as shown in Fig. 8A, whereby one end (which is shown in Fig. 8A) of a predetermined one of the cutting grooves 25 is positioned under the focusing means 522. Furthermore, the focus point P of the pulsed laser beam to be applied is positioned by the focusing means 522 near the bottom of the predetermined cutting groove 25, as shown in Fig. 8C. Thereafter, the pulsed laser beam having an absorption wavelength for the substrate 20 and the functional layer 21 is applied to the wafer 2 by the focusing means 522, and the chuck table 51 is rotated in the direction indicated by an arrow X1 in Fig. 8A at a predetermined feed speed. When the other end, ie the one shown in Fig. 8B, the right end of the predetermined cutting groove 25, as shown in Fig. 8B, the position directly below the focusing means 522 is reached, the application of the pulsed laser beam is stopped and the movement of the chuck table 51 is also stopped (laser processing step).

[0038] Thereafter, the clamping table 51 is moved in the direction of the arrow Y (in the direction corresponding to the sheet plane of the Fig. 8B perpendicular direction) by an amount corresponding to the pitch of the cutting grooves 25, ie, corresponding to the pitch of the dividing lines 23. Thereafter, the pulsed laser beam is applied from the focusing means 522 to the semiconductor wafer 2, and the chuck table 51 is moved in the direction indicated by an arrow X2 in Fig. 8B at a predetermined feed speed. When the clamping table 51 is moved to the direction shown in Fig. 8A, the application of the pulsed laser beam is stopped and the movement of the chuck table 51 is also stopped.

[0039] By performing the above-mentioned laser processing step, a laser-processed groove 24 having a depth reaching the protective tape 3 is formed in the part 201 of the substrate 20 and the functional layer 21 of the semiconductor wafer 2 along each cut groove 25, as shown in Fig. 8D. As a result, the portion 201 of the substrate 20 and the functional layer 21 in the area corresponding to each dividing line 23 are divided by the laser-machined groove 24. The above-mentioned laser processing step is performed similarly along all the dividing lines 23 (along all the cutting grooves 25), whereby a similar laser-machined groove 24 is formed in the portion 201 of the substrate 20 and the functional layer 21 of the semiconductor wafer 2 along each dividing line 23, each laser-machined groove 24 having a depth reaching the protective tape 3. That is, the semiconductor wafer 2 is divided into the individual device chips corresponding to the devices 22.In this laser processing step, by applying the pulsed laser beam having an absorption wavelength suitable for the substrate 20 and the functional layer 21, residual particles can be generated from the semiconductor wafer 2. However, since the protective tape 3 is applied close to the functional layer 21 in such a way that the adhesive layer 31 of the protective tape 3 is in close contact with the components 22 formed by the functional layer 21, adhesion of the residual particles to the components 22 can be prevented. The residual particles generated by applying the pulsed laser beam adhere to the adhesive layer 31 of the protective tape 3 in an area corresponding to each parting line 23.

[0040] In the second preferred embodiment of the above-mentioned wafer dividing step, the cutting groove 25 is formed in the substrate 20 of the semiconductor wafer 2 along each dividing line 23. Compared with the first preferred embodiment of the wafer dividing step described in the Fig. 4A to 4C, the power of the pulsed laser beam can be reduced accordingly in the second preferred embodiment of the wafer dividing step.

[0041] After performing the above-mentioned wafer dividing step, a wafer supporting step is performed in such a manner that the back side 20B of the substrate 20 of the semiconductor wafer 2 is attached to a dicing tape, and the dicing tape is supported in its edge portion on an annular frame having an inner opening capable of receiving the semiconductor wafer 2 and thereby supporting the semiconductor wafer 2 through the dicing tape on the annular frame. More specifically, as shown in FIGS. Fig. 9A and Fig. 9B, a separation tape T is preliminarily supported at its peripheral portion on an annular frame F having an inner opening, so that the inner opening of the annular frame F is closed by the separation tape T. In this state, the back surface 20b of the substrate 20 of the semiconductor wafer 2 is attached to the front surface (adhesive side) of the separation tape T, which is exposed to the inner opening of the annular frame F. Accordingly, the protective tape 3 attached to the front surface 21a of the functional layer 21 of the semiconductor wafer 2 is directed upward in the state where the semiconductor wafer 2 is attached to the separation tape T. For example, the separation tape T is composed of a polyethylene film having a thickness of 100 μm and an adhesive layer formed on the front surface of the polyethylene film.While the back side 20b of the substrate 20 of the semiconductor wafer 2 is attached to the separation tape T which is preliminarily bonded at its peripheral portion to the annular frame F in this embodiment. Fig. 9A and Fig. 9B, the separation tape T may be attached to the back surface 20b of the substrate 20 of the semiconductor wafer 2 and at the same time, the edge portion of the separation tape T may be mounted to the annular frame F.

[0042] After performing the above wafer supporting step, a protective tape peeling step as shown in Fig. 10A, in such a way that the protective tape 3, which is attached to the front side 21a of the functional layer 21 of the semiconductor wafer 2, is peeled off. As shown in Fig. As shown in Fig. 10B, residual particles 26 generated by applying the pulsed laser beam in the wafer dividing step adhere to the adhesive layer 31 of the protective tape 3 peeled off from the front surface 21a of the functional layer 21 of the semiconductor wafer 2. In this way, residual particles 26 generated by applying the pulsed laser beam in the wafer dividing step adhere to the adhesive layer 31 of the protective tape 3, so that the residual particles 26 are prevented from adhering to the devices 22 formed by the functional layer 21 of the semiconductor wafer 2.

[0043] After performing the above-mentioned wafer support step and the protective tape peeling step, a tape expanding step is performed in such a way that the separation tape T attached to the backside 20b of the substrate 20 of the semiconductor wafer 2 is expanded to increase the spacing between the devices 22 that are already separated from each other. This tape expanding step is performed using a tape expanding device 7, which is shown in Fig. 11. The Fig. The band expanding device 7 shown in Fig. 11 includes a frame holding means 71 for holding the annular frame F, a band expanding means 72 for expanding the separating band T supported by the annular frame F held via the frame holding means 71, and a fetching clamp 73. The frame holding means 71 comprises an annular frame holding member 711 and a plurality of clamps 712 as fastening means provided on the outer periphery of the frame holding member 711. The upper surface of the frame holding member 711 functions as a holding surface 711a for mounting the annular frame F thereon. The annular frame F mounted on the mounting surface 711a is fixed to the frame holding member 711 by the clamps 712. The frame holding means 71 is supported by the band expanding means 72 so as to be movable in the vertical direction.

[0044] The belt expanding means 72 includes an expanding drum 721 provided within the annular frame holding member 711. The expanding drum 721 has an outer diameter smaller than the inner diameter of the annular frame F and an inner diameter larger than the outer diameter of the semiconductor wafer 2 attached to the separation belt T supported by the annular frame F. The expanding drum 721 has a supporting flange 722 at the lower end of the drum 721. The belt expanding means 72 further includes a supporting means 723 for vertically movably supporting the annular frame holding member 711. The supporting means 723 is composed of a plurality of air cylinders 723a provided on the supporting flange 722. Each air cylinder 723a is provided with a piston rod 723b connected to the lower surface of the annular frame holding member 711.The support means 723 composed of this plurality of air cylinders 723a has the function of moving the annular frame holding member 711 in the vertical direction to selectively form a position in . Fig. 12A, in which the mounting surface 711a is substantially at the same height as the upper end of the expanding drum 721, and to assume an expanding position in which the mounting surface 711a, as in Fig. 12B, is at a lower height than the upper end of the expanding drum 721 by a predetermined amount.

[0045] The tape expanding step using the tape expanding device 7 will now be described with reference to the Fig. 12A to 12C. As described in Fig. As shown in Fig. 12A, the annular frame F, which supports the semiconductor wafer 2 by the separation tape T, is mounted on the mounting surface 711A of the frame holding member 711 by the frame holding means 71 and fixed to the frame holding member 711 by the clamps 712 (frame holding step). At this time, the frame holding member 711 is Fig. 12A. Thereafter, the air cylinders 723a as supporting means 723 of the belt expanding means 72 are actuated to move the frame holding member 711 to the position shown in Fig. 12B. Accordingly, the annular frame F, which is fixed to the mounting surface 711a of the frame holding member 711, is also lowered, so that the separating belt T, which is supported on the annular frame F, comes to rest against the upper end of the expanding drum 721 and, as shown in Fig. 12B (tape expanding step). As a result, a force acts on the semiconductor wafer 2 attached to the separation tape T in the radial direction of the semiconductor wafer 2. Accordingly, an increased distance S is formed between any adjacent ones of the individual components 22, which, as shown in Fig. 12B, are already separated from each other.

[0046] Thereafter, the pickup clamp 73 is actuated to hold each component 22 by means of a suction force and pull it off the separating belt T, whereby the components 22, as shown in Fig.12C, are individually picked up or picked up (picking up step). Thereafter, each component 22, as a component die thus picked up, is transferred to a tray or a position (not shown) where a die bonding step is performed. By performing the tape expanding step, the increased distance S is formed between any adjacent ones of the individual components 22 attached to the separation tape T, so that each component 22 can be easily picked up without contact with its adjacent component 22 in the picking up step.

Claims

[1] Wafer processing method for dividing a wafer (2) into a plurality of individual components (22) along a plurality of separation lines (23) formed on a front side of the wafer (2), wherein the wafer (2) is composed of a substrate (20) and a functional layer (21) formed on a front side (20a) of the substrate (20), and individual components (22) are formed from the functional layer (21) and are divided by the separation lines (23), the wafer processing method comprising: a protective tape applying step for applying a protective tape (3) having an adhesive layer (31) to a front side (21a) of the functional layer (21) of the wafer (2) in a state in which the adhesive layer (31) of the protective tape (3) is in contact with the front side (21a) of the functional layer (21); and a wafer dividing step of holding the wafer (2) processed by the protective tape applying step on a holding surface of a chuck table (51) in a state where the protective tape (3) is in contact with the holding surface, and then applying a laser beam (LB) having an absorption wavelength for the substrate (20) and the functional layer (21) of the wafer (2) from a back side (20b) of the substrate (20) along each dividing line (23) to form a laser-machined groove (24) with a depth that reaches the protective tape (3) along each dividing line (23), thereby dividing the wafer (2) into individual component chips corresponding to the individual components (22), wherein the protective tape (3) is applied close to the front side (21a) of the functional layer (21) in the protective tape application step in such a way that the adhesive layer (31) of the protective tape (3) comes into close contact with the components (22) formed by the functional layer (21) to prevent the adhesion of residual particles to a front side of each component (22), the residual particles being generated from the wafer (2) by applying the laser beam (LB) in the wafer dividing step along each dividing line (23), wherein, after applying the protective tape (3) to the front side (21a) of the functional layer (21) of the semiconductor wafer (2) in the protective tape applying step, pressure is exerted on the protective tape (3), whereby the adhesive layer (31) is brought into close contact with the components (22) formed by the functional layer (21), and wherein a pressure roller (4) is rolled on the protective tape (3) to exert pressure on the protective tape (3). [2] A wafer processing method according to claim 1, wherein the residual particles generated by the application of the pulsed laser beam (LB) adhere to the adhesive layer (31) of the protective tape (3) in a region corresponding to each parting line (23). [3] A wafer processing method according to any preceding claim, wherein in the wafer separating step, the focal point (P) of the pulsed laser beam (LB) to be applied by the focusing means (522) is set near the back surface (20B) of the substrate (20) to form a laser-machined groove (24) having a depth reaching the protective tape (3) from the back surface (20B) of the substrate (2).

Images