TWI813624B - Wafer processing method - Google Patents

Abstract

[Project] Smoothly divide wafers with functional layers including low-k. [Solution] It includes: a grinding step of grinding the back surface of the wafer; and irradiating the first laser beam with a wavelength that is absorptive with respect to the functional layer to the surface of the wafer to form a laser processing groove to break the surface of the wafer. The laser processing groove formation step of the functional layer; and condensing the second laser beam with a wavelength penetrating relative to the wafer from the back side of the wafer to a specific depth position to form the groove inside the wafer. In the modification layer formation step, in the grinding step, the outer peripheral remaining area hangs downward relative to the device area, and then, by grinding from the back side, the thickness of the wafer is increased from the inner peripheral edge of the outer peripheral remaining area toward The outer peripheral edge becomes thicker. In the laser processing groove forming step, the first laser beam is focused at a height position of the surface of the peripheral edge outside the outer peripheral remaining area, forming a shape from the inner peripheral edge to the outer peripheral edge. The laser machined trench becomes deeper.

Description

Wafer processing methods

The present invention relates to a method of processing a wafer having devices formed on its surface.

On the surface of a device wafer such as an IC chip on which a semiconductor device is mounted, functional layers including various layers constituting the device are laminated. The functional layer includes various layers such as a wiring layer that transmits electrical signals or an interlayer insulating layer that insulates between wiring layers. In recent years, in order to reduce parasitic capacitance formed between wiring layers, so-called Low-k films with low dielectric constants are used as interlayer insulating layers included in functional layers, in order to improve the processing capabilities of device chips.

When manufacturing a device wafer, a plurality of intersecting planned division lines are set on the surface of a disc-shaped semiconductor wafer, and devices are formed in each area divided by the planned division lines, and then divided along the planned division lines. wafer. However, the Low-k film is very fragile. When the wafer is divided together with the Low-k film, there is a risk that the Low-k film will peel off from the wafer and cause damage to the device.

Therefore, the functional layer is removed by ablation processing in advance, and processing grooves are formed along the planned division lines. When performing ablation processing, a laser beam with a wavelength that is absorptive to the functional layer is focused on the surface of the wafer, and the wafer and the laser processing unit are relatively moved along the planned division line.

Furthermore, the wafer is irradiated from the back side with a laser beam having a wavelength that is penetrating to the wafer, and the light is focused inside the wafer along the planned division line to form a modified layer through a multi-photon absorption process. After that, when the crack is extended from the modified layer at the bottom of the processing trench, the wafer containing the Low-k film can be divided appropriately. With this method, the functional layer is not divided during the division of the wafer. , so peeling off of the Low-k film or damage to the device is suppressed.

However, as a semiconductor packaging technology, bumps made of metal or the like that are electrically connected to the device are formed on the surface of the device, the surface of the wafer and the bumps are covered and sealed with a sealing material, and the wafer is divided into individual devices. Chip technology has been put into practical use. This packaging technology is called WL-CSP (Wafer-level Chip Size Package). In WL-CSP, the size of the device chip formed by dividing the wafer remains the same as the package size, which contributes to the miniaturization and weight reduction of the device.

Furthermore, in order to form a thin device wafer, the wafer before being divided is ground from the back side. When performing grinding, the wafer is loaded onto a chuck mounting table provided in the grinding device, and the rotating grinding wheel is brought into contact with the back surface of the wafer. At this time, the surface side of the wafer with the protruding bumps faces the upper surface of the chuck mounting table.

The surface side of the wafer includes a device area where devices are formed, and an outer peripheral remaining area where devices are not formed. The bumps are provided in the device area of the wafer and are not placed in the remaining peripheral areas. Therefore, during grinding of the wafer, the device area of the wafer is supported on the chuck mounting table via the bumps, while the remaining peripheral area is not supported. As a result, when the wafer is placed on the chuck mounting table, the remaining outer peripheral area hangs down relative to the device area (see Patent Document 1).

When the back side of the wafer is ground while the remaining peripheral area is hanging down, the thickness of the wafer after grinding is that the remaining peripheral area is thicker than the device area. In this state, when a processing groove is formed on the front side of the wafer and a laser beam is irradiated from the back side to form a modified layer inside the wafer, the processing groove and modified layer remain in the peripheral area compared to the device area. The distance between layers becomes larger. When the distance between the processing groove and the modified layer becomes larger, it is difficult for cracks to reach the processing groove when the wafer is divided, and it is difficult for the wafer to break in the remaining peripheral area.

Therefore, a processing method has been proposed in which the relative movement speed of the nip mounting table and the laser processing unit is changed during ablation processing (see Patent Document 2). In this processing method, when the laser beam is irradiated, the relative movement speed is reduced as it approaches the outer periphery. In this way, as it approaches the outer periphery, ablation processing is performed with high intensity, forming a deep processing groove. Therefore, in the remaining peripheral area, cracks can easily extend from the modified layer to the processing groove, and the remaining peripheral area can be broken like the device area. [Prior technical literature] [Patent Document]

[Patent Document 1] Japanese Patent Application Publication No. 2013-21017 [Patent Document 2] Japanese Patent Application Publication No. 2017-45965

[Problem to be solved by the invention]

In this processing method, it is necessary to appropriately change the relative movement speed of the cartridge mounting table and the laser processing unit during laser beam irradiation, but it is not easy to control the relative movement speed with high precision. Furthermore, since the structure of a system that can appropriately control changes in the relative moving speed is complicated, it is costly to introduce the system into a laser processing device.

The present invention was created in view of such problems, and an object thereof is to provide a wafer processing method that can smoothly divide a wafer provided with bumps without changing the control system of the laser processing apparatus. [Means used to solve problems]

According to one aspect of the present invention, a wafer processing method is provided, in which a functional layer is laminated on the surface, a device area including a plurality of devices including the functional layer is formed, and a device area surrounding the functional layer is provided. There is a remaining area outside the device area, and a plurality of intersecting planned division lines are set on the surface side so as to partition a plurality of the devices, and a plurality of bumps electrically connected to each device are arranged on the surface side. The circle is processed along the planned dividing line together with the functional layer, and is characterized in that it is provided with a grinding step in which the crystal is ground in a grinding device equipped with a nip plate mounting table with the surface facing downwards. The wafer is held on the chuck mounting table, and the back surface of the wafer is ground to thin the wafer; the laser processing groove forming step is to, after the grinding step, absorb the wavelength with respect to the functional layer The first laser beam is irradiated to the surface of the wafer to form a laser processing groove to separate the functional layer; and a modified layer forming step, which is to perform the step of forming the laser processing groove relative to the modified layer after the laser processing groove forming step is performed. The second laser beam with a wavelength penetrating the wafer is focused from the back side of the wafer along the planned division line at a specific depth position, and is formed inside the wafer as a starting point for division of the wafer. The modified layer, in the grinding step, when the wafer is held on the chuck mounting platform, the device area of the wafer is supported on the chuck mounting platform through the bump, and then, according to the The wafer is ground from the back side by the grinding unit, and the thickness of the wafer becomes thicker from the inner peripheral edge toward the outer peripheral edge of the outer peripheral remaining area. In the laser processing groove forming step, the first laser beam is focused on The height position of the surface of the outer peripheral edge of the outer peripheral remaining area is irradiated along the planned division line to form the laser processing groove that becomes deeper as it approaches the outer peripheral edge from the inner peripheral edge.

In the step of forming the laser processing groove, before or after the first laser beam is focused on a height position of the surface of the peripheral edge outside the outer peripheral remaining area and irradiated along the planned dividing line, further, It is preferred that the first laser beam is focused at a height position on the surface of the device area of the wafer and irradiated along the planned division line. [Effects of the invention]

In the wafer processing method according to one aspect of the present invention, in the laser processing groove forming step, the first laser beam having absorptivity with respect to the functional layer is irradiated along the planned division line. When the first laser beam is irradiated to the surface of the wafer, the wafer is ablated and laser-processed grooves are formed on the surface.

Here, the smaller the difference between the height position of the surface of the wafer irradiated with the first laser beam and the height position of the focusing point of the first laser beam, the ablation process is performed with high intensity, and the laser processing is formed The trench becomes deeper. In the laser processing groove forming step, since the first laser beam is focused at a height position on the surface of the wafer at the periphery outside the outer peripheral remaining region, the laser processing groove is formed at the periphery outside the outer peripheral remaining region. deep.

When the first laser beam is irradiated to the surface of the wafer from the outer peripheral edge of the outer peripheral remaining area to the inner peripheral edge along the planned dividing line without changing the height of the focusing point of the first laser beam, this difference will It gradually becomes larger, the intensity of the ablation process gradually becomes weaker, and the laser processing groove is formed and gradually becomes shallower. Furthermore, in the device area, a relatively shallow laser processing groove is formed as it is.

When the laser-processed groove is formed in this way in the laser-processed groove forming step, the height position of the bottom of the laser-processed groove is relatively uniform over the entire length of the laser-processed groove. Therefore, the distance between the modified layer formed in the modified layer forming step and the laser processing groove is relatively uniform. Therefore, cracks can easily extend from the modified layer to the laser-processed groove along the entire length of the planned dividing line.

In the laser processing groove forming step, the wafer and the laser processing unit can be relatively moved at a certain speed. At this time, there is no need to change the focusing height of the first laser beam. That is, the control of the laser processing unit or the nip stage becomes extremely easy, and the wafer can be smoothly divided without making changes to the control system of the laser processing device.

Therefore, according to the present invention, the wafer processing method of dividing the wafer provided with bumps can be smoothly performed without changing the control system of the laser processing device.

Embodiments of the present invention will be described with reference to the attached drawings. First, a wafer, which is a workpiece of the processing method according to this embodiment, will be described using FIG. 1(A) . FIG. 1(A) schematically shows a perspective view of a wafer with a functional layer formed on the surface. The wafer 1 is a substantially disc-shaped substrate made of, for example, silicon, SiC (silicon carbide), or other semiconductor materials, or sapphire, glass, quartz, or other materials.

The functional layer 3 is stacked on the surface 1a of the wafer 1 . The functional layer 3 includes various layers such as a wiring layer for transmitting electrical signals or an interlayer insulating layer for insulating between wiring layers. In order to reduce the parasitic capacitance formed between wiring layers, for example, a so-called Low-k film having a low dielectric constant is used as an interlayer insulating film. Low-k films are known to include inorganic films such as SiOF and SiOB (borosilicate glass) or polymer films such as polyimide and p-xylene films, which are also organic films. They are very fragile. membrane.

The surface 1a of the wafer 1 is divided into a plurality of areas by a plurality of planned division lines (dicing lanes) 9 arranged in a grid shape. In each area divided by the plurality of planned division lines 9, an IC (Integrated circuit) is formed. Devices such as 11. The functional layer 3 is included in the device 11, and the various layers included in the functional layer 3 are formed into specific shapes through photolithography processes to perform specific functions. Finally, the wafer 1 is thinned and divided along the planned dividing lines 9 to form individual device wafers.

In order to reduce the manufacturing cost of the device wafer, as many devices 11 as possible are formed on the surface 1 a of the wafer 1 . However, due to the shapes of the wafer 1 and the device 11 , and for the convenience of the manufacturing process of the device wafer, a remaining area 7 without the outer periphery of the device 11 is disposed near the outer periphery of the wafer 1 . In this regard, the area surrounded by the outer peripheral remaining area 7 and forming the device 11 is called a device area 5 . However, the boundary 15 between the outer peripheral remaining area 7 and the device area 5 does not necessarily need to be clear.

On the surface 1 a of the wafer 1 , a plurality of protruding bumps 13 serving as connection terminals of the device chip are formed. The bumps 13 are made of metal such as gold or copper and are electrically connected to the device 11 . In addition, the bumps 13 are formed only in the device region 5 of the surface 1 a of the wafer 1 and are not formed in the outer peripheral remaining region 7 where the device 11 is not formed.

Next, each processing device used in the processing method of the wafer 1 according to this embodiment will be described. In addition, the processing method is not limited to this, and some or all of the processing devices described below may not be used.

FIG. 2(A) schematically shows a grinding device 2 for grinding the back surface 1 b side of the wafer 1 to thin the wafer 1 . The grinding device 2 includes a chuck mounting base 4 that attracts and holds a wafer 1 as a workpiece, and a grinding unit 6 that grinds the wafer 1 held on the chuck mounting base 4 .

The nip holding table 4 is provided with a porous member (not shown) exposed on the upper surface side, and a suction source (not shown) connected to the porous member. The upper surface of the porous member serves as a holder for holding the wafer 1 . noodle. When the wafer 1 is placed on the holding surface and negative pressure generated by the suction source acts on the wafer 1 through the holes of the porous member, the wafer 1 is attracted and held on the chuck mounting table 4 . Furthermore, the clamping plate mounting base 4 can rotate around an axis perpendicular to the holding surface.

The grinding unit 6 includes a spindle 10 extending in a direction perpendicular to the holding surface of the clamping table 4 , a roller bracket 12 disposed at the lower end of the spindle 10 , and a grinding wheel 14 mounted on the roller bracket 12 . The grinding wheel 14 includes a base 16 and a grinding stone 18 installed below the base 16 . When the spindle 10 is rotated around a direction perpendicular to the holding surface, the grinding wheel 14 can be rotated. Furthermore, the grinding unit 6 can be raised and lowered in a direction perpendicular to the holding surface of the clamp plate mounting table 4 .

When the wafer 1 is held on the holding surface of the nip holder 4, the grinding wheel 14 and the nip holder 4 are rotated, the grinding unit 6 is lowered, and the grinding stone 18 is brought into contact with the wafer 1, the wafer 1 Be ground and processed.

FIG. 3(B) schematically shows a first laser processing device 20 that irradiates the surface 1 a side of the wafer 1 with a first laser beam and forms a laser processing groove 19 in the wafer 1 by ablation processing. The first laser processing apparatus 20 includes a nip holding table 22 that attracts and holds the wafer 1 , and a first laser processing unit 24 that irradiates the wafer 1 held on the nip holding table 22 with a first laser beam. The nip plate mounting base 22 is the same as the nip plate mounting base 4 provided in the above-mentioned grinding device 2 .

The first laser processing unit 24 can oscillate the first laser beam 26 a with a wavelength that is penetrative to the functional layer 3 . The first laser beam 26a is focused to a specific height position by the processing head 26. The first laser processing unit 24 can move up and down in a direction perpendicular to the holding surface of the clamping table 22 . When the first laser processing unit 24 is raised and lowered, the focusing height of the first laser beam 26a can be changed. The clamp mounting table 22 and the first laser processing unit 24 move relatively in a direction parallel to the holding surface.

First, the first laser processing unit 24 is positioned at a specific height on the extension line of the planned division line 9 of the wafer 1 . Then, while oscillating the first laser beam 26 a in the first laser processing unit 24 , the first laser processing unit 24 and the nip plate mounting table 22 are relatively moved along the planned division line 9 . In this way, the ablation process is performed along the planned division line 9 .

After processing along one planned dividing line 9 , the pallet mounting table 22 and the first laser processing unit 24 are relatively moved in a direction parallel to the holding surface and perpendicular to the planned dividing line 9 , and then the other dividing lines are moved along the other planned dividing lines 9 . Predetermined line 9 is processed. After processing is performed along the planned dividing line 9 in one direction, the clamp tray mounting table 22 is rotated at a specific angle, and processing is similarly performed along the divided dividing line 9 in the other direction. In this way, the laser processing grooves 19 can be formed along all the planned dividing lines 9 .

FIG. 6 schematically shows that the second laser beam is focused from the back surface 1b side of the wafer 1 to the inside of the wafer 1. Through the multi-photon absorption process, a modification that becomes the starting point for segmentation is formed inside the wafer 1. The second laser processing device 32 of the second layer. The second laser processing device 32 includes a chuck mounting table 34 and a second laser processing unit 36 . The nip plate mounting base 34 is the same as the nip plate mounting base 4 provided in the above-mentioned grinding device 2 .

The second laser processing unit 36 can oscillate the second laser beam 38a with a wavelength that is penetrative to the functional layer 1 . The second laser beam 38a is focused to a specific height position inside the wafer 1 by the processing head 38 . The second laser processing unit 36 can be raised and lowered in a direction perpendicular to the holding surface of the clamping table 34 . When the second laser processing unit 36 is raised and lowered, the focusing height of the second laser beam 38a can be changed.

The second laser processing unit 36 and the chuck mounting base 34 are relatively movable in a direction parallel to the holding surface of the chuck mounting base 34 . In the second laser processing device 32, like the first laser processing device 20, the second laser beam 38 can be irradiated into the inside of the wafer 1 along each planned division line 9 of the wafer 1. The multi-photon absorption process forms a modified layer 27 along the planned division line 9 inside the wafer 1 .

Next, each step of the wafer processing method related to this embodiment will be described. First, in the grinding device 2 , a grinding step of grinding the back surface 1 b of the wafer 1 to thin the wafer 1 is performed. Before the wafer 1 is loaded into the grinding device 2 , a protective member 17 is attached to the surface 1 a side of the wafer 1 to protect the surface 1 a side. FIG. 1(B) is a perspective view schematically showing the protective member 17 attached to the surface 1 a of the wafer 1 .

The protective member 17 includes an adhesive layer 17b (see FIG. 2(A)) having an adhesive surface, and a film-like base material layer 17A (see FIG. 2(A)) that supports the adhesive layer 17b. The adhesive layer 17b side is adhered to Surface 1a side of wafer 1. For example, PO (polyolefin), PET (polyethylene terephthalate), polyvinyl chloride, polystyrene, etc. are used for the base material layer 17a. Furthermore, for the adhesive layer 17b, for example, silicone rubber, acrylic material, epoxy material, or the like is used.

When the protective member 17 is adhered to the surface 1a side, the adhesive layer 17b deforms following the uneven shape of the surface 1a. Furthermore, the adhesive layer 17 b is, for example, an ultraviolet curable resin. When the protective member 17 is peeled off from the wafer 1 , the protective member 17 is irradiated with ultraviolet rays to harden the adhesive layer 17 b , and the peeling becomes easy.

FIG. 2(A) is a cross-sectional view schematically showing the grinding step. In the grinding device 2, the wafer 1 with the protective member 17 attached to the surface 1a is loaded. With the surface 1a facing downward, the wafer 1 is placed on the holding surface of the nip plate 4 with the protective member 17 interposed therebetween, and the suction source of the nip plate 4 is actuated to attract and hold the wafer 1 on the nip plate. Placing table 4. In this way, the surface to be ground, that is, the back surface 1 b of the wafer 1 is exposed upward.

Next, the grinding unit 6 is lowered toward the clamp mounting base 4 while rotating the clamp mounting base 4 and the grinding wheel 14 about an axis perpendicular to the holding surface. In this way, the grinding stone 18 comes into contact with the back surface 1b of the wafer 1, and the grinding process starts. Then, the grinding unit 6 is lowered to a specific height position so that the wafer 1 has a specific thickness. In this way, the wafer 1 is thinned to a specific thickness.

FIG. 2(B) is an enlarged cross-sectional view schematically showing the wafer 1 to be ground. The functional layer 3 is omitted in Fig. 2(B). As shown in FIG. 2(B) , when the chuck stage 4 holds the wafer 1 , the adhesive layer 17 b of the protective member 17 deforms following the shape of the surface 1 a side of the wafer 1 . Furthermore, the device area 5 of the wafer 1 is supported on the chuck mounting table 4 mainly through the bumps 13 .

In contrast, no bump 13 is formed in the outer peripheral remaining area 7 . Compared with the bumps 13 , the adhesive layer 17 b has higher flexibility and weaker ability to support the wafer 1 . Therefore, when the wafer 1 is placed on the chuck mounting table 4 , the outer peripheral remaining area 7 hangs down relative to the device area 5 . When the backside 1b side of the wafer 1 is ground while the remaining outer peripheral area 7 hangs down, as shown in FIG. 2(B) , the thickness of the wafer 1 after grinding is from the inner peripheral edge of the outer peripheral remaining area 7 toward the outer peripheral edge. become thicker.

In the processing method according to this embodiment, a laser processing groove forming step is performed next. The laser processing groove forming step is implemented by the first laser processing device 20 described above. In addition, before the wafer 1 is loaded into the first laser processing apparatus 20 , the protective member 17 adhered to the surface 1 a side of the wafer 1 is peeled off in advance.

FIG. 3(A) is a perspective view schematically showing the protective member 17 peeling off the surface 1 a of the wafer 1 . When the protective member 17 uses an ultraviolet curable resin, peeling becomes easy when the protective member 17 is cured by irradiating ultraviolet rays. Furthermore, as shown in FIG. 3(A) , the dicing tape 21 is adhered to the back surface 1b side of the wafer 1 . This dicing tape 21 has the same structure as the protective member 17 .

FIG. 3(B) is a perspective view schematically showing the laser processing groove forming step. The wafer 1 with the surface 1 a facing upward is placed on the holding surface of the nip table 22 of the first laser processing apparatus 20 , and the wafer 1 is attracted and held on the nip table 22 . In this way, the surface 1 a of the wafer 1 to be processed is exposed upward. Furthermore, the camera 28 of the first laser processing unit 24 is used to capture the planned division line 9 on the surface 1 a of the wafer 1 , and the relative position of the chuck mounting table 4 and the first laser processing unit 24 is adjusted.

Next, the surface 1 a of the wafer 1 is irradiated with the first laser beam 26 a having a wavelength that is absorptive with respect to the functional layer 3 along the planned division line 9 . In this way, the functional layer 3 is removed by the ablation process, and a laser-processed groove 19 for dividing the functional layer 3 is formed on the surface 1 a of the wafer 1 . FIG. 4(A) schematically shows the wafer 1 after the laser processing groove forming step. As shown in FIG. 4(A) , when the laser processing groove forming step is performed, the laser processing groove 19 is formed on the surface 1 a side of the wafer 1 .

FIG. 4(A) is a cross-sectional view schematically showing the condensing position of the first laser beam 26a, and FIG. 4(B) is a cross-sectional view schematically showing an enlarged outer peripheral remaining area 7. 4(A) and 4(B) show the height position of the processing head 26 when the first laser beam 26a is irradiated near the peripheral edge outside the peripheral remaining area 7. When performing the laser processing groove forming step, the first laser processing unit 24 is set such that the focusing point 30 of the first laser beam 26 a is aligned with the height position of the surface 1 a of the outer peripheral edge of the outer peripheral remaining area 7 the height.

FIG. 4(C) is an enlarged cross-sectional view schematically showing the device region 5 . 4(A) and 4(C) show the height position of the processing head 26 when the device area 5 is irradiated with the first laser beam 26a. During the ablation process, the height position of the processing head 26 is maintained. Therefore, when the ablation process is performed on the device region 5 of the wafer 1 , the focusing point 30 of the first laser beam 26 a is arranged at a height position deviated from the height position of the surface 1 a of the wafer 1 .

The smaller the difference between the height position of the surface 1a of the wafer 1 and the height position of the focusing point 30 of the first laser beam 26a, the deeper the laser processing groove 19 formed when the ablation process is performed with high intensity. Therefore, in the laser processing groove forming step, a deep laser processing groove 19 is formed on the outer peripheral edge of the remaining peripheral area 7 , and the formed laser processing groove 19 becomes shallower as it approaches the inner peripheral edge from the outer peripheral edge. Furthermore, in the device area 5, a shallow laser processing groove 19 is formed.

Therefore, when the laser-processed trench forming step is performed, the laser-processed trench 19 is formed with a depth corresponding to the thickness of the wafer 1 where the laser-processed trench 19 is formed. Therefore, the height position of the bottom of the laser processing groove 19 becomes the same height position over the entire length.

In the processing method related to this embodiment, a modified layer forming step is performed next. The modified layer forming step is implemented using the second laser processing device 32 described above. In addition, before loading the wafer 1 into the second laser processing device 32 , an expansion tape is adhered to the surface 1 a side of the wafer 1 in advance.

FIG. 5 is a perspective view schematically showing the expansion tape 23 attached to the surface 1 a side of the wafer 1 . The expansion tape 23 is, for example, adhered to the annular frame 25 at its outer periphery, and is adhered to the surface 1 a of the wafer 1 inside the opening of the frame 25 . In addition, the expansion tape 23 has the same structure as the protective member 17 mentioned above, for example. Furthermore, the annular frame 25 is made of metal or other materials.

The wafer 1 is loaded into the second laser processing apparatus 32 in a state where the frame unit 33 in which the tape 23 and the frame 25 are integrated is expanded. Fig. 6 is a cross-sectional view schematically showing the steps of forming a modified layer. With the surface 1 a side facing downward, the wafer 1 is placed on the holding surface of the nip holding table 34 via the expansion tape 23 , and the wafer 1 is attracted and held on the nip holding table 34 . In this way, the back surface 1b side of the wafer 1, which is the irradiated surface of the second laser beam 38a, faces upward.

Next, the second laser beam 38 a having a wavelength penetrating to the wafer 1 is focused on a specific height position of the wafer 1 along the planned division line 9 through the dicing tape 21 . In this way, the modified layer 27 is formed near the focusing point 40 of the second laser beam 38a through the multi-photon absorption process.

In addition, the dicing tape 21 may be peeled off before the irradiation of the second laser beam 38a. In this case, the second laser beam 38 a is irradiated to the wafer 1 without passing through the dicing tape 21 .

The modified layer 27 serves as a starting point for dividing the wafer 1 . That is, when an external force or the like is applied to the wafer 1 to cause the crack to extend from the modified layer 27 to the bottom of the laser processing groove 19 , the wafer 1 can be divided along the planned division line 9 . In addition, the cracks may be extended simultaneously with the formation of the modified layer 27 . Here, the height position of the bottom of the laser processing groove 19 is the same height over the entire length as described above. That is, the distance between the bottom of the laser-processed groove 19 and the modified layer 27 is the same anywhere along the planned division line 9 .

When the crack is extended from the modified layer 27 toward the laser-processed groove 19, the shape of the extension changes depending on the distance between the modified layer 27 and the laser-processed groove 19. Furthermore, the larger this distance is, the more difficult it is for the crack to properly extend, making it difficult for the wafer 1 to break.

Here, for the sake of comparison, a processing method related to a comparative example in which the height position of the bottom of the laser-processed groove 19 is not the same over the entire length will be described. FIG. 9(A) is a cross-sectional view schematically showing the condensing position of the first laser beam 26a in the processing method related to this comparative example, and FIG. 9(B) is a schematic view showing an enlarged outer peripheral remaining area. 9(C) is a sectional view schematically showing an enlarged device area.

In the processing method related to this comparative example, when performing the laser processing groove forming step, as shown in FIG. 9(C) , the first laser beam 26a is focused on the first laser processing device 20. The light spot 30 is aligned at a height position of the surface 1 a of the wafer 1 in the device area 5 .

In this case, since the focusing point 30 of the first laser beam 26a is positioned at a high position on the surface of the wafer 1 in the device area 5, the ablation process is performed with a stronger intensity to form a relatively deep laser beam. Processing ditch 19. On the other hand, as shown in FIG. 9(B) , in the outer peripheral remaining area 7, the height position of the focusing point 30 of the first laser beam 26a is greatly different from the height position of the surface 1a of the wafer 1. The ablation process is performed with a weak intensity to form a relatively shallow laser processing groove 19 .

In the peripheral remaining area 7, the thickness of the wafer 1 is larger than that of the wafer 1 in the device 5, and the formed laser processing groove 19 becomes shallower. Therefore, as shown in FIGS. 9(B) and 9(C) , the height position of the bottom of the laser processing groove 19 will not be the same throughout its length. Therefore, when the modified layer formation step is subsequently performed to form the modified layer 27 inside the wafer 1 , the distance between the modified layer 27 and the bottom of the laser processing groove 19 will not be the same over the entire length of the planned division line. .

In particular, this distance greatly changes from the inner peripheral edge to the outer peripheral edge within the outer peripheral remaining area 7 . Accordingly, it is difficult to appropriately extend the cracks from the modified layer 27 in the laser processing groove 19 in the vicinity of the outer peripheral remaining area 7 , and when the wafer 1 is divided along the planned division line 9 , the outer peripheral remaining area 7 Near the outer periphery, it becomes difficult to properly break the wafer 1 .

On the other hand, in the processing method according to this embodiment, since the distance between the modified layer 27 and the laser-processed groove 19 is the same over the entire length, the wafer 1 appropriately includes the vicinity of the outer peripheral edge of the outer peripheral remaining region 7 Be broken. Furthermore, in the laser processing groove forming step, there is no need to change the relative speeds of the first laser processing unit 24 and the nip mounting table 22 during the ablation process. Therefore, the wafer 1 provided with the bumps 13 can be smoothly divided without changing the control system of the laser processing apparatus.

After the modified layer forming step is performed, the expansion tape 23 is expanded in order to divide the wafer 1 to form individual device wafers. When the expansion tape 23 is expanded, the dicing tape 21 is peeled off from the back surface 1b side of the wafer 1 as shown in FIG. 7 . Then, the frame unit 33 including the wafer 1 is loaded into the expansion device.

FIG. 8(A) is a cross-sectional view schematically showing the wafer loaded into the expansion device, and FIG. 8(B) is a cross-sectional view schematically showing how the wafer is expanded by the expansion device. The structure of the expansion device 42 will be described. The expansion device 42 includes a cylindrical expansion drum 44 and a frame holding unit 46 surrounding the expansion drum 44 from the outer peripheral side.

The expansion drum 44 is provided with: a push-up portion 54 having an upper surface in contact with the expansion tape 23; a rod 50 that supports the push-up portion 54 from below; and a cylinder 52 that raises and lowers the rod 50. . When the air cylinder 52 is actuated, the push-up part 54 can be raised and lowered between the frame unit loading position 56 and the expansion position 58 . The frame holding unit 46 includes a clamp 48 that holds the frame 25 of the frame unit 33 .

When the frame unit 33 is moved into the expansion device 42, as shown in FIG. 8(A) , the push-up part 54 is arranged at the frame unit carry-in position 56, and the frame unit 33 is placed on the push-up part 54. Furthermore, the frame 25 is held by the clamp 48 .

Next, as shown in FIG. 8(B) , the air cylinder 52 is actuated to raise the push-up portion 54 to the expansion position 58 . In this way, the expansion tape 23 is expanded radially outward. In this way, the distance between the device wafers 31 formed by dividing the wafer 1 becomes wider, making it easier to pick up the device wafers 31 .

By the above, the wafer 1 in which the functional layer 3 including the Low-k film is formed can be smoothly divided to form each device wafer 31 .

In addition, the present invention is not limited to the description of the above embodiment, and can be implemented with various modifications. For example, one aspect of the present invention is not limited to the case where the bottom of the laser processing groove 19 formed in the laser processing groove forming step is at the same height position over the entire length of the planned division line 9 . Depending on the irradiation conditions of the first laser beam 26a in the first laser processing device 20, the bottom of the formed laser processing groove 19 may not be at the same height position over the entire length of the planned division line 9. .

Even in this case, for example, if the change in the height of the bottom of the laser processing groove 19 formed in the laser processing groove forming step from the inner peripheral edge of the outer peripheral remaining area 7 to the outer peripheral edge is smaller than the change in the thickness of the wafer 1 , can also enjoy the effects of the present invention. That is, the wafer 1 in the remaining peripheral area 7 can be divided more smoothly.

Furthermore, in the above embodiment, in the laser processing groove forming step, the focusing point 30 of the first laser beam 26a is arranged at the height of the surface 1a of the wafer 1 in the peripheral remaining area 7, but one aspect of the present invention The form is not limited to this. For example, the light condensing point 30 may be arranged at a height higher than the surface 1 a of the wafer 1 in the outer peripheral remaining area 7 .

Even in this case, the intensity of the ablation process performed in the laser processing groove forming step becomes higher from the inner peripheral edge of the outer peripheral remaining area 7 toward the outer peripheral edge, so that the formed laser processing groove 19 gradually increases from the inner peripheral edge of the outer peripheral remaining area 7 to the outer peripheral edge. The inner periphery becomes darker approaching the outer periphery.

In addition, in the laser processing groove forming step, when the first laser beam 26a is focused on the height position of the surface 1a of the wafer 1 on the outer periphery of the remaining area 7, there is no formation in the device area 5. The case of laser processing trench 19 with sufficient depth. This is because the intensity of the ablation process becomes insufficient in the device area 5 due to reasons such as the distance between the light condensing point 30 and the surface 1 a of the wafer 1 becoming too large.

Therefore, even before or after the first laser beam 26 a is irradiated at a height position of the surface 1 a of the wafer 1 at the outer periphery of the remaining peripheral area 7 , the light is further focused on the surface of the device area 5 The first laser beam 26a may be irradiated at a height position of 1a. That is, in the laser processing groove forming step, the first laser beam 26a may be irradiated a plurality of times by changing the height of the converging position to form the laser processing groove 19 with a uniform height at the bottom.

The structures, methods, etc. related to the above-described embodiments may be appropriately modified and implemented as long as they do not deviate from the scope of the purpose of the present invention.

1‧‧‧wafer 1a‧‧‧Surface 1b‧‧‧Back 3‧‧‧Functional layer 5‧‧‧Installation area 7‧‧‧Remaining area around the periphery 9‧‧‧Scheduled dividing line 11‧‧‧Device 13‧‧‧Bump 15‧‧‧realm 17‧‧‧Protective components 17a‧‧‧Substrate 17b‧‧‧Adhesive layer 19‧‧‧Laser processing trench 21‧‧‧Cutting tape 23‧‧‧Expansion tape 25‧‧‧Frame 27‧‧‧Modification layer 29‧‧‧Crack 31‧‧‧Device chip 33‧‧‧Frame unit 2‧‧‧Grinding device 4, 22, 34‧‧‧Plate holding table 6‧‧‧Grinding unit 10‧‧‧Spindle 12‧‧‧Bracket 14‧‧‧Grinding wheel 16‧‧‧Abutment 18‧‧‧Grinding stone 20. 32‧‧‧Laser processing device 24, 36‧‧‧Laser processing unit 26, 38‧‧‧ processing head 26a, 38a‧‧‧Laser Beam 28‧‧‧Camera unit 30, 40 focus point 42‧‧‧Expansion device 44‧‧‧Expansion drum 46‧‧‧Frame holding unit 48‧‧‧Holding tools 50‧‧‧Bar 52‧‧‧Cylinder 54‧‧‧Push up part 56‧‧‧Frame unit move-in location 58‧‧‧Expanded position

FIG. 1(A) is a perspective view schematically showing a wafer with a functional layer formed on its surface, and FIG. 1(B) is a perspective view schematically showing a protective member attached to the surface of the wafer. FIG. 2(A) is a cross-sectional view schematically showing the cutting step, and FIG. 2(B) is an enlarged cross-sectional view schematically showing the wafer to be ground. FIG. 3(A) is a perspective view schematically showing the peeling of the protective member from the surface of the wafer, and FIG. 3(B) is a perspective view schematically showing the step of forming a laser processing groove. 4(A) is a cross-sectional view schematically showing the focus position of the laser beam during the laser processing groove formation step, and FIG. 4(B) is a cross-sectional view schematically showing an enlarged peripheral remaining area, FIG. 4(C) is a cross-sectional view schematically showing an enlarged device area. FIG. 5 is a perspective view schematically showing a wafer with expansion tape attached to its surface. Fig. 6 is a cross-sectional view schematically showing the steps of forming a modified layer. FIG. 7 is a perspective view schematically showing peeling of the dicing tape from the back surface of the wafer. FIG. 8(A) is a cross-sectional view schematically showing the wafer loaded into the expansion device, and FIG. 8(B) is a cross-sectional view schematically showing how the wafer is expanded by the expansion device. 9(A) is a cross-sectional view schematically showing the focusing position of the laser beam in the laser processing groove forming step in the processing method according to the comparative example, and FIG. 9(B) is an enlarged peripheral remaining area pre-print As for the schematic cross-sectional view, FIG. 9(C) is a schematic cross-sectional view in which the device area is enlarged.

1‧‧‧wafer

1a‧‧‧Surface

1b‧‧‧Back

5‧‧‧Installation area

7‧‧‧Remaining area around the periphery

13‧‧‧Bump

19‧‧‧Laser processing trench

20‧‧‧Laser processing device

21‧‧‧Cutting tape

22‧‧‧Plate holding table

24‧‧‧Laser processing unit

26‧‧‧Processing head

26a‧‧‧Laser Beam

30‧‧‧Focusing point

Claims (2)

A method of processing a wafer, in which a device area having a functional layer laminated on the surface and a plurality of devices including the functional layer is formed, and a plurality of intersecting remaining areas surrounding the outer peripheral area of the device area are formed. The planned dividing line is set on the surface side to partition a plurality of the devices, and a wafer with a plurality of bumps electrically connected to each device is arranged on the surface side, along the planned dividing line along with the function It is characterized in that it has: a grinding step, which is in a grinding device provided with a chuck mounting table, holding the wafer on the chuck mounting table with the surface facing downward, and grinding the wafer The back side of the wafer is thinned; the laser processing groove forming step is to irradiate the first laser beam with a wavelength that is absorptive with respect to the functional layer to the wafer after the grinding step. forming a laser processing groove on the surface to separate the functional layer; and a modified layer forming step, which is to convert a third wavelength having penetrability relative to the wafer after the laser processing groove forming step. 2. The laser beam is focused from the back side of the wafer at a specific depth position along the planned division line to form a modified layer inside the wafer that becomes the starting point for division of the wafer. In the grinding step When the wafer is held on the chuck mounting table, the device area of the wafer is supported on the chuck mounting table via the bump, and then is ground from the back side by the grinding unit. The thickness of the wafer becomes thicker from the inner circumference toward the outer circumference within the remaining area of the outer circumference, In the laser processing groove forming step, the first laser beam is focused at a height position of the surface of the outer peripheral edge of the outer peripheral remaining area and deviated from a height position of the surface of the device area. At a high position, the laser processing groove is irradiated along the planned dividing line and becomes deeper in the remaining outer peripheral area as it approaches the outer peripheral edge from the inner peripheral edge. The wafer processing method as set forth in claim 1, wherein in the laser processing groove forming step, the first laser beam is focused at a height position of the surface of the outer periphery in the remaining area of the outer periphery. , and the height position is deviated from the height position of the surface of the device area, before or after irradiation along the planned division line, and further, the first laser beam is focused on the device area of the wafer. The height position of the surface is irradiated along the planned division line.