Wafer processing method

Abstract

To provide a method for processing a wafer which can suppress breakage of a device in association with formation of a thinned wafer and also can suppress reduction of the number of chips which can be manufactured from the wafer.SOLUTION: A method for processing a wafer by forming a separation layer in the wafer and dividing the wafer by the separation layer includes: a separation layer formation step of forming a first separation layer along a side surface of a circular truncated cone and forming a second separation layer along the upper surface of a circular truncated cone; a division step of dividing a wafer to form a thinned wafer having a first separation surface exposed by being divided by the first separation layer and a second separation surface exposed by being divided by the second separation layer; a dry etching step of performing dry etching on the first separation surface; a grinding step of grinding the second separation surface; and a polishing step of polishing the first separation surface and the second separation surface.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a wafer processing method, comprising: forming a peeling layer inside a wafer by irradiating a laser beam having a wavelength that transmits through the wafer onto the wafer, wherein a plurality of devices are formed on a surface side of the wafer and an outer peripheral region of the wafer is chamfered; and then dividing the wafer at the peeling layer by applying an external force to the wafer along the thickness direction of the wafer. Background Art

[0002] Chips of devices such as integrated circuits (ICs) or memories are generally manufactured using disk-shaped wafers made of semiconductor materials such as silicon (Si). Specifically, the wafer is partitioned by division lines set in a grid pattern, and a device is formed on the surface side of each of a plurality of regions.

[0003] Chips are then manufactured by dividing the wafer along the predetermined dividing lines. Furthermore, through-silicon vias (TSVs) are sometimes provided in a wafer for the purpose of achieving high integration of packages including a plurality of chips, and the like. In this package, for example, electrodes included in different chips can be electrically connected via TSVs.

[0004] TSVs are provided on a wafer in the following order, for example. First, grooves are formed on the surface side of the wafer. Next, TSVs are provided in the grooves. Next, the surface side of the wafer is bonded to a support wafer. Then, the back surface of the wafer is ground until the TSVs are exposed at the back surface of the wafer.

[0005] Here, for the purpose of preventing cracking and the like, the outer peripheral region of a wafer is often chamfered. When the back surface of a wafer with a chamfered outer peripheral region is ground until the thickness of the wafer becomes half or less of the original thickness, the back surface side of the outer peripheral region becomes shaped like a knife edge.

[0006] In this case, stress concentrates on the back side of the outer peripheral region during wafer grinding, making the wafer prone to cracking and potentially reducing the yield of chips obtained from the wafer. Therefore, it has been proposed to remove a portion of the front side of the outer peripheral region (so-called edge trimming) prior to grinding the back side of the wafer (see, for example, Patent Documents 1 and 2).

[0007] Furthermore, if grinding is performed on the back surface of the wafer until the TSV is exposed, the amount of material removed from the wafer increases, and the amount of wear on the grinding wheel required to grind this wafer increases. In this case, the cost of the chips or packages manufactured using this wafer may increase, and the processing time may also be prolonged.

[0008] In light of these points, a method has been proposed for splitting a wafer using a laser beam with a wavelength that penetrates the wafer (see, for example, Patent Document 3). In this method, first, a laser beam is irradiated in a ring shape onto a region of the wafer where no diffuse reflection occurs when the laser beam is irradiated (in short, the region inside the beveled outer peripheral region), with the focal point positioned on the surface side of the wafer.

[0009] This creates a delamination layer (cylindrical delamination layer) that follows the side of a cylinder, with its bottom surface located on the wafer surface and its top surface located inside the wafer. Next, a laser beam is irradiated onto the region inside this cylindrical delamination layer, with the focal point positioned on the top surface of the cylinder. This creates a delamination layer (disk-shaped delamination layer) that follows the top surface of the cylinder.

[0010] Next, an external force is applied to the wafer, causing it to split in a cylindrical delamination layer and a disc-shaped delamination layer. This forms a thinned wafer having a side surface exposed by the splitting in the cylindrical delamination layer and an upper surface exposed by the splitting in the disc-shaped delamination layer. [Prior art documents] [Patent Documents]

[0011] [Patent Document 1] Japanese Patent Publication No. 2007-158239 [Patent Document 2] Japanese Patent Publication No. 2020-57709 [Patent Document 3] Japanese Patent Publication No. 2020-136442 [Overview of the project] [Problems that the invention aims to solve]

[0012] As described above, when dividing a wafer with a cylindrical or disc-shaped delamination layer, it is necessary to separate the outer peripheral region of the wafer from the device region along the radial direction of the wafer, and to separate the back side of this device region from the front side along the thickness direction of the wafer.

[0013] Here, the direction of the external force required to separate the outer peripheral region of the wafer from the device region is perpendicular to the direction of the external force required to separate the back side of the device region from the front side. Therefore, it is not always easy to simultaneously separate the wafer in cylindrical and disc-shaped delamination layers by applying an external force along a specific direction to the wafer.

[0014] For example, when a wafer is split by applying an external force along the thickness direction of the wafer, there is a risk that the side surface of the thinned wafer formed by the splitting of the wafer and the inner surface of the remaining portion of the wafer (including the outer peripheral region) may come into contact with each other. In this case, there is a risk that cracks may be formed due to the contact between the two.

[0015] Furthermore, as mentioned above, the cylindrical delamination layer is formed in a region of the wafer where diffuse reflection does not occur when a laser beam is irradiated (specifically, the region inside the chamfered outer peripheral region). Therefore, if the wafer is split in this cylindrical delamination layer, cracks may propagate toward the device formed on the surface side of the thinned wafer, potentially damaging the device.

[0016] Furthermore, damage to such devices can be prevented by not forming devices in the area near the side surface of the thinned wafer. However, in this case, the number of chips that can be manufactured from the wafer may decrease.

[0017] In view of the above, the object of the present invention is to provide a wafer processing method that can suppress device damage associated with the formation of a thinned wafer and suppress the reduction in the number of chips that can be manufactured from the wafer. [Means for solving the problem]

[0018] According to the present invention, a wafer processing method is provided, wherein a wafer having a plurality of devices formed on its surface and a chamfered outer peripheral region is irradiated with a laser beam of a wavelength that penetrates the wafer to form a delamination layer inside the wafer, and then the wafer is divided in the delamination layer by applying an external force to the wafer along the thickness direction of the wafer, wherein the wafer has a first bottom surface located on the surface side of the wafer, a side surface having one end that overlaps with the outer peripheral of the first bottom surface and that slopes so as it moves away from the surface side of the wafer, and a second bottom surface having an outer peripheral that overlaps with the other end of the side surface and that is parallel to the first bottom surface, and the first delamination layer and the second A wafer processing method is provided, comprising: a delamination layer formation step of forming a second delamination layer along the bottom surface; a division step of dividing the wafer after the delamination layer formation step so as to form a thinned wafer having a first delamination surface exposed by division in the first delamination layer and a second delamination surface exposed by division in the second delamination layer; a dry etching step of performing dry etching on at least the first delamination surface after the division step; a grinding step of performing grinding on the second delamination surface after the dry etching step; and a polishing step of polishing the first delamination surface and the second delamination surface after the grinding step.

[0019] The wafer processing method of the present invention preferably further comprises a bonding step of bonding the surface side of the wafer to the surface side of a second wafer different from the wafer, before performing the splitting step. Furthermore, dry etching of the first and second peeled surfaces is not required between the grinding step and the polishing step. Moreover, in the polishing step, the first and second peeled surfaces may be polished simultaneously.

[0020] Furthermore, in the peeling layer forming step, it is preferable to form one of the first peeling layer and the second peeling layer and then form the other. Alternatively, the wafer includes a first region and a second region each extending along a predetermined direction, the first peeling layer includes a pair of first inclined portions formed in the first region and a pair of second inclined portions formed in the second region, and the second peeling layer includes a first linear portion formed in the first region and located between the pair of first inclined portions in the predetermined direction, and a second linear portion formed in the second region and located between the pair of second inclined portions in the predetermined direction, and in the peeling layer forming step, it is preferable to form the pair of first inclined portions and the first linear portion, and then form the pair of second inclined portions and the second linear portion.

[0021] Also, in the dry etching step, it is preferable to perform dry etching on the second peeling surface together with the first peeling surface. Alternatively, in the dry etching step, it is preferable to perform dry etching on the first peeling surface in a state where a mask is provided on the second peeling surface.

Effects of the Invention

[0022] In the present invention, the first peeling layer extending along the side surface of a truncated cone and the second peeling layer extending along the second bottom surface are formed inside the wafer, wherein the truncated cone is defined by: a first bottom surface located on the surface side of the wafer; a side surface having one end overlapping the outer circumference of the first bottom surface and inclined such that the distance from the outer circumferential region increases as the distance from the surface side of the wafer increases; and a second bottom surface having an outer circumference overlapping the other end of the side surface and parallel to the first bottom surface.

[0023] Then, in the present invention, the wafer is divided such that a thinned wafer is formed, the thinned wafer having a first peeling surface (the side surface) exposed by division along the first peeling layer, and a second peeling surface (the top surface) exposed by division along the second peeling layer.

[0024] Here, in this wafer, the first exfoliation layer is formed along the side surface of the truncated cone. In this case, when dividing the wafer, the probability that the side surface of the thinned wafer and the inner side surface of the remaining portion of the wafer (the portion including the outer peripheral region) contact each other is low. Therefore, in the present invention, the occurrence of cracks caused by the contact between the two can be suppressed.

[0025] Furthermore, in this wafer, cracks easily propagate from the first exfoliation layer in a direction along the side surface of the truncated cone. In this case, the probability that cracks propagate toward devices formed on the front surface side of the thinned wafer is reduced. Therefore, in the present invention, damage to devices caused by the formation of the thinned wafer can be suppressed, and a decrease in the number of chips that can be manufactured from the wafer can be suppressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] [Figure 1] Fig. 1(A) is a top view schematically showing an example of a wafer, and Fig. 1(B) is a cross-sectional view schematically showing an example of a wafer. [Figure 2] Fig. 2 is a flowchart schematically showing an example of a wafer processing method. [Figure 3] Fig. 3 is a cross-sectional view schematically showing a state where the front surface side of a wafer is bonded to the front surface side of a support wafer. [Figure 4] Fig. 4 is a perspective view schematically showing an example of a laser processing apparatus. [Figure 5] Fig. 5 is a diagram schematically showing how a laser beam travels in a laser beam irradiation unit. [Figure 6] Fig. 6 is a partial cross-sectional front view schematically showing a state where a laser beam is irradiated to a region inside the outer peripheral region of a wafer. [Figure 7] Fig. 7 is a partial cross-sectional front view schematically showing a state where a laser beam is irradiated to a region inside the first exfoliation layer of a wafer. [Figure 8] Each of Fig. 8(A) and Fig. 8(B) is a partial cross-sectional front view schematically showing a state where a wafer is divided. [Figure 9] Figure 9 schematically shows the process of dry etching on the side and top surfaces of a thinned wafer. [Figure 10] Figure 10 is a schematic perspective view showing an example of a processing apparatus. [Figure 11] Figure 11 is a schematic partial cross-sectional side view showing the grinding process on the upper surface of a thinned wafer. [Figure 12] Figure 12 is a schematic cross-sectional side view showing the polishing process applied to the side and top surfaces of a thinned wafer. [Modes for carrying out the invention]

[0027] Embodiments of the present invention will be described with reference to the attached drawings. Figure 1(A) is a schematic top view showing an example of a wafer, and Figure 1(B) is a schematic cross-sectional view showing an example of a wafer. In Figure 1(A), the crystal orientation of the single-crystal material (in this case, single-crystal silicon) constituting this wafer is also shown.

[0028] The wafer 11 shown in Figures 1(A) and 1(B) consists of cylindrical single-crystal silicon in which a specific crystal plane (referred to here as crystal plane (100) for convenience) included in the crystal plane {100} is exposed on the front surface 11a and the back surface 11b, respectively. In other words, this wafer 11 consists of cylindrical single-crystal silicon in which the perpendiculars (crystal axes) of the front surface 11a and the back surface 11b are aligned with the crystal orientation

[0100] .

[0029] Although the wafer 11 is manufactured so that crystal planes (100) are exposed on both the front surface 11a and the back surface 11b, due to processing errors during manufacturing, the front surface 11a and the back surface 11b may be slightly inclined from the crystal planes (100). Specifically, the front surface 11a and the back surface 11b of the wafer 11 may be surfaces in which the acute angle formed between them and the crystal planes (100) is 1° or less.

[0030] In other words, the crystal axis of wafer 11 may be aligned in a direction where the acute angle formed with the crystal orientation

[0100] is 1° or less. Also, a notch 13 is formed on the side surface 11c of wafer 11, and the crystal orientation as seen from this notch 13 <110> The center of wafer 11 is located on a specific crystal plane (for convenience, this will be referred to as crystal orientation

[0011] ) contained within it.

[0031] Furthermore, the wafer 11 is divided into multiple regions by multiple intersecting division lines, and a device 15 such as an IC or memory is formed on the surface 11a side of each region. In addition, grooves for providing TSVs may be formed on the surface 11a side of the wafer 11.

[0032] Furthermore, the outer periphery of the wafer 11 is chamfered. That is, the side surface 11c of the wafer 11 is curved so as to be convex outwards. Note that the device 15 is not formed in the outer periphery of the wafer 11. In other words, the area of ​​the wafer 11 where the device 15 is formed (the device area) is surrounded by its outer periphery.

[0033] There are no restrictions on the material, shape, structure, or size of the wafer 11. The wafer 11 may be made of a semiconductor material other than silicon (for example, silicon carbide (SiC) or gallium nitride (GaN)). Similarly, there are no restrictions on the type, quantity, shape, structure, size, or arrangement of the devices 15.

[0034] Figure 2 is a schematic flowchart illustrating an example of a wafer processing method. In this method, first, the surface 11a side of wafer 11 is bonded to the surface side of a support wafer (second wafer) (bonding step S1). Figure 3 is a schematic cross-sectional view showing the bonding of the surface 11a side of wafer 11 to the surface side of the support wafer.

[0035] The support wafer 17, which is bonded to the wafer 11, has a shape similar to that of the wafer 11. Also, similar to the wafer 11, multiple devices may be formed on the surface 17a of the support wafer 17. Furthermore, an adhesive 19, such as an acrylic adhesive or an epoxy adhesive, is provided on the surface 17a of the support wafer 17.

[0036] Then, in bonding step S1, with the back surface 17b side of the support wafer 17 supported, the surface 11a of wafer 11 is pressed against the surface 17a of support wafer 17 via adhesive 19. This forms a bonded wafer in which the surface 11a side of wafer 11 is bonded to the surface 17a side of support wafer 17.

[0037] After performing the bonding step S1, a delamination layer is formed inside the wafer 11 by irradiating the wafer 11, which is positioned with its back surface 11b facing upward, with a laser beam of a wavelength that penetrates the wafer 11 from above. Specifically, a first delamination layer along the side surface of the frustum and a second delamination layer along the top surface are formed inside the wafer 11 (delamination layer formation step S2).

[0038] This frustum of a cone is a hypothetical structure that includes a bottom surface (first base) located on the surface 11a side of the wafer 11, a bottom end (one end) that coincides with the outer circumference of the bottom surface, a side surface that slopes away from the outer circumference region as it moves away from the surface 11a side of the wafer 11, and an outer circumference that coincides with the top end (other end) of the side surface, and an upper surface (second base) that is parallel to the bottom surface. In this frustum of a cone, the angle between the first base and the side surface is, for example, between 1° and 80°.

[0039] Figure 4 is a schematic perspective view showing an example of a laser processing apparatus used in the peel layer formation step S2. Note that the X-axis direction (left-right direction) and Y-axis direction (front-back direction) shown in Figure 4 are mutually orthogonal directions on the horizontal plane, and the Z-axis direction (up-down direction) is a direction perpendicular to the X-axis direction and the Y-axis direction, respectively (vertical direction).

[0040] The laser processing apparatus 2 shown in Figure 4 has a base 4 that supports each component. A horizontal movement mechanism 6 is positioned on the upper surface of this base 4. The horizontal movement mechanism 6 has a pair of Y-axis guide rails 8 that are fixed to the upper surface of the base 4 and extend along the Y-axis direction.

[0041] A Y-axis moving plate 10 is connected to the upper surface of a pair of Y-axis guide rails 8 in a manner that allows it to slide along the pair of Y-axis guide rails 8. A screw shaft 12 extending along the Y-axis direction is positioned between the pair of Y-axis guide rails 8. A motor 14 for rotating the screw shaft 12 is connected to the front end (one end) of this screw shaft 12.

[0042] Furthermore, a nut (not shown) is provided on the surface of the screw shaft 12, where a helical groove is formed, to accommodate a large number of balls that roll on the surface of the rotating screw shaft 12, thus forming a ball screw. That is, as the screw shaft 12 rotates, the large number of balls circulate within the nut, causing the nut to move along the Y-axis direction.

[0043] Furthermore, this nut is fixed to the underside of the Y-axis moving plate 10. Therefore, when the screw shaft 12 is rotated by the motor 14, the Y-axis moving plate 10 moves along the Y-axis direction together with the nut. In addition, a pair of X-axis guide rails 16 extending along the X-axis direction are fixed to the upper surface of the Y-axis moving plate 10.

[0044] An X-axis moving plate 18 is connected to the upper surface of a pair of X-axis guide rails 16 in a manner that allows it to slide along the pair of X-axis guide rails 16. A screw shaft 20 extending along the X-axis direction is positioned between the pair of X-axis guide rails 16. A motor 22 for rotating the screw shaft 20 is connected to one end of this screw shaft 20.

[0045] Furthermore, a nut (not shown) is provided on the surface of the screw shaft 20, where a helical groove is formed, to accommodate a large number of balls that roll on the surface of the rotating screw shaft 20, thus forming a ball screw. That is, as the screw shaft 20 rotates, the large number of balls circulate within the nut, causing the nut to move along the X-axis direction.

[0046] Furthermore, this nut is fixed to the underside of the X-axis moving plate 18. Therefore, when the motor 22 rotates the screw shaft 20, the X-axis moving plate 18 moves along the X-axis direction together with the nut.

[0047] A cylindrical table base 24 is positioned on the upper side of the X-axis moving plate 18. A holding table 26 for holding the bonded wafers described above is positioned on the upper part of this table base 24. This holding table 26 has, for example, a circular upper surface (holding surface) parallel to the X-axis and Y-axis directions, and the porous plate 26a is exposed on this holding surface.

[0048] Furthermore, a rotational drive source (not shown), such as a motor, is connected to the lower part of the table base 24. When this rotational drive source is activated, the holding table 26 rotates around a straight line passing through the center of the holding surface and parallel to the Z-axis direction as its axis of rotation. Also, when the horizontal movement mechanism 6 described above is activated, the holding table 26 moves along the X-axis direction and / or the Y-axis direction.

[0049] Furthermore, the porous plate 26a is in communication with a suction source (not shown), such as an ejector, via a flow path or the like provided inside the holding table 26. When this suction source is activated, a suction force acts on the space near the holding surface of the holding table 26. Therefore, for example, when the aforementioned bonded wafer is placed on the holding table 26 and the suction source is activated, the bonded wafer can be held by the holding table 26.

[0050] Furthermore, a support structure 30 is provided behind the horizontal movement mechanism 6, having sides that are generally parallel to the Y-axis and Z-axis directions. A vertical movement mechanism 32 is positioned on the side of this support structure 30. The vertical movement mechanism 32 has a pair of Z-axis guide rails 34 that are fixed to the side of the support structure 30 and extend along the Z-axis direction.

[0051] A Z-axis moving plate 36 is connected to the surface side of a pair of Z-axis guide rails 34 in a manner that allows it to slide along the pair of Z-axis guide rails 34. A screw shaft (not shown) extending along the Z-axis direction is positioned between the pair of Z-axis guide rails 34. A motor 38 for rotating the screw shaft is connected to the upper end (one end) of this screw shaft.

[0052] Furthermore, a nut (not shown) is provided on the surface of the screw shaft where a helical groove is formed, which houses a number of balls that roll on the surface of the rotating screw shaft, thus forming a ball screw. In other words, as this screw shaft rotates, the number of balls circulate within the nut, causing the nut to move along the Z-axis direction.

[0053] Furthermore, this nut is fixed to the back side of the Z-axis moving plate 36. Therefore, when the motor 38 rotates the screw shaft located between the pair of Z-axis guide rails 34, the Z-axis moving plate 36 moves along the Z-axis direction together with the nut.

[0054] A support 40 is fixed to the surface side of the Z-axis moving plate 36. This support 40 supports a part of the laser beam irradiation unit 42. Figure 5 is a schematic diagram showing how the laser beam LB propagates in the laser beam irradiation unit 42. In Figure 5, some of the components of the laser beam irradiation unit 42 are shown as functional blocks.

[0055] The laser beam irradiation unit 42 has a laser oscillator 44 fixed to the base 4. This laser oscillator 44 has, for example, Nd:YAG as the laser medium and emits a laser beam LB with a wavelength (e.g., 1064 nm or 1342 nm) that penetrates the wafer 11. This laser beam LB is, for example, a pulsed laser beam with a frequency of 60 kHz.

[0056] The laser beam LB is then supplied to the spatial light modulator 48 after its output is adjusted in the attenuator 46. In the spatial light modulator 48, the laser beam LB is split. For example, the spatial light modulator 48 splits the laser beam LB, which has been adjusted in the attenuator 46, so that the laser beam LB emitted from the irradiation head 52 (described later) forms multiple focal points whose positions (coordinates) in a plane parallel to the X-axis and Y-axis (XY coordinate plane) and / or positions (heights) in the Z-axis direction are different from each other.

[0057] Furthermore, the laser beam LB, which is branched in the spatial light modulator 48, is reflected by the mirror 50 and guided to the irradiation head 52. The irradiation head 52 houses a focusing lens (not shown) for focusing the laser beam LB. The laser beam LB, focused by this focusing lens, is then emitted towards the holding surface side of the holding table 26.

[0058] As shown in Figure 4, the irradiation head 52 is located at the front end of the cylindrical housing 54. A support 40 is fixed to the rear side of the housing 54. Furthermore, an imaging unit 56 is fixed to the front side of the housing 54.

[0059] This imaging unit 56 includes, for example, a light source such as an LED (Light Emitting Diode), an objective lens, and an image sensor such as a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor.

[0060] When the vertical movement mechanism 32 described above is activated, the laser beam irradiation unit 42 and the imaging unit 56 move along the Z-axis direction. Furthermore, a cover (not shown) is provided on the base 4 to cover the above-mentioned components. A touch panel 58 is also located on the front of this cover.

[0061] This touch panel 58 is composed of, for example, an input device such as a capacitive or resistive touch sensor and a display device such as a liquid crystal display or an organic EL (Electro-Luminescence) display, and functions as a user interface.

[0062] In the laser processing apparatus 2, for example, the delamination layer formation step S2 is carried out in the following order. Specifically, first, the bonded wafer is placed on the holding table 26 so that the center of the back surface 17b of the support wafer 17 of the bonded wafer described above coincides with the center of the holding surface of the holding table 26.

[0063] Next, a suction source communicating with the porous plate 26a is activated so that the bonded wafer is held by the holding table 26. Then, the imaging unit 56 images the back surface 11b of the wafer 11 of the bonded wafer to form an image.

[0064] Next, referring to this image, the horizontal movement mechanism 6 is operated so that the irradiation head 52 of the laser beam irradiation unit 42 is positioned directly above an area slightly inside the outer peripheral region of the wafer 11.

[0065] Alternatively, referring to this image, the rotation drive source connected to the lower part of the table base 24 may be operated so that the holding table 26 is rotated so that the crystal orientation

[0010] of the single crystal silicon constituting the wafer 11 is parallel to the X-axis direction and the crystal orientation

[0001] is parallel to the Y-axis direction.

[0066] Next, a delamination layer (first delamination layer) is formed in a region inside the outer peripheral region of the wafer 11, following the side surface of the frustum described above. Specifically, the laser beam irradiation unit 42 is first operated to irradiate the wafer 11 with the branched laser beam LB. At this time, the laser beam LB is branched such that the multiple focal points are arranged further away from the outer peripheral region as they move away from the surface 11a of the wafer 11.

[0067] Figure 6 is a schematic partial cross-sectional front view showing how the laser beam LB is irradiated onto a region inside the outer periphery of the wafer 11. This laser beam LB forms multiple (for example, 8) focal points in the radial direction (perpendicular to the Z-axis direction) and thickness direction (Z-axis direction) of the wafer 11, with the positions of adjacent focal points shifted by 10 μm inside the wafer 11.

[0068] In this case, modified portions 21 are formed inside the wafer 11, with the crystal structure of the material constituting the wafer 11 being disordered, centered around each of the multiple focal points. That is, multiple modified portions 21 are formed such that they are aligned linearly along the radial direction of the wafer 11 in a plan view, and the acute angle between this line and the surface 11a of the wafer 11 is 45°.

[0069] Furthermore, the acute angle formed between the linearly arranged modified portions 21 and the surface 11a of the wafer 11 is not limited to 45°. That is, the laser beam LB may be irradiated onto the wafer 11 such that multiple focal points are formed, where the radial spacing of adjacent focal points differs from the thickness spacing of the wafer 11.

[0070] However, this angle is preferably 1° or more. That is, the angle between the lower surface and the side surface of the frustum of the cone is preferably 1° or more. This prevents damage to the device 15 formed on the surface 11a side of the wafer 11 by the laser beam LB when forming a delamination layer along the upper surface of the frustum of the cone.

[0071] Furthermore, cracks 23 extend from each of the multiple modified sections 21, connecting adjacent pairs of modified sections 21. As a result, a delamination layer is formed inside the wafer 11, which includes the multiple modified sections 21 and the cracks 23 extending from each of the multiple modified sections 21.

[0072] Next, while the laser beam irradiation unit 42 is still operating, the rotation drive source connected to the bottom of the table base 24 is operated so that the holding table 26 that holds the bonded wafer rotates at least once. As a result, a delamination layer (first delamination layer) that follows the side surface of the frustum is formed in the region inside the outer peripheral region of the wafer 11.

[0073] Next, a delamination layer (second delamination layer) is formed in a region of the wafer 11 that is inside the first delamination layer, following the upper surface (second base surface) of the frustum of the cone. Specifically, first, the horizontal movement mechanism 6 is operated so that the irradiation head 52 of the laser beam irradiation unit 42 is positioned directly above a region slightly inside the first delamination layer of the wafer 11.

[0074] At this time, if necessary, the rotation drive source connected to the lower part of the table base 24 may be operated to rotate the holding table 26 so that the crystal orientation

[0010] of the single crystal silicon constituting the wafer 11 becomes parallel to the X-axis direction and the crystal orientation

[0001] becomes parallel to the Y-axis direction. This rotation of the holding table 26 is performed, for example, by referring to the image of the back surface 11b of the wafer 11 formed by imaging by the imaging unit 56.

[0075] Next, the laser beam irradiation unit 42 is operated to irradiate with the branched laser beam LB. At this time, the laser beam LB is branched so that all of the multiple focusing points are located on the upper surface (second base) of the frustum of the cone.

[0076] Figure 7 is a schematic partial cross-sectional front view showing the irradiation of the laser beam LB to the region of wafer 11 inside the first exfoliation layer. This laser beam LB is branched to form multiple (e.g., 2 to 10) focal points arranged at equal intervals (e.g., 5 μm to 15 μm) in the Y-axis direction (crystal orientation

[0001] ).

[0077] In this case, as described above, multiple modified portions 21 and cracks 23 extending from each of the multiple modified portions 21 are formed inside the wafer 11. Next, while the laser beam LB is still irradiating the wafer 11, the horizontal movement mechanism 6 is operated to move the holding table 26 that holds the bonded wafer along the X-axis.

[0078] As a result, a delamination layer is formed in a linear region along the X-axis direction (crystal orientation

[0010] ) that is located inside the first delamination layer of wafer 11. Furthermore, when a delamination layer is formed by irradiating a linear region along the crystal orientation

[0010] with a laser beam LB that forms multiple focal points aligned along the crystal orientation

[0001] , the delamination layer can be made wider and thinner.

[0079] The following explains this point in detail. First, single-crystal silicon is generally most easily cleaved at a specific crystal plane within crystal plane {111}, and second most easily cleaved at a specific crystal plane within crystal plane {110}. Therefore, for example, the crystal orientation of the single-crystal silicon constituting wafer 11 <110> When a modified portion is formed along a specific crystal orientation (for example, crystal orientation

[0011] ) contained within the material, many cracks are generated from this modified portion that extend along the specific crystal plane contained within the crystal plane {111}.

[0080] On the other hand, the crystal orientation of single-crystal silicon <100> When multiple modified regions are formed in a region along a specific crystal orientation within a given material, such that they are aligned in a direction perpendicular to the direction in which the region extends when viewed from a plan perspective, many cracks are generated from each of these modified regions, extending along the crystal plane {N10} (where N is a natural number less than or equal to 10) that is parallel to the direction in which the region extends.

[0081] For example, as described above, if multiple modified portions 21 are formed in a region along the crystal orientation

[0010] so as to be aligned along the crystal orientation

[0001] , then many cracks will extend from each of these multiple modified portions 21 along the crystal plane {N10} that is parallel to the crystal orientation

[0010] .

[0082] Specifically, when multiple modified regions 21 are formed in this manner, cracks tend to propagate more easily in the following crystal planes.

number

number

[0083] Furthermore, the angle that the crystal planes (100) exposed on the front surface 11a and back surface 11b of the wafer 11 make with the crystal planes parallel to the crystal orientation

[0010] within the crystal plane {N10} is 45° or less. On the other hand, the angle that the crystal planes (100) make with a specific crystal plane included in the crystal plane {111} is approximately 54.7°.

[0084] Therefore, when a delamination layer is formed as described above, the delamination layer can be made wider and thinner compared to the case where multiple modified portions are formed in a region along the crystal orientation

[0011] of the single-crystal silicon, such that they are aligned in a direction perpendicular to the direction in which the region extends when viewed in plan.

[0085] Next, the horizontal movement mechanism 6 is operated to move the holding table 26 that holds the bonded wafers along the Y-axis. Then, the laser beam irradiation unit 42 and the horizontal movement mechanism 6 are operated so that a delamination layer is formed in a linear region parallel to the linear region where a delamination layer has already been formed.

[0086] Furthermore, the same operation is repeated until a delamination layer is formed in the entire area inside the first delamination layer of wafer 11. As a result, a delamination layer (second delamination layer) is formed in the area inside the first delamination layer of wafer 11, following the upper surface (second base surface) of the frustum of the cone. With this, the delamination layer formation step S2 is completed.

[0087] In step S2, the second delamination layer may be formed before the first delamination layer. That is, in step S2, the second delamination layer may be formed by sequentially irradiating a plurality of linear regions contained in the wafer 11 with the laser beam LB, and then the first delamination layer may be formed by irradiating an annular region located outside these regions with the laser beam LB.

[0088] Alternatively, in step S2, the first and second peeling layers may be formed simultaneously. Specifically, the first and second peeling layers may be formed as follows.

[0089] First, the irradiation head 52 of the laser beam irradiation unit 42 is positioned directly above a region slightly inside the outer peripheral region of the wafer 11. Next, the holding table 26 that holds the bonded wafer is rotated so that the crystal orientation

[0010] of the single crystal silicon constituting the wafer 11 is parallel to the X-axis direction and the crystal orientation

[0001] is parallel to the Y-axis direction.

[0090] Next, the laser beam irradiation unit 42 is operated to irradiate with the branched laser beam LB. At this time, the laser beam LB is branched such that, for example, all of the multiple focal points are located near one end in the X-axis direction on the lower surface of the frustum of the cone.

[0091] Next, the holding table 26 that holds the bonded wafer is moved along the X-axis while raising the position of the multiple focal points until all of the multiple focal points are positioned near one end of the upper surface of the frustum in the X-axis direction. As a result, a part of the first delamination layer (first inclined portion) is formed inside a linear region (first region) along the crystal orientation

[0010] contained in the wafer 11.

[0092] Next, the holding table 26 that holds the bonded wafer is moved along the X-axis direction, keeping all of the multiple focal points at the same height as the top surface of the frustum of the cone, until all of the multiple focal points are positioned near the other end in the X-axis direction of the top surface of the frustum of the cone. As a result, a portion of the second delamination layer (the first straight section) is formed inside the first region of the wafer 11.

[0093] Next, the holding table 26 that holds the bonded wafer is moved along the X-axis while lowering the positions of the multiple focal points until all of the multiple focal points are positioned near the other end in the X-axis direction on the lower surface of the frustum of the cone. As a result, another part of the first delamination layer (the first inclined portion) is formed inside the first region of the wafer 11.

[0094] In other words, by irradiating the wafer 11 with the laser beam LB as described above, a pair of first inclined portions and a first straight portion located between the pair of first inclined portions in the crystal orientation

[0010] are formed inside the first region of the wafer 11. Next, the holding table 26 that holds the bonded wafers is moved along the Y-axis.

[0095] Next, the laser beam irradiation unit 42 and the horizontal movement mechanism 6 are operated so that another pair of inclined sections (second inclined section) and another straight section (second straight section) are formed in a straight region (second region) parallel to the first region where a pair of inclined sections and a first straight section have already been formed. By repeating the same operation, the first and second peeling layers are formed simultaneously and in parallel.

[0096] After performing the delamination layer formation step S2, the wafer 11 is divided (dividing step S3) so that a thinned wafer is formed having a side surface (first delamination surface) that is exposed by being divided in the first delamination layer and an upper surface (second delamination surface) that is exposed by being divided in the second delamination layer.

[0097] Figures 8(A) and 8(B) are schematic cross-sectional side views illustrating the process of dividing a wafer 11 on which a first and second delamination layer have been formed. This dividing step S3 is carried out, for example, in the dividing apparatus 60 shown in Figures 8(A) and 8(B).

[0098] The splitting device 60 has a holding table 62 that holds the bonded wafer including the wafer 11. The holding table 62 has a circular top surface (holding surface), on which a porous plate (not shown) is exposed. Furthermore, this porous plate is in communication with a suction source (not shown), such as an ejector, via a flow path or the like provided inside the holding table 62.

[0099] When this suction source is activated, a suction force acts on the space near the holding surface of the holding table 62. Therefore, for example, if the suction source is activated while the bonded wafer described above is placed on the holding table 62, the bonded wafer can be held by the holding table 62.

[0100] Furthermore, a splitting unit 64 is provided above the holding table 62. This splitting unit 64 has a cylindrical support member 66. A ball screw type lifting mechanism (not shown), for example, is connected to the upper part of this support member 66, and the splitting unit 64 moves up and down by operating this lifting mechanism.

[0101] Furthermore, the lower end of the support member 66 is fixed to the center of the upper part of the disc-shaped support base 68. On the lower side of the outer peripheral region of the support base 68, a plurality of movable members 70 are provided at intervals of approximately equal angles along the circumferential direction of the support base 68. Each of these movable members 70 has a plate-shaped vertical portion 70a that extends downward.

[0102] The upper end of the vertical section 70a is connected to an actuator such as an air cylinder built into the support base 68, and by operating this actuator, the movable member 70 moves along the radial direction of the support base 68. In addition, the inner surface of the lower end of the vertical section 70a is provided with a plate-shaped claw portion 70b that extends toward the center of the support base 68 and whose thickness decreases as it approaches the center of the support base 68.

[0103] In the splitting apparatus 60, for example, the splitting step S3 is carried out in the following order. Specifically, first, the bonded wafer is placed on the holding table 62 such that the center of the back surface 17b of the support wafer 17 included in the bonded wafer, which includes the wafer 11 on which the first and second release layers are formed, coincides with the center of the holding surface of the holding table 62.

[0104] Next, a suction source communicating with the porous plate exposed on the holding surface is activated so that the bonded wafer is held by the holding table 62. Then, actuators are activated to position each of the multiple movable members 70 radially outward from the support base 68.

[0105] Next, the lifting mechanism is operated to position the tips of the claw portions 70b of each of the multiple movable members 70 at a height corresponding to the adhesive 19 of the bonding wafer. Next, the actuator is operated to bring the claw portions 70b into contact with the bonding wafer. Next, the lifting mechanism is operated to raise the claw portions 70b (see Figure 8(A)).

[0106] This applies an upward external force to the outer peripheral region of the wafer 11, that is, an external force along the thickness direction of the wafer 11. As a result, the cracks 23 contained in the first and second delamination layers extend further, causing the wafer 11 to split, separating the region where the multiple devices 15 are formed from the outer peripheral region, and also separating the surface 11a side from the back surface 11b side (see Figure 8(B)).

[0107] In other words, a thinned wafer 25 is formed having a side surface (first delamination surface) 25a that is exposed by separation in the first delamination layer, and an upper surface (second delamination surface) 25b that is exposed by separation in the second delamination layer. The side surface 25a has irregularities that reflect the shape of the cracks 23 contained in the first delamination layer, and the upper surface 25b also has irregularities that reflect the shape of the cracks 23 contained in the second delamination layer. Thus, the splitting step S3 is completed.

[0108] After performing the splitting step S3, dry etching is performed on the side surface 25a and top surface 25b of the thinned wafer 25 (dry etching step S4). Figure 9 is a schematic diagram showing the dry etching process on the side surface 25a and top surface 25b of the thinned wafer 25.

[0109] This dry etching step S4 is carried out, for example, in the etching apparatus 72 shown in Figure 9. In Figure 9, some of the components of the etching apparatus 72 are shown as functional blocks. The etching apparatus 72 has a chamber 74 made of a conductive material and grounded.

[0110] The chamber 74 has an inlet / outlet 74a for loading a bonded wafer, including the thinned wafer 25, into and out of the chamber. The inlet / outlet 74a is provided with a gate valve 76 that can separate or connect the internal space of the chamber 74 to the external space. The chamber 74 also has an exhaust port 74b for exhausting its internal space.

[0111] This exhaust port 74b is connected to an exhaust device 80 such as a vacuum pump via piping 78, etc. A support member 82 is provided on the inner surface of the chamber 74, and this support member 82 supports the table 84. An electrostatic chuck (not shown) is provided on the top of the table 84.

[0112] Furthermore, a disc-shaped opening is formed in the chamber 74 at a position opposite the upper surface of the table 84, and a gas ejection head 92 is provided in this opening, supported by the chamber 74 via a bearing 90. This gas ejection head 92 is made of a conductive material and is connected to a high-frequency power supply 96 via a matching unit 94.

[0113] Furthermore, a cavity (gas diffusion space) 92a is formed inside the gas ejection head 92. In addition, a plurality of gas outlets 92b are formed in the inner part of the gas ejection head 92 (for example, the lower part) that connect the gas diffusion space 92a with the internal space of the chamber 74. In addition, two gas supply ports 92c and 92d are formed in the outer part of the gas ejection head 92 (for example, the upper part) for supplying a predetermined gas to the gas diffusion space 92a.

[0114] Furthermore, the gas supply port 92c is connected via piping 98a, etc., to a gas supply source 100a that supplies, for example, a sulfur fluoride-based gas such as SF6. Also, the gas supply port 92d is connected via piping 98b, etc., to a gas supply source 100b that supplies, for example, an inert gas such as Ar.

[0115] In the etching apparatus 72, for example, the dry etching step S4 is carried out in the following order. Specifically, first, with the gate valve 76 connecting the internal and external spaces of the chamber 74, the bonded wafers are brought onto the table 84 with the thinned wafer 25 facing upwards.

[0116] Next, the bonded wafer is held by the electrostatic chuck of the table 84. Then, the internal space of the chamber 74 is evacuated by the exhaust device 80 to create a vacuum. Next, plasma etching (dry etching) is performed on the side surface 25a and the top surface 25b of the thinned wafer 25.

[0117] Specifically, this plasma etching is carried out, for example, by supplying a gas containing SF6 from a gas supply source 100a to the internal space of the chamber 74, and supplying Ar gas from a gas supply source 100b, while providing high-frequency power from a high-frequency power supply 96 to the gas ejection head 92.

[0118] As a result, the side surface 25a and top surface 25b of the thinned wafer 25 are etched by F-based radicals and the like generated in the internal space of the chamber 74, thereby mitigating the irregularities present on the side surface 25a and top surface 25b. With this, the dry etching step S4 is completed.

[0119] In addition, in the dry etching step S4, dry etching may be performed on the upper surface 25b of the thinned wafer 25 with a mask placed on the upper surface 25b. That is, in the dry etching step S4, dry etching may be performed on at least the side surface 25a of the thinned wafer 25.

[0120] After performing the dry etching step S4, the upper surface (second peeling surface) 25b of the thinned wafer 25 is ground (grinding step S5). After performing this grinding step S5, the side surface 25a and upper surface 25b of the thinned wafer 25 are polished (polishing step S6).

[0121] Figure 10 is a schematic perspective view showing an example of a processing apparatus used in grinding step S5 and polishing step S6. Note that the U-axis and V-axis directions shown in Figure 10 are mutually orthogonal directions on the horizontal plane, and the W-axis direction (vertical direction) is perpendicular to the U-axis and V-axis directions, respectively.

[0122] The processing apparatus 102 shown in Figure 10 includes a base 104 that supports each structure. An opening 104a is formed on the front upper surface of the base 104, and a transport mechanism 106 is provided within this opening 104a for transporting the bonded wafer, including the thinned wafer 25, while holding it in place. Furthermore, the transport mechanism 106 can also invert the bonded wafer while holding it.

[0123] Furthermore, cassette tables 108a and 108b are provided in front of the opening 104a. Cassettes 110a and 110b, capable of accommodating multiple bonded wafers, are placed on these cassette tables 108a and 108b, respectively. In addition, a position adjustment mechanism 112 for adjusting the position of the bonded wafers is provided diagonally behind the opening 104a.

[0124] The position adjustment mechanism 112 includes, for example, a table 112a configured to support the central portion of the bonded wafer, and a plurality of pins 112b configured to move closer to and further away from the table 112a, located outside the table 112a. The bonded wafer, for example, unloaded from the cassette 110a by the transport mechanism 106, is loaded onto the table 112a.

[0125] Then, the position adjustment mechanism 112 aligns the bonded wafer that has been loaded onto the table 112a. Specifically, by bringing multiple pins 112b closer to the table 112a until they contact the side surface of the bonded wafer loaded onto the table 112a, the position of the center of the bonded wafer is aligned to a predetermined position in a plane parallel to the U-axis and V-axis directions (UV plane).

[0126] Furthermore, a transport mechanism 114 is provided near the position adjustment mechanism 112, which rotates and transports the bonded wafer while holding it in place by suction. This transport mechanism 114 is equipped with a suction pad that can suck the upper surface of the bonded wafer and transports the bonded wafer, whose position has been adjusted by the position adjustment mechanism 112, to the rear. Behind the transport mechanism 114, a disc-shaped turntable 116 is provided.

[0127] The turntable 116 is connected to a rotational drive source (not shown), such as a motor. When this rotational drive source is operated, the turntable 116 rotates around a straight line passing through its center and parallel to the W-axis direction as its axis of rotation. In addition, multiple (for example, four) holding tables 118 are provided on the upper surface of the turntable 116 at roughly equal intervals along the circumferential direction of the turntable 116.

[0128] The transport mechanism 114 then unloads the bonded wafer from the table 112a of the position adjustment mechanism 112 and loads it into the holding table 118 located at an loading / unloading position near the transport mechanism 114. The turntable 116 rotates, for example, in the direction of the arrow shown in Figure 10, moving each holding table 118 in the order of loading / unloading position, rough grinding position, finish grinding position, and polishing position.

[0129] Furthermore, the holding table 118 is connected to a suction source (not shown), such as an ejector. When this suction source is activated, a suction force acts on the space near the upper surface (holding surface) of the holding table 118. Therefore, when the suction source is activated with the bonded wafer placed on the holding table 26, the bonded wafer can be held by the holding table 118.

[0130] Furthermore, the holding table 118 is connected to a rotational drive source (not shown), such as a motor. When this rotational drive source is operated, the holding table 118 rotates around a straight line passing through its center and parallel to the W-axis direction as its axis of rotation.

[0131] A columnar support structure 120 is provided behind both the rough grinding position and the finish grinding position (behind the turntable 116). A W-axis movement mechanism 122 is provided on the front surface of the support structure 120 (the surface facing the turntable 116). This W-axis movement mechanism 122 has a pair of W-axis guide rails 124 that are fixed to the front surface of the support structure 120 and extend along the W-axis direction.

[0132] Furthermore, a W-axis moving plate 126 is connected to the front side of the pair of W-axis guide rails 124 in a manner that allows it to slide along the pair of W-axis guide rails 124. A screw shaft 128 extending along the W-axis direction is positioned between the pair of W-axis guide rails 124. A motor 130 for rotating the screw shaft 128 is connected to the upper end of this screw shaft 128.

[0133] Furthermore, a nut (not shown) is provided on the surface of the screw shaft 128, where a helical groove is formed, to accommodate a number of balls that roll on the surface of the rotating screw shaft 128, thus forming a ball screw. That is, when the screw shaft 28 rotates, the number of balls circulate within the nut, causing the nut to move along the W-axis direction.

[0134] Furthermore, this nut is fixed to the rear (back) side of the W-axis moving plate 126. Therefore, when the screw shaft 128 is rotated by the motor 130, the W-axis moving plate 126 moves along the W-axis direction together with the nut. In addition, a fixing device 132 is provided on the front surface of the W-axis moving plate 126.

[0135] The fixture 132 supports a grinding unit 134 for grinding the upper surface 25b of the thinned wafer 25. The grinding unit 134 includes a spindle housing 136 fixed to the fixture 132. The spindle housing 136 houses a spindle 138 that extends along the W-axis direction in a rotatable manner.

[0136] A rotational drive source (not shown), such as a motor, is connected to the upper end of the spindle 138, and the spindle 138 can rotate on a straight line parallel to the W-axis direction using the power of this rotational drive source. The lower end of the spindle 138 is exposed from the lower surface of the spindle housing 136, and a disc-shaped mount 140 is fixed to this lower end.

[0137] A grinding wheel 142a for rough grinding is mounted on the underside of the mount 140 of the grinding unit 134 on the rough grinding position side. This grinding wheel 142a for rough grinding has a disc-shaped wheel base that is approximately the same diameter as the mount 140. Multiple grinding wheels (grinding wheels for rough grinding), each shaped like a rectangular parallelepiped, are fixed to the underside of this wheel base.

[0138] Similarly, a grinding wheel 142b for finish grinding is mounted on the underside of the mount 140 of the grinding unit 134 on the finish grinding position side. This grinding wheel 142b for finish grinding has a disc-shaped wheel base that is approximately the same diameter as the mount 140. Multiple rectangular parallelepiped grinding wheels (grinding wheels for finish grinding) are fixed to the underside of this wheel base.

[0139] Furthermore, each of the coarse grinding wheel and the finish grinding wheel contains abrasive grains, for example, diamond or cBN (cubic boron nitride), and a binder that holds these abrasive grains. Examples of binders used include metal bonds, resin bonds, or vitrified bonds.

[0140] Generally, the average particle size of abrasive grains in a finishing grinding wheel is smaller than that of a rough grinding wheel. For example, the average particle size of abrasive grains in a rough grinding wheel is between 0.5 μm and 30 μm, while the average particle size of abrasive grains in a finishing grinding wheel is less than 0.5 μm.

[0141] Furthermore, a grinding fluid supply nozzle (not shown) is positioned near the grinding wheels 142a and 142b to supply a liquid (grinding fluid) such as pure water to the processing point when grinding the upper surface 25b of the thinned wafer 25. Alternatively, instead of or in addition to this grinding fluid supply nozzle, an opening for supplying liquid may be provided on the grinding wheels 142a and 142b, and the grinding fluid may be supplied to the processing point through this opening.

[0142] Furthermore, a support structure 144 is provided on the side of the polishing position (on the side of the turntable 116). A U-axis movement mechanism 146 is provided on the side of the support structure 144 that faces the turntable 116. This U-axis movement mechanism 146 is fixed to the side of the support structure 144 that faces the turntable 116 and has a pair of U-axis guide rails 148 that extend along the U-axis direction.

[0143] Furthermore, a U-axis moving plate 150 is connected to the turntable 116 side of the pair of U-axis guide rails 148 in a manner that allows it to slide along the pair of U-axis guide rails 148. A screw shaft 152 extending along the U-axis direction is positioned between the pair of U-axis guide rails 148. A motor 154 for rotating the screw shaft 152 is connected to the front end of this screw shaft 152.

[0144] Furthermore, a nut (not shown) is provided on the surface of the screw shaft 152, where a helical groove is formed, to accommodate a number of balls that roll on the surface of the rotating screw shaft 152, thus forming a ball screw. That is, when the screw shaft 152 rotates, the number of balls circulate within the nut, causing the nut to move along the U-axis direction.

[0145] Furthermore, this nut is fixed to the side (back) of the U-axis moving plate 150 that faces the support structure 144. Therefore, when the screw shaft 152 is rotated by the motor 154, the U-axis moving plate 150 moves along the U-axis direction together with the nut. In addition, a W-axis moving mechanism 156 is provided on the side (front) of the U-axis moving plate 150 that faces the turntable 116.

[0146] This W-axis movement mechanism 156 has a pair of W-axis guide rails 158 fixed to the surface of the U-axis movement plate 150 and extending along the W-axis direction. A W-axis movement plate 160 is connected to the turntable 116 side of the pair of W-axis guide rails 158 in a manner that allows it to slide along the pair of W-axis guide rails 158.

[0147] Furthermore, a screw shaft 162 extending along the W-axis direction is positioned between a pair of W-axis guide rails 158. A motor 164 for rotating the screw shaft 162 is connected to the upper end of this screw shaft 162. A nut (not shown) is provided on the surface of the screw shaft 162, where a helical groove is formed, to accommodate a number of balls that roll on the surface of the rotating screw shaft 162, thus forming a ball screw.

[0148] In other words, as the screw shaft 162 rotates, numerous balls circulate within the nut, causing the nut to move along the W-axis direction. This nut is fixed to the side (back) of the W-axis moving plate 160 that faces the U-axis moving plate 150. Therefore, when the motor 164 rotates the screw shaft 162, the W-axis moving plate 160 moves along the W-axis direction along with the nut.

[0149] Furthermore, a fixing device 166 is provided on the surface of the W-axis moving plate 160 that faces the turntable 116. This fixing device 166 supports a polishing unit 168 for polishing the side surface 25a and top surface 25b of the thinned wafer 25. The polishing unit 168 includes a spindle housing 170 that is fixed to the fixing device 166.

[0150] The spindle housing 170 houses a spindle 172 that extends along the W-axis direction in a rotatable manner. A rotational drive source (not shown), such as a motor, is connected to the upper end of the spindle 172, and the spindle 172 rotates due to the power of this rotational drive source.

[0151] Furthermore, the lower end of the spindle 172 is exposed from the lower surface of the spindle housing 170, and a disc-shaped mount 174 is fixed to this lower end. A disc-shaped polishing pad 176 is mounted on the lower surface of the mount 174. This polishing pad 176 has a disc-shaped base that is approximately the same diameter as the mount 174.

[0152] A disc-shaped polishing layer, roughly the same diameter as mount 174, is fixed to the underside of this base. This polishing layer is a fixed abrasive layer in which abrasive grains are dispersed. For example, the polishing layer is manufactured by impregnating a polyester nonwoven fabric with a urethane solution containing abrasive grains with an average particle size of 0.4 μm to 0.6 μm, and then drying it.

[0153] The abrasive particles dispersed within the polishing layer consist of materials such as SiC, cBN, diamond, or metal oxide fine particles. These metal oxide fine particles may include those made of SiO2 (silica), CeO2 (ceria), ZrO2 (zirconia), or Al2O3 (alumina).

[0154] Furthermore, the polishing layer is flexible and elastically deforms in response to the pressure applied during polishing. For example, when polishing the upper surface 25b of the thinned wafer 25, the polishing layer elastically deforms in response to the pressure applied, so that the thinned wafer 25 becomes embedded in the polishing layer, that is, so that it covers the side surface 25a of the thinned wafer 25. In this case, both the upper surface 25b and the side surface 25a of the thinned wafer 25 are polished.

[0155] Furthermore, in order to polish the side surface 25a along with the top surface 25b of the thinned wafer 25 in this manner, it is preferable that the angle between the top surface 25b and the side surface 25a be 100° or more. In other words, it is preferable that the angle between the side surface and the top surface of the frustum of cone is 100° or more, that is, that the angle between the bottom surface and the side surface of the frustum of cone is 80° or less.

[0156] Furthermore, the radial center positions of the spindle 172, mount 174, and the base and polishing layer of the polishing pad 176 are roughly coincidental, and a cylindrical through-hole is formed to penetrate these center positions. This through-hole is connected to a polishing fluid supply source (not shown) that supplies a liquid (polishing fluid) such as pure water to the processing point when polishing the side surface 25a and top surface 25b of the thinned wafer 25.

[0157] This polishing fluid supply source includes a storage tank and a fluid delivery pump for the polishing fluid. The polishing fluid supply source also supplies the polishing fluid to the holding table 118 positioned at the polishing position through through holes formed in the spindle 172, etc. The polishing fluid may or may not contain abrasive particles.

[0158] Furthermore, a transport mechanism 178 is provided to the side of the transport mechanism 114, which rotates and transports the bonded wafer while holding it in place by suction. This transport mechanism 178 is equipped with a suction pad that can suck the upper surface of the bonded wafer and transports the bonded wafer, which is placed on the holding table 118 positioned at the loading / unloading position, forward.

[0159] Furthermore, a cleaning mechanism 180 is positioned in front of the transport mechanism 178 and behind the opening 104a, configured to clean the upper surface of the bonded wafers discharged by the transport mechanism 178. The bonded wafers cleaned by this cleaning mechanism 180 are then transported by the transport mechanism 106 and, for example, stored in a cassette 110b.

[0160] In the processing apparatus 102, for example, the grinding step S5 and polishing step S6 are performed in the following order. First, the transport mechanism 106 unloads the bonded wafer, including the thinned wafer 25, from the cassette 110a and places it on the table 112a of the position adjustment mechanism 112 so that the thinned wafer 25 is facing upwards. Next, the bonded wafer is aligned by bringing a plurality of pins 112b into contact with the bonded wafer.

[0161] Next, the transport mechanism 114 unloads the bonded wafer from the table 112a and loads it into the holding table 118, which is positioned at the loading / unloading location, with the thinned wafer 25 facing upwards. Then, the holding table 118, into which the bonded wafer has been loaded, holds the bonded wafer by suction on the support wafer 17 side.

[0162] Next, as shown in Figure 11, grinding is performed on the upper surface (second delamination surface) 25b of the thinned wafer 25. Specifically, first, the turntable 116 is rotated so that the holding table 118 that holds the bonded wafer is positioned in the rough grinding position.

[0163] Next, while rotating both the holding table 118 and the spindle 138 of the grinding unit 134 on the rough grinding position side, the grinding unit 134 on the rough grinding position side is lowered so that the grinding wheel of the grinding wheel 142a and the upper surface 25b of the thinned wafer 25 come into contact.

[0164] This rough grinding is performed on the upper surface 25b of the thinned wafer 25, thereby reducing the unevenness of the upper surface 25b. At this time, grinding fluid is supplied to the contact interface (processing point) between the grinding wheel 142a and the upper surface 25b of the thinned wafer 25.

[0165] Next, the grinding unit 134 on the rough grinding position side is raised so that the grinding wheel 142a and the upper surface 25b of the thinned wafer 25 are separated. Then, the rotation of both the holding table 118 and the spindle 138 of the grinding unit 134 on the rough grinding position side is stopped.

[0166] Next, the turntable 116 is rotated so that the holding table 118, which holds the bonded wafers, is positioned at the finish grinding position. Then, while rotating both the holding table 118 and the spindle 138 of the grinding unit 134 on the finish grinding position side, the grinding unit 134 on the finish grinding position side is lowered so that the grinding wheel 142b and the upper surface 25b of the thinned wafer 25 come into contact.

[0167] This allows the upper surface 25b of the thinned wafer 25 to undergo finish grinding, further reducing the unevenness of the upper surface 25b. At this time, grinding fluid is supplied to the contact interface (processing point) between the grinding wheel 142b and the upper surface 25b of the thinned wafer 25.

[0168] Next, the grinding unit 134 on the finish grinding position side is raised so that the grinding wheel 142b and the upper surface 25b of the thinned wafer 25 are separated. Then, the rotation of both the holding table 118 and the spindle 138 of the grinding unit 134 on the finish grinding position side is stopped. With this, grinding step S5 is completed.

[0169] Next, as shown in Figure 12, the side surface (first delamination surface) and the top surface (second delamination surface) 25b of the thinned wafer 25 are polished. Specifically, first, the turntable 116 is rotated so that the holding table 118 that holds the bonded wafer is positioned at the polishing position.

[0170] Next, while rotating both the holding table 118 and the spindle 172 of the polishing unit 168, the polishing unit 168 is lowered until the polishing layer of the polishing pad 176 is elastically deformed so that the thinned wafer 25 is embedded in the polishing layer, that is, so that it covers the side surface 25a and top surface 25b of the thinned wafer 25.

[0171] As a result, the side surface 25a and top surface 25b of the thinned wafer 25 are polished and flattened. At this time, polishing fluid P is supplied to the side surface 25a and top surface 25b of the thinned wafer 25 from a polishing fluid supply source through a through hole 182 that penetrates the spindle 172, mount 174, and polishing pad 176.

[0172] Next, the polishing unit 168 is raised so that the polishing layer of the polishing pad 176 is separated from the side surface 25a and the top surface 25b of the thinned wafer 25. Then, the rotation of both the holding table 118 and the spindle 172 is stopped. With this, polishing step S6 is completed.

[0173] Here, the polishing step S6 is performed after the dry etching step S4 to alleviate the irregularities on at least the side surface 25a of the thinned wafer 25 and the grinding step S5 to alleviate the irregularities on its upper surface 25b. This is preferable because it reduces the burden on the polishing layer of the polishing pad 176 compared to the case where the grinding step S5 and polishing step S6 are performed without performing the dry etching step S4.

[0174] Specifically, if the dry etching step S4 is not performed, the presence of irregularities on the side surface 25a of the thinned wafer 25 causes severe wear of the polished layer in the portion that comes into contact with the side surface 25a. In contrast, if the dry etching step S4 is performed, the irregularities on the side surface 25a of the thinned wafer 25 are mitigated, thereby preventing localized wear of the polished layer.

[0175] Next, the turntable 116 is rotated so that the holding table 118 that holds the bonded wafer is positioned at the loading / unloading position. Then, the holding table 118, now positioned at the loading / unloading position, stops suctioning the bonded wafer from the support wafer 17 side.

[0176] Next, the transport mechanism 178 unloads the bonded wafer from the holding table 118 and loads it into the cleaning mechanism 180 with the thinned wafer 25 facing upwards. Then, the cleaning mechanism 180 cleans the sides 25a and top surface 25b of the bonded wafer. Next, the transport mechanism 106 loads the bonded wafer into the cassette 110b.

[0177] In the wafer processing method shown in Figure 2, a first delamination layer is formed inside the wafer 11 along the side surface of a frustum defined by a bottom surface (first bottom surface) located on the surface 11a side of the wafer 11, a side surface having one end that overlaps with the outer circumference of the bottom surface and that slopes away from the outer circumference region as it moves away from the surface 11a side of the wafer 11, and an upper surface (second bottom surface) having an outer circumference that overlaps with the other end of the side surface and that is parallel to the bottom surface.

[0178] In this method, the wafer 11 is divided such that a thinned wafer 25 is formed having a side surface (first delamination surface) 25a that is exposed by being divided in the first delamination layer and an upper surface (second delamination surface) 25b that is exposed by being divided in the second delamination layer.

[0179] In this wafer 11, the first delamination layer is formed along the side surface of the frustum of the cone. In this case, when the wafer 11 is divided, the probability of the side surface 25a of the thinned wafer 25 and the inner surface of the remaining portion of the wafer 11 (including the outer peripheral region) coming into contact with each other is low. Therefore, this method can suppress the occurrence of cracks caused by contact between the two.

[0180] Furthermore, in this wafer 11, cracks tend to propagate from the first delamination layer in a direction along the side surface of the frustum. In this case, the probability of cracks propagating toward the device 15 formed on the surface side of the thinned wafer 25 is reduced. Therefore, this method suppresses damage to the device 15 associated with the formation of the thinned wafer 25 and also suppresses the reduction in the number of chips that can be manufactured from wafer 11.

[0181] The wafer processing method described above is one aspect of the present invention, and the present invention is not limited to the method described above. For example, in the present invention, the bonding step S1 may not be performed. That is, the present invention may be a wafer processing method in which a single wafer 11 is divided to form a thinned wafer 25.

[0182] Furthermore, in the bonding step S1 of the present invention, the surface 11a of the wafer 11 and the surface 17a of the support wafer 17 may be bonded together without using the adhesive 19. For example, the bonding step S1 of the present invention may be carried out by removing foreign matter from each surface 11a and 17a and bringing both surfaces 11a and 17a into contact in an activated state.

[0183] Furthermore, in the present invention, after the first and second release layers are formed on the wafer 11, the surface 11a side of the wafer 11 may be bonded to the surface 17a side of the support wafer 17. In other words, in the present invention, the bonding step S1 may be performed after the release layer formation step S2.

[0184] Furthermore, the structure of the laser processing apparatus used in the peeling layer formation step S2 of the present invention is not limited to the structure of the laser processing apparatus 2 described above. For example, the peeling layer formation step S2 of the present invention may be carried out using a laser processing apparatus equipped with a horizontal movement mechanism that moves the irradiation head 52 of the laser beam irradiation unit 42, etc., along the X-axis and / or Y-axis directions, respectively.

[0185] Alternatively, step S2 of the present invention may be carried out using a laser processing apparatus in which a scanning optical system capable of changing the direction of the laser beam LB irradiated from the irradiation head 52 is provided in the laser beam irradiation unit 42. This scanning optical system includes, for example, a galvanometer scanner, an acousto-optic element (AOD), and / or a polygon mirror.

[0186] In other words, in the laser processing apparatus used in step S2 of the present invention, it is sufficient that the holding table 26 and the focal point where the laser beam LB is focused can move relative to each other along the X-axis, Y-axis, and Z-axis directions, and there are no limitations on the structure for this purpose.

[0187] Furthermore, in step S2 of the present invention, the method for forming the first peeling layer is not limited to the method described above. For example, in step S2 of the present invention, the first peeling layer may be formed by irradiating the wafer 11 with a laser beam LB in an annular shape by operating a horizontal movement mechanism 6 instead of a rotational drive source connected to the lower part of the table base 24.

[0188] Furthermore, in step S2 of the present invention, the first delamination layer may be formed by irradiating the wafer 11 with an unbranched laser beam LB. In this case, for example, the focal point where the laser beam LB is focused and the wafer 11 can be moved relative to each other so that a trajectory like a spiral staircase is drawn inside the wafer 11, with the radius of rotation decreasing as it goes upwards.

[0189] Furthermore, in step S2 of the present invention, the method for forming the second peel layer is not limited to the method described above. For example, in step S2 of the present invention, the second peel layer may be formed by irradiating the wafer 11 with a laser beam LB in a spiral manner by operating a rotary drive source connected to the lower part of the table base 24 in addition to the horizontal movement mechanism 6.

[0190] Furthermore, in the splitting step S3 of the present invention, ultrasound may be applied to the wafer 11 prior to splitting the wafer 11 on which the first and second delamination layers are formed. In this case, the cracks 23 contained in the first and second delamination layers will extend, making it easier to separate the wafer 11.

[0191] Furthermore, the structures and methods of the embodiments described above can be modified as appropriate without departing from the scope of the present invention. [Explanation of Symbols]

[0192] 2: Laser processing equipment 4: Base 6: Horizontal movement mechanism 8: Y-axis guide rail 10: Y-axis moving plate 11: Wafer (11a: front surface, 11b: back surface, 11c: side surface) 12: Screw shaft 13: Notch 14: Motor 15: Device 16: X-axis guide rail 17: Support wafer (second wafer) (17a: front side, 17b: back side) 18: X-axis movement plate 19: Adhesive 20: Screw shaft 21: Modification section 22: Motor 23: Crack 24: Table base 25: Thinned wafer (25a: side surface (first delamination surface), 25b: top surface (second delamination surface)) 26: Holding table (26a: Porous plate) 30:Support structure 32: Vertical movement mechanism 34: Z-axis guide rail 36: Z-axis movement plate 38: Motor 40: Support 42: Laser beam irradiation unit 44: Laser oscillator 46: Attenuator 48: Spatial Light Modulator 50: Miller 52: Irradiation head 54: Housing 56: Imaging Unit 58: Touch panel 60:Dividing device 62: Holding Table 64: Split Unit 66: Support member 68: Support stand 70: Movable member (70a: Vertical part, 70b: Claw part) 72: Etching equipment 74: Chamber (74a: Inlet / Outlet, 74b: Exhaust Port) 76: Gate valve 78: Piping 80: Exhaust system 82: Support member 84: Table 90: Bearing 92: Gas ejection head (92a: gas diffusion space, 92b: gas outlet, 92c, 92d: gas supply port) 94: Matching box 96:High frequency power supply 98a,98b: Piping 100a, 100b: Gas supply source 102: Processing equipment 104: Base (104a: Opening) 106: Conveying mechanism 108a, 108b: Cassette Table 110a, 110b: Cassette 112: Position adjustment mechanism (112a: Table, 112b: Pin) 114: Conveying mechanism 116: Turntable 118: Holding Table 120: Support structure 122:W-axis movement mechanism 124: W-axis guide rail 126: W-axis moving plate 128: Screw shaft 130: Motor 132: Fixtures 134: Grinding Unit 136: Spindle Housing 138: Spindle 140: Mount 142a, 142b: Grinding wheels 144: Support structure 146:U-axis movement mechanism 148: U-axis guide rail 150: U-axis moving plate 152: Screw shaft 154: Motor 156:W-axis movement mechanism 158: W-axis guide rail 160: W-axis moving plate 162: Screw shaft 164: Motor 166: Fixtures 168: Polishing Unit 170: Spindle Housing 172: Spindle 174: Mount 176: Polishing pad 178: Conveying mechanism 180: Cleaning mechanism 182: Through hole

Claims

1. A wafer processing method comprising: irradiating a wafer, on which multiple devices are formed on the surface and whose outer peripheral region is chamfered, with a laser beam of a wavelength that penetrates the wafer to form a delamination layer inside the wafer; and then applying an external force to the wafer along the thickness direction of the wafer to divide the wafer in the delamination layer, wherein A delamination layer forming step of forming a first delamination layer along the side surface of a frustum defined by a first bottom surface located on the surface side of the wafer, a side surface having one end overlapping the outer circumference of the first bottom surface and inclined so as it moves away from the surface side of the wafer, and a second bottom surface having an outer circumference overlapping the other end of the side surface and parallel to the first bottom surface, and a second delamination layer along the second bottom surface, After performing the delamination layer formation step, a dividing step is performed to divide the wafer so that a thinned wafer is formed having a first delamination surface exposed by being divided in the first delamination layer and a second delamination surface exposed by being divided in the second delamination layer. After performing the splitting step, a dry etching step is performed on at least the first peeled surface, After performing the dry etching step, a grinding step is performed on the second peeled surface, After performing the grinding step, a polishing step is performed on the first peeled surface and the second peeled surface, A wafer processing method comprising the following.

2. The wafer processing method according to claim 1, further comprising a bonding step of bonding the surface side of the wafer to the surface side of a second wafer different from the wafer, before performing the splitting step.

3. The wafer processing method according to claim 1, wherein dry etching is not performed on the first peeled surface and the second peeled surface between the grinding step and the polishing step.

4. The wafer processing method according to claim 1, wherein in the polishing step, the first peeled surface and the second peeled surface are polished simultaneously.

5. The wafer processing method according to any one of claims 1 to 4, wherein in the step of forming the delamination layer, one of the first delamination layer and the second delamination layer is formed before the other.

6. The wafer includes a first region and a second region, each extending along a predetermined direction. The first peeling layer includes a pair of first inclined portions formed in the first region and a pair of second inclined portions formed in the second region. The second peel layer includes a first straight section formed in the first region and located between the pair of first inclined sections in the predetermined direction, and a second straight section formed in the second region and located between the pair of second inclined sections in the predetermined direction. The wafer processing method according to any one of claims 1 to 4, wherein in the peel layer formation step, the pair of first inclined portions and the first straight portion are formed before the pair of second inclined portions and the second straight portion is formed.

7. The wafer processing method according to any one of claims 1 to 4, wherein in the dry etching step, dry etching is performed on the second peel surface together with the first peel surface.

8. The wafer processing method according to any one of claims 1 to 4, wherein in the dry etching step, dry etching is performed on the first peeling surface while a mask is provided on the second peeling surface.

Images