Wafer manufacturing method

Abstract

To suppress a wafer from warping, in manufacturing the wafer from an ingot having warpage using a laser beam.SOLUTION: A wafer manufacturing method, which manufactures a wafer from a work-piece that has a first surface, a second surface positioned at the opposite side of the first surface and a side surface connected to an outer peripheral edge of the first surface and an outer peripheral edge of the second surface, and which includes warpage of the first surface and the second surface, comprises: a first grinding step in which the first surface of the work-piece is ground in a state where the work-piece is held on a first holding table so that the second surface faces the first holding table and the first surface is exposed, by making two or more fixing members which the first holding table has tuck and fix the side surface of the work-piece without making the first holding table suction and hold the second surface; a second grinding step of grinding the second surface of the work-piece; a peeled layer forming step of forming a peeled layer inside the work-piece; and a peeling step of peeling the wafer from the work-piece, with the peeled layer set as a starting point.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method for manufacturing a wafer from a workpiece having a first surface, a second surface opposite the first surface, and a side surface connecting an outer peripheral edge of the first surface and an outer peripheral edge of the second surface, the workpiece including warpage on the first surface and the second surface. [Background technology]

[0002] Device chips mounted on electronic devices are manufactured by forming multiple devices on the surface of a wafer (e.g., a silicon wafer), which is a single crystal substrate of semiconductor material such as silicon, and then dividing the wafer into device units. The silicon wafers used in manufacturing these device chips are usually manufactured by cutting a cylindrical ingot with a wire saw.

[0003] However, the diameter of the wire saw is relatively large compared to the thickness of the wafers to be manufactured, and furthermore, the front and back surfaces of the wafers cut by the wire saw must be polished to reduce unevenness on those surfaces, resulting in a problem that the volume of ingots discarded per wafer is relatively large.

[0004] Therefore, instead of using a wire saw, a method of manufacturing wafers from an ingot using a laser beam has been proposed (see, for example, Patent Document 1). Specifically, a laser beam is used to form a peeling layer at a predetermined depth inside the ingot, and then the wafer is peeled from the ingot starting from the peeling layer, and the peeled surface of the ingot from which the wafer has been peeled is ground to flatten it. By repeating this series of steps, multiple wafers are manufactured from the ingot.

[0005] When forming a peeling layer on an ingot using a laser beam in this manner, one surface of the ingot is usually ground to make the surface of the ingot irradiated with the laser beam approximately flat.

[0006] However, when the ingot used in the laser processing is an as-sliced ​​ingot produced by cutting the straight body portion, excluding the tail, cone, etc., of a cylindrical ingot produced by a single crystal growth method such as the Czochralski method, into several pieces in its longitudinal direction, the ingot usually has warping on its front and back surfaces.

[0007] Therefore, when this ingot is suction-held on the holding surface of the chuck table, the ingot deforms to conform to the shape of the holding surface so as to correct the warp, and therefore, when the ingot is removed from the holding surface after grinding, the warp remains in the ingot after grinding.

[0008] Therefore, if a delamination layer is formed on an ingot that still has warpage after grinding, and then a wafer is delaminate from the ingot starting from the delamination layer, there is a problem that the delaminate wafer still has warpage. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] JP 2019-12765 A Summary of the Invention [Problem to be solved by the invention]

[0010] The present invention has been made in consideration of the above problems, and has an object to suppress warping of a wafer when a wafer is manufactured from a warped ingot using a laser beam. [Means for solving the problem]

[0011] According to one aspect of the present invention, there is provided a method for manufacturing a wafer from a workpiece having a first surface, a second surface located opposite to the first surface, and a side surface connecting an outer periphery of the first surface and an outer periphery of the second surface, the first surface and the second surface including warpage, the method including a first grinding step of grinding the first surface of the workpiece in a state in which the workpiece is held by the first holding table such that the second surface faces the first holding table and the first surface is exposed by sandwiching and fixing the side surface of the workpiece with two or more fixing members of the first holding table without holding the second surface by suction with the first holding table; a second grinding step of grinding the second surface of the workpiece after the grinding step; a delamination layer forming step of forming a delamination layer inside the workpiece, the delamination layer including a modified layer and cracks extending from the modified layer, by positioning a focal point of a laser beam having a wavelength that transmits the workpiece at a predetermined depth inside the workpiece corresponding to a thickness of a wafer after the second grinding step and relatively moving the focal point and the workpiece along a predetermined direction perpendicular to the thickness direction of the workpiece; and a delamination step of peeling the wafer from the workpiece using the delamination layer as a starting point after the delamination layer forming step.

[0012] Preferably, in the second grinding step, the first surface ground in the first grinding step is suction-held by a holding surface of a second holding table so that the second surface is exposed.

[0013] Preferably, the first surface exposed in the first grinding step has a convex warp, and in the first grinding step, the first surface is ground to reduce the warp of the first surface.

[0014] Preferably, the wafer manufacturing method further includes, prior to the peeling layer formation step, a planarization step for reducing surface roughness of one of the first surface ground in the first grinding step and the second surface ground in the second grinding step, which is to be irradiated with the laser beam in the peeling layer formation step. Effect of the Invention

[0015] In one embodiment of the wafer manufacturing method of the present invention, in a first grinding step, the second surface of a warped workpiece is not held by suction on a first holding table, but the sides of the workpiece are clamped and fixed with two or more fixing members, so that the first surface of the workpiece is ground while the workpiece is held on the first holding table so that the first surface is exposed.

[0016] Therefore, compared to grinding the first surface of the workpiece in a state in which the second surface is deformed by holding the second surface by suction with the first holding table, it is possible to reduce warpage remaining on the first surface when the workpiece is removed from the first holding table after grinding. Therefore, it is possible to reduce warpage of the wafer peeled off from the workpiece after the peeling layer forming step and the peeling step. [Brief description of the drawings]

[0017] [Figure 1] FIG. 2 is a flow diagram showing a wafer manufacturing method. [Diagram 2] FIG. 2(A) is a perspective view of the ingot, and FIG. 2(B) is a side view of the ingot. [Diagram 3] FIG. 3(A) is a plan view of the first holding table, and FIG. 3(B) is a side view of the first holding table and an ingot fixed by the first holding table. [Figure 4] FIG. 13 is a diagram showing a state at the start of a first grinding step. [Diagram 5] FIG. 13 is a diagram showing the state at the end of the first grinding process. [Figure 6] 4 is an example of a cross-sectional curve of a first surface of an ingot before and after a first grinding step. [Figure 7] FIG. 13 is a diagram showing a state at the start of a second grinding step. [Figure 8] FIG. 13 is a diagram showing the state at the end of the second grinding process. [Figure 9] FIG. 11 is a perspective view showing a release layer forming step. [Figure 10] FIG. 10(A) is a plan view showing the separation layer forming step, and FIG. 10(B) is a cross-sectional view of the ingot on which the separation layer has been formed. [Figure 11] FIG. 11A is a diagram showing the state at the start of the peeling process, and FIG. 11B is a diagram showing the state at the end of the peeling process. [Figure 12] FIG. 12(A) is a flow diagram showing a first modified example of the method for producing a wafer, and FIG. 12(B) is a flow diagram showing a second modified example of the method for producing a wafer. [Figure 13] 13A to 13C are diagrams illustrating a second grinding step in a third modified example of the wafer manufacturing method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] (First embodiment) An embodiment according to one aspect of the present invention will be described with reference to the accompanying drawings. Fig. 1 is a flow diagram showing a method for manufacturing a wafer 13 (see Fig. 11(B)) from an ingot (i.e., workpiece) 11 (see Fig. 2(A)) as a workpiece.

[0019] In this embodiment, as shown in Fig. 1, the steps are performed in the order of a first grinding step S10, a second grinding step S20, a peeling layer forming step S30, a peeling step S40, a determination step S50, and a peeled surface grinding step S60. First, with reference to Figs. 2(A) and 2(B), an ingot 11 to be subjected to each step will be described.

[0020] 2(A) is a perspective view of an ingot 11. The ingot 11 is a single crystal of silicon (Si). However, the conductivity type of the ingot 11 is not particularly limited, and may be p-type or n-type. The ingot 11 of this embodiment is cylindrical and has a diameter of 12 inches (about 300 mm) and a thickness of 14 mm.

[0021] However, the diameter and thickness of the ingot 11 are not limited to these values. However, the thickness of the ingot 11 is preferably 3 mm or more in view of the structure of the first holding table 12 described later, and is preferably 1000 mm or less in view of the safe execution of processing.

[0022] The ingot 11 has a first surface 11a and a second surface 11b located on the opposite side to the first surface 11a in the thickness direction 11c of the ingot 11, and an outer peripheral edge 11a1 of the first surface 11a and an outer peripheral edge 11b1 of the second surface 11b are connected by a side surface 11d.

[0023] In the present embodiment, a notch 11d1 is formed on the side surface 11d along the thickness direction 11c, but an orientation flat (not shown) may be formed instead of the notch 11d1.

[0024] As shown in Fig. 2(B), the ingot 11 has warpage on the first surface 11a and the second surface 11b. Fig. 2(B) is a side view of the ingot 11. The first surface 11a of this embodiment has a convex warpage 11e1.

[0025] The warpage 11e1 of the first surface 11a is, for example, about 50 μm. The second surface 11b of this embodiment has a concave warpage 11e2. The warpage 11e2 of the second surface 11b is also, for example, about 50 μm.

[0026] The warping of the first surface 11a and the second surface 11b can be measured, for example, using a measuring device (for example, a contact-type microshape measuring device, or a non-contact height measuring device that uses a laser beam for measurement rather than processing) that moves a stage (not shown) on which the ingot 11 is placed and a sensor head (not shown) relatively.

[0027] Next, the grinding device 10 (see FIG. 4) used in the first grinding step S10 will be described with reference to FIG. 3(A) to FIG. 5. The grinding device 10 has a first holding table 12 (see FIG. 3(A)).

[0028] FIG. 3(A) is a plan view of first holding table 12, and FIG. 3(B) is a side view of first holding table 12 and ingot 11 fixed by first holding table 12.

[0029] The first holding table 12 has a disk-shaped inner frame 14 made of non-porous ceramics. A disk-shaped recess (not shown) is formed on the top of the inner frame 14 and is concentric with the inner frame 14, and a porous plate 16 made of porous ceramics is fixed to this recess.

[0030] The upper surface of the porous plate 16 is approximately flat. The upper surfaces of the inner frame 14 and the porous plate 16 are approximately flush with each other to form an approximately flat holding surface. However, the holding surface of the first holding table 12 may be conical in shape, with the center protruding slightly compared to the outer periphery.

[0031] A suction source (not shown) such as a vacuum pump is connected to the inner frame 14 via a predetermined flow path (not shown), and negative pressure generated by the suction source can be transmitted to the porous plate 16 via the predetermined flow path. A valve (not shown) such as an electromagnetic valve is provided in this predetermined flow path.

[0032] By opening and closing the valve, the transmission and non-transmission of negative pressure to the porous plate 16 is controlled. For example, when the valve is opened, the first holding table 12 functions as a chuck table that suctions and holds the ingot 11 with negative pressure. On the other hand, when the valve is closed, the negative pressure is not transmitted to the porous plate 16.

[0033] The first holding table 12 is configured to be rotatable around a predetermined rotation shaft 12a (see FIG. 4). For example, a driven pulley (not shown) is fixed to the rotation shaft 12a, and power is transmitted to the driven pulley from a drive pulley (not shown) fixed to the output shaft of a motor (not shown) via a belt (not shown).

[0034] In addition, a tilt adjustment mechanism (not shown) is provided under the first holding table 12. The tilt adjustment mechanism adjusts the tilt of the first holding table 12 so that the upper surfaces (i.e., holding surfaces) of the inner frame 14 and the porous plate 16 are approximately parallel to a horizontal plane perpendicular to the Z-axis direction (i.e., a direction parallel to the vertical direction).

[0035] When the holding surface is conical, the tilt of the first holding table 12 is adjusted by the tilt adjustment mechanism so that a portion of the holding surface is approximately parallel to the horizontal plane. At this time, the rotation axis 12a is slightly tilted with respect to the Z-axis direction.

[0036] 3, a clamping jig 18 is fixed to the outer periphery of the inner frame 14. The clamping jig 18 has an annular outer frame 20 made of a metal material or ceramics. The outer frame 20 has four screw holes (not shown) that penetrate from its outer periphery side surface to its inner periphery side surface.

[0037] The multiple screw holes in the outer frame body 20 are arranged at approximately equal intervals along the circumferential direction of the outer frame body 20, and the shaft portions 22b of the bolts 22 are inserted into each screw hole along the radial direction of the outer frame body 20. By rotating the head portions 22a of the bolts 22 in a predetermined direction, the tips of the shaft portions 22b are pressed against the outer peripheral side surface of the inner frame body 14.

[0038] In this manner, the outer frame body 20 is fixed to the inner frame body 14 by sandwiching the inner frame body 14 with the multiple bolts 22. At this time, the upper surface of the outer frame body 20 and the upper surface of the inner frame body 14 are positioned, for example, on the same plane.

[0039] Clamps (fixing members) 24 are provided above each screw hole on the upper surface of the outer frame 20. In this embodiment, four (i.e., two or more) clamps 24 are provided. Each clamp 24 has a rectangular parallelepiped base 26 fixed to the outer frame 20.

[0040] Base 26 includes a cylindrical screw hole (not shown) that opens on its upper surface, and shaft 28b of bolt 28 is rotatably fastened to this screw hole. Bolt 28 can advance and retreat relative to base 26 by rotating head 28a of bolt 28.

[0041] A rectangular parallelepiped movable member 30 is provided between the head 28a of the bolt 28 and the upper surface of the base 26 in a manner that allows it to move in a direction perpendicular to the longitudinal direction of the shaft 28b of the bolt 28. The movable member 30 has a groove along the longitudinal direction of the movable member 30, and the shaft 28b of the bolt 28 described above is inserted into this groove.

[0042] The bolt 28 is rotated to sandwich the movable member 30 between the head 28a and the upper surface of the base 26, thereby fixing the position of the movable member 30 relative to the outer frame 20. A flat pressing plate 32 is fixed to one end of each movable member 30 near the center of the first holding table 12.

[0043] In this embodiment, the four pressing plates 32 are arranged at approximately equal intervals, so that two pressing plates 32 face each other, and the remaining two pressing plates 32 also face each other. The tip surface of each pressing plate 32 close to the center of the first holding table 12 becomes a contact surface that comes into contact with the side surface 11d of the ingot 11.

[0044] The contact surface of each pressing plate 32 may be a flat surface or a curved surface having a curvature corresponding to the curvature of the side surface 11d of the ingot 11. In addition, the contact surface of the pressing plate 32 may be formed of a relatively soft material having elasticity, such as a resin.

[0045] In addition, although the first holding table 12 in this embodiment has four clamp portions 24, it may have two or three clamp portions 24, or may have five or more clamp portions 24, as long as they are able to clamp and fix the side surface 11d of the ingot 11 in the radial direction of the ingot 11.

[0046] However, since the first holding table 12 rotates at, for example, 100 rpm or more when processing the ingot 11, in order to properly hold the ingot 11, it is preferable that the first holding table 12 has three or more clamp portions 24.

[0047] 4, a grinding unit 34 is provided above the first holding table 12. A Z-axis direction moving mechanism (not shown) including a ball screw is connected to the grinding unit 34, and the grinding unit 34 is movable along the Z-axis direction.

[0048] The grinding unit 34 has a cylindrical spindle 36. The longitudinal direction of the spindle 36 is disposed substantially parallel to the Z-axis direction. A part of the spindle 36 is rotatably accommodated in a spindle housing (not shown). The spindle 36 functions as a rotor that constitutes a motor.

[0049] A disk-shaped mount 38 is provided at the lower end of the spindle 36. An annular grinding wheel 40 is attached to the underside of the mount 38 by using a bolt (not shown). The grinding wheel 40 has a wheel base 40a having approximately the same outer diameter as the mount 38.

[0050] A plurality of grinding wheels 40b each having a block shape are arranged at approximately equal intervals on the lower surface side of the wheel base 40a. Each grinding wheel 40b has abrasive grains and a bond material for fixing the abrasive grains. The grain size of the abrasive grains in the grinding wheel 40 is, for example, #240 to #1200. That is, the grinding wheel 40 is a so-called rough grinding wheel.

[0051] The grain sizes identified by the combination of # and number are described in the Japanese Industrial Standards (i.e., JIS (Japanese Industrial Standards) standard R6001-2:2017 (Grain size of abrasives for grinding wheels - Part 2: Fine powder). Grain sizes from #240 to #1200 may be evaluated by the sedimentation test method or the electrical resistance test method.

[0052] Here, a description will be given of the first grinding step S10 for grinding the first surface 11a of the ingot 11. In the first grinding step S10, first, an operator uses the above-mentioned measuring device to identify which of the two surfaces of the ingot 11 is the first surface 11a having a convex shape (and which is the second surface 11b having a concave shape).

[0053] 3(B), an operator places the ingot 11 on the first holding table 12 so that the concave second surface 11b faces the first holding table 12 and the convex first surface 11a is exposed upward. At this time, the center of the first surface 11a and the center of the porous plate 16 are approximately aligned.

[0054] Then, each movable member 30 is slid toward the center of the first holding table 12 so that each pressing plate 32 comes into contact with the side surface 11d of the ingot 11. In this state, the bolt 28 is screwed in to fix the position of the movable member 30 relative to the outer frame body 20.

[0055] Particularly, in the first grinding step S10, a valve provided in a predetermined flow path between the suction source and the porous plate 16 is closed, and negative pressure is not transmitted to the porous plate 16. Therefore, a disk-shaped gap 17 is formed between the second surface 11b and the upper surface of the porous plate 16.

[0056] The ingot 11 is clamped and fixed between multiple clamp portions 24 of the first holding table 12 with the first surface 11a exposed upward, without the second surface 11b being held by suction with the porous plate 16 of the first holding table 12.

[0057] With the ingot 11 held by the first holding table 12 in this manner, grinding water such as pure water is supplied to the contact area between the grinding wheel 40b and the ingot 11, while the first holding table 12 and the spindle 36 are each rotated in a predetermined direction and the grinding unit 34 is ground (i.e., moved downward) at a predetermined grinding feed speed.

[0058] In this manner, in-feed grinding is performed to grind the entire first surface 11a with the grinding wheel 40. Fig. 4 is a diagram showing a state at the start of the first grinding step S10. The processing conditions in the first grinding step S10 are set, for example, as follows.

[0059] Spindle speed: 3200 rpm Rotational speed of the first holding table: 300 rpm Grinding feed rate: 3.0μm / s

[0060] In this embodiment, the convex first surface 11a is exposed upward, and therefore tilting, misalignment, etc. of the ingot 11 are suppressed compared to a case in which the concave second surface 11b is exposed upward and the convex first surface 11a is positioned downward before grinding is performed, thereby enabling grinding to be performed stably.

[0061] After the grinding unit 34 is fed for a predetermined time to grind the first surface 11a and reduce the warpage 11e1 of the first surface 11a (for example, to make the first surface 11a substantially flat), the grinding feed of the grinding unit 34 is stopped to end the first grinding step S10. Fig. 5 is a diagram showing a state at the end of the first grinding step S10.

[0062] After the grinding feed is stopped, the grinding unit 34 is moved upward to move the grinding wheel 40 sufficiently away from the ingot 11. Next, after stopping the rotation of the first holding table 12, an operator loosens each bolt 28 to move each movable member 30 away from the ingot 11 and remove the ingot 11 from the first holding table 12.

[0063] 6 is an example of a cross-sectional curve of the first surface 11a (i.e., actual surface) of the ingot 11 before and after the first grinding step S10. The horizontal axis indicates the position (mm) in the radial direction of the first surface 11a, and the vertical axis indicates the height (μm) of the first surface 11a relative to a reference height (0 μm on the vertical axis) in the radial direction of the first surface 11a. Although the right end portion of the graph in FIG. 6 is missing, the diameter of the first surface 11a is 300 mm.

[0064] The graph shown in Fig. 6 was measured using a microprofile measuring instrument (model name: ET4100S) sold by Kosaka Laboratory Co., Ltd. As is clear from Fig. 6, the warpage 11e1 of the first surface 11a before the first grinding step S10 was about 50 µm, whereas the warpage 11e1 of the first surface 11a after the first grinding step S10 was reduced to about 5 µm.

[0065] After the first grinding step S10, a second grinding step S20 is performed to grind the second surface 11b of the ingot 11. First, a grinding device 50 used in the second grinding step S20 will be described with reference to Fig. 7. The grinding device 50 has a second holding table 52 different from the first holding table 12.

[0066] The second holding table 52 differs from the first holding table 12 in that it does not have the clamp jig 18, but the other structures are the same as those of the first holding table 12. Therefore, the same reference numerals as those of the first holding table 12 are used for elements common to the first holding table 12, and detailed descriptions thereof will be omitted.

[0067] The second holding table 52 functions as a chuck table. In the second holding table 52, the upper surface of the porous plate 16 is also approximately flat, as in the first holding table 12. In the second holding table 52, the upper surface of the inner frame 14 and the upper surface of the porous plate 16 form a holding surface 52a that is approximately flush with each other.

[0068] Similar to the first holding table 12, the holding surface 52a may have a conical shape with a slight protrusion at the center of the porous plate 16. Similar to the grinding device 10, a grinding unit 34 is provided above the second holding table 52.

[0069] In the second grinding process S20, when grinding the second surface 11b of the ingot 11, first, an operator places the ingot 11 on the second holding table 52 so that the second surface 11b is exposed upward and the first surface 11a after grinding faces the second holding table 52 (see Figure 7).

[0070] Then, the valve is opened to transmit negative pressure from the suction source to the porous plate 16, whereby the first surface 11a, which has been ground and become substantially flat in the first grinding step S10, is sucked and held by the holding surface 52a.

[0071] Next, with the ingot 11 held by suction on the second holding table 52, grinding water is supplied to the contact area between the grinding wheel 40b and the ingot 11 while the second holding table 52 and the spindle 36 are rotated in predetermined directions and the grinding unit 34 is ground at a predetermined grinding feed speed.

[0072] 7 is a diagram showing a state at the start of the second grinding step S20. The processing conditions in the second grinding step S20 are set to be, for example, substantially the same as those in the above-mentioned first grinding step S10.

[0073] The grinding unit 34 is fed for grinding for a predetermined time to grind the second surface 11b and reduce the warpage 11e2 of the second surface 11b (for example, to make the second surface 11b substantially flat), and then the grinding feed of the grinding unit 34 is stopped. Fig. 8 is a diagram showing a state at the end of the second grinding step S20.

[0074] After the grinding feed is stopped, the grinding unit 34 is moved upward to move the grinding wheel 40 sufficiently away from the ingot 11. Next, after the rotation of the second holding table 52 is stopped, the suction holding on the holding surface 52a is released, and the worker removes the ingot 11 from the second holding table 52.

[0075] After the second grinding step S20, a peeling layer forming step S30 is performed. In the peeling layer forming step S30, a laser processing device 60 (see FIG. 9) is used to form a peeling layer 15 (see FIG. 10(B)) over one side of the inside of the ingot 11 at a predetermined depth 11f of the ingot 11. First, the laser processing device 60 will be described with reference to FIGS. 9, 10(A) and 10(B).

[0076] The laser processing apparatus 60 has a third holding table 62. The third holding table 62 also functions as a chuck table. The third holding table 62 includes a frame and a porous plate corresponding to the inner frame 14 and the porous plate 16 of the first holding table 12 described above.

[0077] The upper surfaces of the frame and the porous plate in the third holding table 62 are also approximately flat and approximately flush, forming an approximately flat holding surface 62a (see FIG. 10(B)) that suction-holds the ingot 11. The third holding table 62 is configured to be movable along the X-axis direction and the Y-axis direction, which are each perpendicular to the Z-axis direction (vertical direction).

[0078] For example, the third holding table 62 can be moved along the X-axis direction by a ball screw type X-axis direction moving mechanism (not shown), and can be moved along the Y-axis direction by a ball screw type Y-axis direction moving mechanism (not shown). Furthermore, the third holding table 62 is configured to be rotatable within a predetermined angle range around a rotation axis along the Z-axis direction.

[0079] A condenser 66 constituting a laser beam irradiation unit 64 is provided above the holding surface 62a. The laser beam irradiation unit 64 has a laser medium (not shown) such as an Nd:YAG crystal or an Nd:YVO4 crystal, an excitation light source (not shown) such as a lamp that irradiates the laser medium with excitation light, and a laser oscillator (not shown) including a Q switch (not shown).

[0080] The laser oscillator emits a pulsed laser beam L having a wavelength (e.g., 1064 nm) that is transmitted through the ingot 11. The laser beam L is reflected by a mirror (not shown) and then irradiated toward the holding surface 62a via a condenser lens (not shown) provided in the condenser 66.

[0081] A ball screw type Z-axis direction moving mechanism (not shown) is connected to the condenser 66, and the condenser 66 is movable along the Z-axis direction. By adjusting the position of the condenser 66 in the Z-axis direction, the position of the focal point P of the laser beam L in the Z-axis direction is adjusted.

[0082] Fig. 9 is a perspective view showing the separation layer forming step S30. Fig. 10(A) is a plan view showing the separation layer forming step S30, and Fig. 10(B) is a cross-sectional view of the ingot 11 on which the separation layer 15 is formed.

[0083] In the peeling layer forming process S30, the position of the focal point P of the laser beam L in the Z-axis direction is positioned at a predetermined depth 11f inside the ingot 11, which corresponds to the thickness (see FIG. 11(B)) of the wafer 13. That is, the distance from the first surface 11a to the predetermined depth 11f corresponds to the thickness of the wafer 13.

[0084] In this manner, the position of the focal point P is adjusted, and the focal point P and the ingot 11 are moved relatively along the X-axis direction. In this embodiment, the thickness direction 11c of the ingot 11 is disposed approximately parallel to the Z-axis direction, so that the X-axis direction corresponds to a predetermined direction perpendicular to the thickness direction 11c of the ingot 11.

[0085] When moving the focal point P and the ingot 11 relative to each other, for example, as shown in FIG. 9, with the position of the focal point P fixed, the third holding table 62 is moved along the processing feed direction (i.e., the X-axis direction) by the X-axis direction moving mechanism.

[0086] The X-axis direction moving mechanism moves the third holding table 62 so that the focal point P crosses the outer peripheral edge 11a1 of the first surface 11a along the processing feed direction, and then the Y-axis direction moving mechanism moves the third holding table 62 a predetermined amount along the indexing feed direction (i.e., the Y-axis direction).

[0087] Next, in the same manner, the third holding table 62 is moved so that the light condensing point P crosses the outer peripheral edge 11a1 of the first surface 11a along the processing feed direction. In this manner, the processing feed and the indexing feed are repeated in sequence.

[0088] As a result, a peeling layer 15 (see FIG. 10(A)) including a modified layer 15a and cracks 15b extending from the modified layer 15a is formed over substantially the entire surface (i.e., inside) of the ingot 11 to a predetermined depth 11f. The processing conditions in the peeling layer formation step S30 are set, for example, as follows.

[0089] Laser beam wavelength: 900nm to 1400nm Average power of laser beam: 2.5W Pulse frequency: 40kHz Machining feed speed: 240mm / s Indexing feed amount: 500μm

[0090] The modified layer 15a is a region of the ingot 11 having a crystallinity (e.g., amorphous) different from the crystallinity (single crystal) of the non-irradiated region that is not irradiated with the laser beam L, and is a region having reduced mechanical strength compared to the non-irradiated region.

[0091] As shown in Figure 10(A), crack 15b is formed along the indexing feed direction (perpendicular to thickness direction 11c and processing feed direction, respectively) starting from modified layer 15a, which is formed in a straight line when viewed in a plane of ingot 11.

[0092] After the separation layer forming step S30, a separation step S40 is performed in which the wafer 13 is separated from the ingot 11 starting from the separation layer 15. In the separation step S40, a separation device 70 is used. First, the separation device 70 will be described with reference to FIG. 11(A). The separation device 70 has a fourth holding table 72.

[0093] The fourth holding table 72 has approximately the same diameter and structure as the third holding table 62, and its upper surface functions as a approximately flat holding surface 72a that suction-holds the ingot 11. In other words, the fourth holding table 72 also functions as a chuck table.

[0094] A peeling unit 74 is provided above the fourth holding table 72. The peeling unit 74 has a cylindrical movable part 76 whose longitudinal side is disposed along the Z-axis direction. A Z-axis direction moving mechanism (not shown) is connected to the movable part 76.

[0095] The Z-axis direction moving mechanism is, for example, a ball screw type moving mechanism, but may be composed of other actuators. The Z-axis direction moving mechanism allows the movable part 76 to move along the Z-axis direction. A disk-shaped suction head 78 is provided at the bottom of the movable part 76.

[0096] The suction head 78 has a frame and a porous plate, similar to the fourth holding table 72. The lower surfaces of the frame and the porous plate are disposed substantially flush and substantially parallel to the XY plane, and function as a holding surface 78a.

[0097] 11(A) is a diagram showing a state at the start of the peeling step S40. In the peeling step S40, the ground second surface 11b is suction-held by the holding surface 72a, and the ground first surface 11a is suction-held by the holding surface 78a of the suction head 78. Next, the suction head 78 is raised.

[0098] As a result, the wafer 13 is peeled off from the ingot 11 starting from the peeling layer 15. Fig. 11(B) is a diagram showing a state at the end of the peeling step S40. Note that an external force may be applied to the ingot 11 before the suction head 78 is raised.

[0099] The external force is applied, for example, by an operator or by using a dedicated device to drive a wedge (not shown) into the side surface 11d of the ingot 11 at a height position of the peeling layer 15. At this time, it is preferable to drive the wedge into multiple places along the circumferential direction of the ingot 11, rather than just one place on the side surface 11d.

[0100] By applying an external force, the crack 15b can be further developed at the depth position where the peeling layer 15 is formed, making it easier to peel off the wafer 13. Instead of driving a wedge into the side surface 11d, an external force may be applied by applying ultrasonic waves (i.e., elastic vibration waves in a frequency band exceeding 20 kHz) to the ingot 11.

[0101] When applying ultrasonic waves, ultrasonic waves are applied to the exposed surface of the ingot 11 (first surface 11a after grinding in FIG. 11(A)) via liquid such as pure water before the ingot 11 is sucked and held by the suction head 78. Specifically, the liquid to which ultrasonic waves have been applied is sprayed from a nozzle onto the ingot 11, or ultrasonic waves are applied from a horn to the ingot 11 via the liquid.

[0102] When using a nozzle or a horn, an external force is first applied by using ultrasonic waves to a localized area of ​​the ground first surface 11a having a diameter of about 5 mm to 50 mm. Next, the nozzle or the horn and the fourth holding table 72 are moved relatively to apply the external force to other areas of the ground first surface 11a.

[0103] In this manner, by gradually moving the nozzle or horn, the external force is finally applied to the entire surface of the first surface 11a. As a result, the cracks 15b between the adjacent modified layers 15a are connected, so that the mechanical strength of the peeling layer 15 can be reduced compared to when the cracks 15b are not connected.

[0104] It is to be noted that both the nozzle and the horn may be used to apply an external force to the first surface 11a. When both the nozzle and the horn are used, the area to which the external force is applied to the first surface 11a may be different between the two, and the frequencies used may be different between the two.

[0105] If peeling of the wafer 13 is not continued after the peeling step S40 (NO in the judgment step S50), the flow is terminated. On the other hand, if peeling of the wafer 13 is continued (YES in the judgment step S50), the peeled surface 11g (see FIG. 11(B)) corresponding to the exposed surface of the ingot 11 after the peeling step S40 is ground using a grinding device 10, 50, etc., to reduce the surface roughness of the peeled surface 11g compared to immediately after the peeling step S40.

[0106] Next, returning to the separation layer formation step S30, a separation layer 15 is similarly formed using a laser processing device 60, and then the wafer 13 is similarly separated using a separation device 70. In this manner, a plurality of wafers 13 can be produced from one ingot 11.

[0107] In this embodiment, as described above, in the first grinding process S10, the second surface 11b of the warped ingot 11 is not held by suction on the first holding table 12, but the side surface 11d of the ingot 11 is clamped and fixed by two or more clamp portions 24, so that the ingot 11 is held on the first holding table 12 so that the first surface 11a is exposed, and the first surface 11a of the ingot 11 is ground.

[0108] Therefore, compared to grinding the first surface 11a in a state in which the ingot 11 is deformed by holding the second surface 11b by suction on the first holding table 12 so that the entire second surface 11b is in contact with the upper surface of the porous plate 16, it is possible to reduce warpage remaining on the first surface 11a when the ingot 11 is removed from the first holding table 12 after grinding. Therefore, it is possible to reduce warpage of the wafer 13 peeled off from the ingot 11 after the peeling layer forming step S30 and the peeling step S40.

[0109] (First Modification) Next, a first modification of the above-mentioned embodiment will be described with reference to Fig. 12(A). Fig. 12(A) is a flow diagram showing a first modification of the method for manufacturing the wafer 13. The first modification includes a planarization step S15 for reducing the surface roughness of the first surface 11a ground in the first grinding step S10.

[0110] The planarization step S15 is an optional step, and is performed before the release layer formation step S30 in order to reduce the surface roughness of the first surface 11a, for example, when the surface roughness of the first surface 11a after the first grinding step S10 is relatively large. Of course, the planarization step S15 may be omitted, as in the first embodiment.

[0111] In the planarization process S15, the surface roughness of the first surface 11a, which is the surface to be irradiated with the laser beam L in the peeling layer formation process S30, is reduced, for example, so that the arithmetic mean roughness Ra of the contour curve on the first surface 11a (see, for example, B0601:2013 of the Japanese Industrial Standards) is set to 20 nm or less.

[0112] This makes it easier for the laser beam L to enter the inside of the ingot 11 in the peeling layer formation step S30, compared to when Ra exceeds 20 nm. That is, compared to when Ra exceeds 20 nm, it is possible to reduce the energy of the laser beam L lost due to scattering at the first surface 11a and increase the energy of the laser beam L entering the inside of the ingot 11.

[0113] In the planarization step S15, a grinding wheel is used that includes a plurality of grinding stones, each of which has a smaller average grain size than the average grain size of the abrasive grains in each of the plurality of grinding stones 40b used in the first grinding step S10.

[0114] For example, the grinding wheel used in the planarizing step S15 is a so-called finish grinding wheel, and each of the finish grinding stones of the finish grinding wheel has a grain size of #2000 to #8000. The processing conditions in the planarizing step S15 may be the same as those in the first grinding step S10, or may be changed as appropriate.

[0115] (Second Modification) Next, a second modification of the above-mentioned embodiment will be described with reference to Fig. 12(B) Fig. 12(B) is a flow diagram showing a second modification of the method for manufacturing the wafer 13.

[0116] The second modified example includes a planarization step S25 for reducing the surface roughness of the second surface 11b ground in the second grinding step S20. The planarization step S25 is performed after the second grinding step S20 and before the release layer forming step S30.

[0117] In the planarization process S25, a grinding wheel (i.e., a finish grinding wheel) is used that includes multiple grinding wheels, each of which has a smaller average grain size than the average grain size of the abrasive grains in each of the multiple grinding wheels 40b used in the second grinding process S20.

[0118] In the flattening step S25, the arithmetic mean roughness Ra of the profile curve of second surface 11b is set to 20 nm or less. Then, in the peeling layer forming step S30, second surface 11b with reduced surface roughness is used as the surface to be irradiated with laser beam L.

[0119] In this manner, in the first and second modified examples, the surface roughness of one of the first surface 11a ground in the first grinding process S10 and the second surface 11b ground in the second grinding process S20, which will be the surface to be irradiated with the laser beam L in the peeling layer formation process S30, is reduced.

[0120] (Third Modification) Next, a third modification of the above-mentioned embodiment will be described with reference to Fig. 13. Fig. 13 is a diagram showing a second grinding step S20 in the third modification of the method for producing the wafer 13.

[0121] In the second grinding process S20 of the above-described embodiment, a second holding table 52 without a clamping jig 18 is used, whereas in the second grinding process S20 of the third modified example, a first holding table 12 having a clamping jig 18 is used, similar to the first grinding process S10.

[0122] However, in the second grinding step S20 of the third modified example, the first surface 11a after grinding is suction-held by the upper surface of the inner frame 14 and the upper surface of the porous plate 16. In this case, the side surface 11d of the ingot 11 is not fixed by the multiple clamp portions 24, but may be fixed.

[0123] In addition, the structures, methods, and the like according to the above-described embodiments and modifications can be appropriately modified without departing from the scope of the present invention. Unlike the above-described embodiment, the first surface 11a may have a concave shape and the second surface 11b may have a convex shape.

[0124] In the first grinding process S10, the grinding itself can be performed even if one of the first surface 11a and the second surface 11b that is concave is exposed upward and the other of the first surface 11a and the second surface 11b that is convex is positioned downward.

[0125] However, when the convex side faces the first holding table 12, the center of the other side comes into contact with the porous plate 16, so that the ingot 11 is more likely to move during grinding and the flatness of the ground surface decreases compared to when the concave side faces the first holding table 12. Therefore, it is preferable to face the concave side downwards to face the first holding table 12.

[0126] Although the grinding apparatuses 10 and 50 are different apparatuses, the grinding apparatus 10 may be used as the grinding apparatus 50 by removing the clamp jig 18 from the first holding table 12 of the grinding apparatus 10.

[0127] In addition, in the release layer formation process S30, the first surface 11a after grinding is exposed, but the second surface 11b after grinding may be exposed to form a release layer 15 at a predetermined depth 11f from the second surface 11b after grinding.

[0128] Each of the first holding table 12, the second holding table 52, the third holding table 62 and the fourth holding table 72 may be an electrostatic chuck that attracts and holds the ingot 11 by electrostatic force instead of negative pressure, or may attract and hold the ingot 11 by a mechanism other than negative pressure or electrostatic force. [Explanation of symbols]

[0129] 10,50: Grinding device 11: Ingot (workpiece) 11a: first surface, 11a1: outer periphery 11b: second surface, 11b1: outer periphery 11c: thickness direction, 11d: side, 11d1: notch, 11e1, 11e2: warpage 11f: depth, 11g: peeled surface 12: first holding table, 12a: rotating shaft, 14: inner frame, 16: porous plate 13: Wafer 15: peeling layer, 15a: modified layer, 15b: crack, 17: gap 18: Clamp jig 20: outer frame body, 22: bolt, 22a: head, 22b: shaft 24: Clamp part (fixing part) 26: base, 28: bolt, 28a: head, 28b: shaft 30: movable member, 32: pressing plate 34: grinding unit, 36: spindle, 38: mount 40: grinding wheel, 40a: wheel base, 40b: grinding stone 52: second holding table, 52a: holding surface 60: laser processing device, 62: third holding table, 62a: holding surface 64: laser beam irradiation unit, 66: condenser 70: peeling device, 72: fourth holding table, 72a: holding surface 74: peeling unit, 76: movable part, 78: suction head, 78a: holding surface L: laser beam, P: focal point S10: First grinding process, S15, S25: Flattening process S20: second grinding step, S30: peeling layer forming step S40: Peeling process, S50: Judgment process, S60: Peeling surface grinding process

Claims

1. A method for manufacturing a wafer from a workpiece having a first surface, a second surface located opposite to the first surface, and a side surface connected to an outer periphery of the first surface and an outer periphery of the second surface, the workpiece including warpage on the first surface and the second surface, a first grinding step of grinding the first surface of the workpiece in a state in which the workpiece is held by the first holding table such that the second surface faces the first holding table and the first surface is exposed by clamping and fixing the side surface of the workpiece with two or more fixing members of the first holding table without holding the second surface by suction with the first holding table; a second grinding step of grinding the second surface of the workpiece after the first grinding step; a peeling layer forming step in which, after the second grinding step, a focal point of a laser beam having a wavelength that transmits the workpiece is positioned at a predetermined depth inside the workpiece corresponding to a thickness of a wafer, and the focal point and the workpiece are moved relatively along a predetermined direction perpendicular to the thickness direction of the workpiece to form a peeling layer inside the workpiece, the peeling layer including a modified layer and cracks extending from the modified layer; a peeling step of peeling the wafer from the workpiece starting from the peeling layer after the peeling layer forming step; A wafer manufacturing method comprising:

2. 2. The method for manufacturing a wafer as described in claim 1, wherein in the second grinding process, the first surface ground in the first grinding process is suction-held by a holding surface of a second holding table so that the second surface is exposed.

3. the first surface exposed in the first grinding step has a convex warp, 3. The method for producing a wafer according to claim 1, wherein in the first grinding step, the first surface is ground to reduce warpage of the first surface.

4. 3. The method for manufacturing a wafer as described in claim 1 or 2, further comprising, prior to the peeling layer formation step, a planarization step for reducing surface roughness of one of the first surface ground in the first grinding step and the second surface ground in the second grinding step, which is to be irradiated with the laser beam in the peeling layer formation step.

Images