The present invention relates to a method of manufacturing a monocrystalline silicon substrate from a workpiece of monocrystalline silicon that has been manufactured such that a particular crystal plane included in crystal planes {100} is exposed on face and reverse sides thereof.
Semiconductor device chips are typically manufactured from a disk-shaped monocrystalline silicon substrate, hereinafter also referred to as a “substrate.” The substrate is sliced from a cylindrical ingot of monocrystalline silicon, hereinafter also referred to as an “ingot,” with use of a wire saw (see, for example, Japanese Patent Laid-open No. H09-262826A).
Saw kerfs that must be taken into account when substrates are sliced from ingots with a wire saw are comparatively large as they are approximately 300 μm wide each. In addition, the sliced substrates leave minute surface irregularities on their surfaces, and are likely to be curved or warped as a whole. Therefore, the surfaces of the sliced substrates need to be lapped, etched, and/or polished to a flat smooth finish.
After an ingot has been sliced into substrates and the substrates have been finished, the amount of monocrystalline silicon that is eventually left in the substrates is approximately ⅔ of the overall amount of monocrystalline silicon of the ingot. In other words, approximately ⅓ of the overall amount of monocrystalline silicon of the ingot turns into sawdust to be disposed of in slicing and planarizing steps. For this reason, the productivity of the process of manufacturing substrates from ingots with use of a wire saw is low.
It is therefore an object of the present invention to provide a method of manufacturing a monocrystalline silicon substrate with high productivity.
In accordance with an aspect of the present invention, there is provided a method of manufacturing a monocrystalline silicon substrate from a workpiece of monocrystalline silicon that has been manufactured such that a particular crystal plane included in crystal planes {100} is exposed on face and reverse sides thereof. The method includes a peel-off layer forming step of moving a focused spot, positioned within the workpiece, of a laser beam that is applied to the workpiece and that has a wavelength transmittable through the monocrystalline silicon and the workpiece relatively to each other along a first direction that is parallel to the particular crystal plane and that is inclined to a particular crystal orientation included in crystal orientations <100> by an angle of 5° or less, thereby forming a peel-off layer in a straight area in the workpiece along the first direction, an indexing-feed step of moving a position where the focused spot of the laser beam is formed and the workpiece relatively to each other along a second direction parallel to the particular crystal plane and perpendicular to the first direction, and a separating step of separating a substrate from the workpiece along peel-off layers in the workpiece that act as separation initiating points, after the peel-off layer forming step and the indexing-feed step have been repeated. The peel-off layer forming step includes a step of moving the focused spot and the workpiece relatively to each other in order to move the focused spot from an inside of the workpiece toward an outside of the workpiece.
According to the present invention, after peel-off layers have been formed in a workpiece of monocrystalline silicon with use of a laser beam having a wavelength transmittable through monocrystalline silicon, a substrate is separated from the workpiece along the peel-off layers acting as separation initiating points. The process results in increased productivity for the manufacture of substrates, compared with a process of manufacturing substrates from a workpiece with use of a wire saw.
The above and other objects, features and advantages of the present invention and the manner of realizing them will become more apparent, and the invention itself will best be understood from a study of the following description and an appended claim with reference to the attached drawings showing a preferred embodiment of the invention.
A preferred embodiment of the present invention will be described below with reference to the accompanying drawings.
The ingot, denoted by 11, illustrated in
Although the ingot 11 is manufactured such that the crystal plane (100) is exposed on each of the face side 11a and the reverse side 11b, a plane slightly inclined to the crystal plane (100) may be exposed on each the face side 11a and the reverse side 11b due to processing errors, etc., that may have occurred in the manufacturing process. Specifically, a plane that is inclined to the crystal plane (100) by an angle of 1° or less may be exposed on each of the face side 11a and the reverse side 11b. In other words, the crystal axis of the ingot 11 may extend along a direction that is inclined to the crystal orientation [100] by an angle of 1° or less.
The ingot 11 has an orientation flat 13 defined on a side surface 11c thereof. The ingot 11 has a center C positioned on a particular crystal orientation, given herein as a crystal orientation [011] for the sake of convenience, included in crystal orientations <110> as viewed from the orientation flat 13. A crystal plane (011) of monocrystalline silicon is exposed on the orientation flat 13.
The X-axis directions and the Y-axis directions extend perpendicularly to each other on a horizontal plane. A +Z-axis direction and a −Z-axis direction illustrated in
The laser processing apparatus, denoted by 2 in
The porous plate is fluidly connected to a suction source, not illustrated, such as a vacuum pump, through a fluid channel or the like, not illustrated, defined in the holding table 4. When the suction source is actuated, it generates a negative pressure through the fluid channel to a space in the vicinity of the holding surface of the holding table 4, thereby holding the ingot 11 under suction on the holding surface.
The laser processing apparatus 2 also has a laser beam applying unit 6 disposed above the holding table 4. The laser beam applying unit 6 includes a laser oscillator 8 that has a laser medium of neodymium-doped yttrium aluminum garnet (Nd:YAG) or the like and that emits a pulsed laser beam LB having a wavelength of 1064 nm, for example, transmittable through monocrystalline silicon.
The laser beam LB has its output power level adjusted by an attenuator 10 and is then applied to a spatial optical modulator 12. The spatial optical modulator 12 branches the adjusted laser beam LB into a plurality of laser beams LB. Specifically, the spatial optical modulator 12 branches the adjusted laser beam LB into a plurality of, e.g., five, laser beams LB that will have respective focused spots in the ingot 11 arrayed at equal intervals along the Y-axis after being emitted from an emission head 16 to be described below.
The laser beams LB emitted from the spatial optical modulator 12 are applied to and reflected by a mirror 14 to travel to the emission head 16. The emission head 16 houses therein a condensing lens, not illustrated, for converging the laser beams LB. The laser beams LB converged by the condensing lens are emitted toward the holding surface of the holding table 4.
The emission head 16 of the laser beam applying unit 6 is coupled to a moving mechanism, not illustrated. The moving mechanism includes ball screws, etc., and moves the emission head 16 in the X-axis directions, the Y-axis directions, and/or the Z-axis directions. The moving mechanism is actuated to adjust the positions or coordinates of the focused spots of the laser beams LB emitted from the emission head 16 in the X-axis directions, the Y-axis directions, and/or the Z-axis directions.
For forming a peel-off layer in the ingot 11 fully thereacross on the laser processing apparatus 2, the ingot 11 with its face side 11a facing upwardly is placed and held on the holding table 4.
The ingot 11 is held on the holding table 4 such that the direction from the orientation flat 13 toward the center C of the ingot 11, i.e., the crystal orientation [011], is inclined to the +X-axis direction and the +Y-axis direction by an angle of 45°. For example, the ingot 11 is held on the holding table 4 such that the ingot 11 has a crystal orientation [010] directed along the +X-axis direction and a crystal orientation [001] directed along the +Y-axis direction.
Then, in order to form a peel-off layer in a straight area, in the ingot 11, that extends along the X-axis at an end of the ingot 11 along the Y-axis, i.e., in the −Y-axis direction, the moving mechanism moves the emission head 16 to a position where the laser beams LB start being applied to the ingot 11, i.e., a first emission starting position. The first emission starting position is a position where, when the laser beams LB are emitted from the emission head 16 to the ingot 11, they form their respective focused spots in the ingot 11 radially inwardly of the side surface 11c thereof, i.e., a position closer to the center C than the side surface 11c.
Then, a peel-off layer is formed in the straight area, in the ingot 11, that extends along the +X-axis direction along the crystal orientation [010] (peel-off layer forming step S1). In the peel-off layer forming step S1, the laser beams LB are applied to the ingot 11 with their focused spots positioned in the ingot 11 while the emission head 16 is being moved in the −X-axis direction.
As described above, the laser beams LB that have been branched by the spatial optical modulator 12 have their focused spots, i.e., five focused spots, arrayed at equal intervals along the Y-axis.
The laser beams LB thus applied to the ingot 11 form modified regions 15a where the crystalline structure of monocrystalline silicon is disrupted respectively around the focused spots in the ingot 11. The modified regions 15a are arrayed along the Y-axis.
At this time, cracks 15b are developed along predetermined crystal planes from the respective modified regions 15a. The modified regions 15a and the cracks 15b developed therefrom jointly make up a peel-off layer 15 in the ingot 11.
In general, monocrystalline silicon is most likely to cleave along a particular crystal plane included in crystal planes {111}, and is second most likely to cleave along a particular crystal plane included in crystal planes {110}. Therefore, when modified regions are formed along a particular crystal orientation, e.g., a crystal orientation [011], included in crystal orientations <110> of monocrystalline silicon of an ingot, for example, many cracks are developed from the modified regions along the particular crystal plane included in the crystal planes {111}.
On the other hand, when a plurality of modified regions are formed in a straight area along a particular crystal orientation included in crystal orientations <100> of monocrystalline silicon such that the modified regions are arrayed along a direction perpendicular to the direction in which the straight area extends, as viewed in plan, many cracks are developed from the modified regions along a crystal plane parallel to the direction in which the straight area extends, among crystal planes {N10} (N represents a natural number of 10 or less).
For example, when the modified regions 15a arrayed along the crystal orientation [001], i.e., the +Y-axis direction, are formed in the straight area along the crystal orientation [010], i.e., the +X-axis direction, many cracks are developed from the modified regions 15a along a crystal plane parallel to the crystal orientation [010] among the crystal planes {N10} (N represents a natural number 10 or less).
Specifically, when the modified regions 15a are thus formed, cracks are likely to develop along the following crystal planes.
(101),(201),(301),(401),(501),(601),(701),(801),(901),(1001) [Math. 1]
(
The angle that the crystal plane (100) exposed on the face side 11a and the reverse side 11b of the ingot 11 forms with the crystal plane parallel to the crystal orientation [010] among the crystal planes {N10} is 45° or less. On the other hand, the angle that the crystal plane (100) forms with the particular crystal plane included in the crystal planes {111} is approximately 54.7°.
Therefore, when the laser beams LB are applied to the ingot 11 along the crystal orientation
(former case), the peel-off layer 15 tends to be wider and thinner than when the laser beams LB are applied to the ingot 11 along the crystal orientation
(latter case). In other words, the value (W/T) of the ratio of the width (W) to the thickness (T) of the peel-off layer 15 illustrated in
Further, the laser beams LB are applied to the ingot 11 while the emission head 16 is being moved in the −X-axis direction until the emission head 16 reaches a position where the laser beams LB finish being applied from the emission head 16 to the ingot 11, i.e., a first emission ending position. The first emission ending position is a position where, when the laser beams LB are emitted from the emission head 16 to the ingot 11, they form their respective focused spots radially outwardly of the side surface 11c of the ingot 11.
When the laser beams LB are applied to an area of the ingot 11 in the vicinity of the side surface 11c thereof, the power of the laser beams LB becomes unstable at their focused spots. In this case, the positions of the focused spots of the laser beams LB that have passed through the face side 11a of the ingot 11 and the positions of the focused spots of the laser beams LB that have passed through the side surface 11c of the ingot 11 deviate from each other due to the difference between the refractive index of the ingot 11 and the refractive index of the atmosphere.
In other words, the focused spots of the laser beams LB emitted from the emission head 16 are not fixed in position. Consequently, modified regions 15a may not sufficiently be formed in the area of the ingot 11 in the vicinity of the side surface 11c thereof. If modified regions 15a are not formed, then cracks 15b are not developed in the ingot 11. As a result, no peel-off layer 15 may possibly be formed in the area of the ingot 11 in the vicinity of the side surface 11c thereof.
On the other hand, providing that cracks 15b have been formed radially inwardly of the area of the ingot 11 in the vicinity of the side surface 11c thereof, cracks 15b tend to be developed toward the side surface 11c in order to release stresses that have been produced in the ingot 11 due to the cracks 15b formed therein. Therefore, it is preferable to apply the laser beams LB to the area of the ingot 11 with the cracks 15b having been formed radially inwardly of the area.
Specifically, the laser beams LB should preferably be applied to the ingot 11 while the focused spots thereof are being moved from the inside of the ingot 11 toward the outside of the ingot 11. The laser beams LB thus applied make it easier to form a peel-off layer 15 in the area of the ingot 11 in the vicinity of the side surface 11c thereof than when the laser beams LB are applied to the ingot 11 while the focused spots thereof are being moved from the outside of the ingot 11 toward the inside of the ingot 11.
Then, the emission head 16 is moved in the +X-axis direction back to the first emission starting position.
Then, the laser beams LB are applied to the ingot 11 while the emission head 16 is being moved in the +X-axis direction.
The area, of the ingot 11, that is irradiated with the laser beams LB upon movement of the emission head 16 in the +X-axis direction may overlap the area of the ingot 11 in which area the peel-off layer 15 has already been formed. Specifically, the emission head 16 may be moved back to a position that is slightly spaced from the first emission starting position in the −X-axis direction, and the laser beams LB may be applied to the ingot 11 while the emission head 16 is being moved in the +X-axis direction from the spaced position.
The laser beams LB are applied to the ingot 11 while the emission head 16 is being moved in the +X-axis direction until the emission head 16 reaches a position where the laser beams LB finish being applied from the emission head 16 to the ingot 11, i.e., a second emission ending position. The second emission ending position is a position where, when the laser beams LB are emitted from the emission head 16 to the ingot 11, they form their respective focused spots radially outwardly of the side surface 11c of the ingot 11.
Then, the position where the focused spots of the laser beams LB are formed in the ingot 11 and the ingot 11 are relatively moved in the +Y-axis direction, i.e., the crystal orientation [001] (indexing-feed step S2). In the indexing-feed step S2, the moving mechanism moves the emission head 16 to a position where the laser beams LB start being applied to the ingot 11, i.e., a second emission starting position, in order to form a peel-off layer 15 in a straight area, in the ingot 11, that lies parallel to the straight area in which the peel-off layer 15 has already been formed.
When the emission head 16 returns to the first emission starting position, the emission head 16 may apply the laser beams LB to the ingot 11 to increase the density of the modified regions 15a and the cracks 15b contained in the peel-off layer 15 that has already been formed in the ingot 11.
The indexed distance that the emission head 16 repeatedly moves along the Y-axis is set to a value equal to or larger than the width (W) of the peel-off layer 15, for example. Specifically, if the width (W) of the peel-off layer 15 is in the range of 250 to 280 μm, then the index distance is set to approximately 530 μm.
Then, the peel-off layer forming step S1 is carried out again. Specifically, the laser beams LB are applied to the ingot 11 with their focused spots positioned in the ingot 11 while the emission head 16 is being moved in the −X-axis direction.
Further, the laser beams LB are applied to the ingot 11 while the emission head 16 is being moved in the −X-axis direction until the emission head 16 reaches a position where the laser beams LB finish being applied from the emission head 16 to the ingot 11, i.e., a third emission ending position. The third emission ending position is a position where, when the laser beams LB are emitted from the emission head 16 to the ingot 11, they form their respective focused spots slightly radially outwardly of the side surface 11c of the ingot 11.
Then, the emission head 16 is moved in the +X-axis direction back to the second emission starting position. At this time, the emission head 16 may return to the second emission starting position while applying the laser beams LB to the ingot 11 to increase the density of the modified regions 15a and the cracks 15b contained in the peel-off layer 15, i.e., 15-2, that has already been formed in the ingot 11.
Then, the laser beams LB are applied to the ingot 11 while the emission head 16 is being moved in the +X-axis direction. The laser beams LB thus applied form a new peel-off layer 15, e.g., 15-2, in the ingot 11, that extends in the +X-axis direction from the peel-off layer 15, i.e., 15-2, already formed in the ingot 11.
The area, of the ingot 11, that is irradiated with the laser beams LB upon movement of the emission head 16 in the +X-axis direction may overlap the area of the ingot 11 in which area the peel-off layer 15, i.e., 15-2, has already been formed. Specifically, the emission head 16 may be moved back to a position that is slightly spaced from the second emission starting position in the −X-axis direction, and the laser beams LB may be applied to the ingot 11 while the emission head 16 is being moved in the +X-axis direction from the spaced position.
Further, the laser beams LB are applied to the ingot 11 while the emission head 16 is being moved in the +X-axis direction until the emission head 16 reaches a position where the laser beams LB finish being applied from the emission head 16 to the ingot 11, i.e., a fourth emission ending position. The fourth emission ending position is a position where, when the laser beams LB are emitted from the emission head 16 to the ingot 11, they form their respective focused spots radially outwardly of the side surface 11c of the ingot 11.
The laser beams LB thus applied to the ingot 11 so far form the peel-off layer 15, i.e., 15-2, in the straight area, in the ingot 11, that extends along the X-axis direction, i.e., the crystal orientation.
The peel-off layer 15-2 is closer to the center C of the ingot 11 than the initially formed peel-off layer 15-1 is and is longer than the peel-off layer 15-1 along the X-axis.
The indexing-feed step S2 and the peel-off layer forming step S1 are repeatedly carried out until a peel-off layer 15 is formed in a straight area, in the ingot 11, that extends along the X-axis at an opposite end of the ingot 11 along the Y-axis. If peel-off layers 15 have been formed fully in the ingot 11 from the area at the end in the −Y-axis direction of the ingot 11 to the area at the opposite end in the +Y-axis direction of the ingot 11 (step S3: YES), then a substrate is separated from the ingot 11 along the peel-off layers 15 that act as separation initiating points (separating step S4).
The holding table 20 has a circular upper surface as a holding surface where a porous plate, not illustrated, is exposed. The porous plate is fluidly connected to a suction source, not illustrated, such as a vacuum pump, through a fluid channel or the like, not illustrated, defined in the holding table 20. When the suction source is actuated, it generates a negative pressure through the fluid channel to a space in the vicinity of the holding surface of the holding table 20, thereby holding the ingot 11 under suction on the holding surface.
A separating unit 22 is disposed above the holding table 20. The separating unit 22 has a cylindrical support member 24 having an upper portion to which there are coupled a ball-screw-type lifting and lowering mechanism, not illustrated, and a rotary actuator, not illustrated, such as an electric motor, for example. When the ball-screw-type lifting and lowering mechanism is actuated, it selectively lifts and lowers the separating unit 22. When the rotary actuator is actuated, it rotates the support member 24 about a straight rotational axis passing through the center of the support member 24 and extending perpendicularly to the holding surface of the holding table 20.
The support member 24 has a lower end fixed centrally to an upper portion of a disk-shaped base 26. A plurality of movable fingers 28 are mounted on a lower surface of an outer circumferential portion of the base 26 and angularly spaced at generally equal intervals circumferentially around the base 26. The movable fingers 28 have respective plate-shaped vertical portions 28a extending downwardly from the lower surface of the base 26.
The vertical portions 28a have respective upper ends coupled to actuators such as air cylinders, not illustrated, housed in the base 26. When the actuators are actuated, they move the movable fingers 28 in radial directions of the base 26. The movable fingers 28 also include respective plate-shaped wedges 28b extending radially inwardly from respectively inner sides of a lower end portion of the vertical portions 28a. The wedges 28b are tapered in such a manner as to be progressively thinner toward their pointed distal ends.
The separating apparatus 18 operates to carry out the separating step S4 according to the following sequence of events. First, the ingot 11 is placed on the holding table 20 such that the center of the reverse side 1b of the ingot 11 with the peel-off layers 15 formed therein and the center of the holding surface of the holding table 20 are aligned with each other.
Then, the suction source fluidly connected to the porous plate exposed on the holding surface is actuated to hold the ingot 11 under suction on the holding table 20. Thereafter, the actuators coupled to the movable members 28 is actuated to position the movable members 28 on a radially outer portion of the base 26.
Next, the lifting and lowering mechanism is operated to position the pointed distal ends of the wedges 28b of the respective movable members 28 at a height horizontally aligned with the peel-off layers 15 in the ingot 11. Then, the actuators are operated to drive the wedges 28b into the side surface 11c of the ingot 11 (see
Then, the lifting and lowering mechanism is operated to lift the wedges 28b (see
If the portion of the ingot 11 closer to the face side 11a of the ingot 11 and the remaining portion of the ingot 11 closer to the reverse side 11b of the ingot 11 are separated from each other at the time when the wedges 28b are driven into the side surface 11c of the ingot 11, then the wedges 28b may not be rotated. The actuators and the rotary actuator may be operated simultaneously to drive the rotating wedges 28b into the side surface 11c of the ingot 11.
In the method of manufacturing the monocrystalline silicon substrate 17 from the ingot 11 according to the present embodiment, after peel-off layers 15 have been formed in the ingot 11 by the laser beams LB whose wavelength is transmittable through monocrystalline silicon, the substrate 17 is separated from the ingot 11 along the peel-off layers 15 that act as separation initiating points. The method is effective to reduce the amount of monocrystalline silicon to be disposed of in the manufacture of the substrate 17 from the ingot 11, resulting in increased productivity for the manufacture of the substrate 17, compared with a process of manufacturing the substrate 17 from the ingot 11 with use of a wire saw.
In the method described above, a plurality of modified regions 15a are formed in an array along the crystal orientation [001], i.e., the +Y-axis direction, in a straight area along the crystal orientation [010], i.e., the +X-axis direction. In this case, many cracks are developed from the respective modified regions 15a along a crystal plane parallel to the crystal orientation [010] among crystal planes {N10} (N represents a natural number 10 or less).
According to the method described above, therefore, the peel-off layer 15 tends to be wider and thinner than when the laser beams LB are applied to the ingot 11 along the crystal orientation [011]. As a result, it is possible to further reduce the amount of monocrystalline silicon to be disposed of in the manufacture of the substrate 17 from the ingot 11, resulting in increased productivity for the manufacture of the substrate 17.
In the method described above, further, the peel-off layer 15 is formed in the ingot 11 by moving the focused spots of the laser beams LB from the inside of the ingot 11 toward the outside of the ingot 11. According to the method described above, therefore, the peel-off layer 15 is formed to a sufficient extent in the area of the ingot 11 in the vicinity of the side surface 11c. As a result, it is easy to separate the substrate 17 from the ingot 11 in the separating step S4.
The method of manufacturing a monocrystalline silicon substrate as described above represents an aspect of the present invention, and the present invention is not limited to the above described method. An ingot used to manufacture a substrate according to the present invention is not limited to the ingot 11 illustrated in
Specifically, according to the present invention, a substrate may be manufactured from an ingot having a notch defined in a side surface thereof. Alternatively, according to the present invention, a substrate may be manufactured from an ingot that is free of an orientation flat and a notch in a side surface thereof.
The structure of a laser processing apparatus that can be used in the present invention is not limited to the structure of the laser processing apparatus 2 described above. According to the present invention, the method of manufacturing a monocrystalline silicon substrate may be carried out using a laser processing apparatus including a moving mechanism for moving the holding table 4 in the X-axis directions, the Y-axis directions, and/or the Z-axis directions.
Specifically, a laser processing apparatus that can be used in the present invention is not limited to any structural details insofar as the holding table 4 for holding the ingot 11 thereon and the emission head 16 of the laser beam applying unit 6 for applying the laser beams LB to the ingot 11 can be moved relatively to each other along the X-axis, the Y-axis, and the Z-axis.
Moreover, in the peel-off layer forming step S1 according to the present invention, the straight area, in the ingot 11, that is to be irradiated with the laser beams LB is not limited to a straight area along the crystal orientation [010]. According to the present invention, the laser beams LB may be applied to a straight area along the crystal orientation [001].
Specifically, when the laser beams LB are applied to the ingot 11 in the manner described above, cracks are likely to develop along the following crystal planes.
(110),(210),(310),(410),(510),(610),(710),(810),(910),(1010) [Math. 3]
(
According to the present invention, further, the laser beams LB may be applied to a straight area, in the ingot 11, that extends along a direction slightly inclined to the crystal orientation [010] or the crystal orientation [001]. This alternative will be described below with reference to
Specifically, in a case where the horizontal axis of the graph represents an angle of 45°, a straight area along the crystal orientation [001] becomes a measurement object. Similarly, in a case where the horizontal axis of the graph represents an angle of 135°, a straight area along the crystal orientation [010] becomes a measurement object. The vertical axis of the graph represents values obtained by dividing the widths of peel-off layers formed in measurement areas by applying the laser beams LB to the measurement areas by the width of peel-off layer formed in reference area by applying the laser beams LB to the reference area.
As illustrated in
In the peel-off layer forming step S1 according to the present invention, consequently, the laser beams LB may be applied to straight areas along a direction inclined at 5° or less, as viewed in plan, to the crystal orientation [001] or the crystal orientation.
In the peel-off layer forming step S1 according to the present invention, more specifically, the laser beams LB may be applied to straight areas along a direction, i.e., a first direction, parallel to the crystal plane, e.g., the crystal plane (100), that is exposed on the face side 11a and the reverse side 11b of the ingot 11, among particular crystal planes included in the crystal planes {100} and inclined to a particular crystal orientation, e.g., the crystal orientation [001] or the crystal orientation [010], included in the crystal orientations <100> at an angle of 5° or less.
When the peel-off layer forming step S1 is carried out, the indexing-feed step S2 is carried out by moving positions, in the ingot 11, where the focused spots are formed by the applied laser beams LB along a direction, i.e., a second direction, parallel to the crystal plane, e.g., the crystal plane (100), that is exposed on the face side 11a and the reverse side 11b of the ingot 11, among particular crystal planes included in the crystal planes {100} and perpendicular to the first direction, and the ingot 11 relatively to each other.
According to the present invention, moreover, after peel-off layers 15 have been formed fully in the ingot 11 from the area at the end in the −Y-axis direction of the ingot 11 to the area at the opposite end in the +Y-axis direction of the ingot 11 (step S3: YES), the peel-off layer forming step S1 and the indexing-feed step S2 may be repeatedly carried out. In other words, the laser beams LB for forming peel-off layers 15 may be applied again to the ingot 11 from the area at the end to the area at the opposite end in the Y-axis directions within the ingot 11 in which peel-off layers 15 have already been formed.
According to the present invention, further, after the peel-off layer forming step S1 but before the indexing-feed step S2, the peel-off layer forming step S1 may be carried out again. In other words, the laser beams LB for forming peel-off layers 15 may be applied to a straight area in the ingot 11 in which straight area peel-off layers 15 have already been formed.
When the peel-off layer forming step S1 is performed on an area of the ingot 11 in which area peel-off layers 15 have already been formed, the density of the modified regions 15a and the cracks 15b contained in the peel-off layer 15 that has already been formed in the ingot 11 is increased, thereby making it easier to separate the substrate 17 from the ingot 11 in the separating step S4.
In this case, further, the cracks 15b contained in the peel-off layers 15 are further developed to increase the widths, i.e., the width (W) illustrated in
According to the present invention, moreover, forming peel-off layers 15 fully in the ingot 11 in the peel-off layer forming step S1 is not an essential feature of the invention. Providing that the separating step S4 carried out using the separating apparatus 18, for example, makes it possible to develop cracks 15b in an area in the vicinity of the side surface 11c of the ingot 11, peel-off layers 15 may not be formed partly or wholly in the area in the vicinity of the side surface 11c of the ingot 11 in the peel-off layer forming step S1.
The separating step S4 according to the present invention may be carried out by an apparatus other than the separating apparatus 18 illustrated in
The holding table 32 has a circular upper surface as a holding surface where a porous plate, not illustrated, is exposed. The porous plate is fluidly connected to a suction source, not illustrated, such as a vacuum pump, through a fluid channel or the like, not illustrated, defined in the holding table 32. When the suction source is actuated, it generates a negative pressure through the fluid channel to a space in the vicinity of the holding surface of the holding table 32, thereby holding the ingot 11 under suction on the holding surface.
A separating unit 34 is disposed above the holding table 32. The separating unit 34 has a cylindrical support member 36 having an upper portion to which there is coupled a ball-screw-type lifting and lowering mechanism, not illustrated, for example. When the ball-screw-type lifting and lowering mechanism is actuated, it selectively lifts and lowers the separating unit 34.
The support member 36 has a lower end fixed centrally to an upper portion of a disk-shaped suction plate 38. The suction plate 38 has a plurality of suction ports defined in a lower surface thereof and fluidly connected to a suction source, not illustrated, such as a vacuum pump, through a fluid channel or the like, not illustrated, defined in the suction plate 38. When the suction source is actuated, it generates a negative pressure through the fluid channel to a space in the vicinity of the lower surface of the suction plate 38, thereby attracting the ingot 11 under suction to the lower surface of the suction plate 38.
The separating apparatus 30 operates to carry out the separating step S4 according to the following sequence of events. First, the ingot 11 is placed on the holding table 32 such that the center of the reverse side 11b of the ingot 11 with the peel-off layers 15 formed therein and the center of the holding surface of the holding table 32 are aligned with each other.
Then, the suction source fluidly connected to the porous plate exposed on the holding surface is actuated to hold the ingot 11 under suction on the holding table 32. Thereafter, the lifting and lowering mechanism is operated to lower the separating unit 34 to bring the lower surface of the suction plate 38 into contact with the face side 11a of the ingot 11.
Then, the suction source fluidly connected to the suction ports in the suction plate 38 is actuated to attract the face side 11a of the ingot 11 under suction to the lower surface of the suction plate 38 (see
At this time, upward forces are exerted on the portion of the ingot 11 closer to the face side 11a, of the ingot 11, that is attracted under suction to the suction plate 38 through the suction ports. As a result, the cracks 18b contained in the peel-off layers 15 are further developed, separating the portion of the ingot 11 closer to the face side 11a of the ingot 11 and the portion of the ingot 11 closer to the reverse side 11b of the ingot 11 from each other. In other words, a substrate 17 is manufactured from the ingot 11 along the peel-off layers 15 that act as separation initiating points.
According to the present invention, in the separating step S4, ultrasonic waves may be applied to the face side 11a of the ingot 11 prior to separating the portion of the ingot 11 closer to the face side 11a of the ingot 11 and the portion of the ingot 11 closer to the reverse side 11b of the ingot 11 from each other. In this case, inasmuch as the cracks 23 contained in the peel-off layers 15 are further developed by the applied ultrasonic waves, the portion of the ingot 11 closer to the face side 11a of the ingot 11 and the portion of the ingot 11 closer to the reverse side 11b of the ingot 11 can be separated more easily from each other.
According to the present invention, moreover, prior to the peel-off layer forming step S1, the face side 11a of the ingot 11 may be planarized by grinding or polishing (planarizing step). The planarizing step may be carried out when a plurality of substrates are manufactured from the ingot 11. Specifically, when a substrate 17 is manufactured by being separated from the ingot 11 along the peel-off layers 15, the newly exposed surface of the ingot 11 has surface irregularities reflecting a distribution of modified regions 15a and cracks 15b contained in the peel-off layers 15.
Consequently, when a new substrate is to be manufactured from the ingot 11, it is preferable to planarize the surface of the ingot 11 prior to the peel-off layer forming step S1. The planarized surface of the ingot 11 reduces irregular reflections of the laser beams LB applied to the ingot 11 in the peel-off layer forming step S1. According to the present invention, the newly exposed surface of the substrate 17 that has been separated from the ingot 11 along the peel-off layers 15 may also be planarized by grinding or polishing.
According to the present invention, further, a substrate may be manufactured from a bare wafer as a workpiece that is made of monocrystalline silicon that has been manufactured such that a particular crystal plane included in crystal planes {100} is exposed on face and reverse sides thereof.
The bare wafer is twice to five times thicker than the substrate to be manufactured therefrom, for example. The bare wafer is manufactured by being separated from the ingot 11 according to the same process as the method described above. It can thus be phrased that the substrate is manufactured from the ingot 11 by repeating the above method twice.
According to the present invention, moreover, a substrate may be manufactured from a device wafer as a workpiece that is fabricated from the above bare wafer with semiconductor devices formed thereon.
The structure, method, etc., according to the above embodiment may be changed or modified appropriately without departing from the scope of the present invention.
The present invention is not limited to the details of the above described preferred embodiment. The scope of the invention is defined by the appended claim and all changes and modifications as fall within the equivalence of the scope of the claim are therefore to be embraced by the invention.
Number | Date | Country | Kind |
---|---|---|---|
2021-169244 | Oct 2021 | JP | national |