Wafer manufacturing method

Abstract

Provided is a wafer manufacturing method for obtaining a wafer from an ingot (2) having a c-axis provided in a state in which a center axis (L) is tilted in an off-angle direction (Dθ) by an off-angle greater than 0 degree, the method comprising the following procedure, steps, or processes. A surface (21) on one end side of the ingot in a height direction thereof is irradiated with a laser beam having a penetrating property, to form a delamination layer (25) at a depth corresponding to the thickness of the wafer from the surface. A load is applied in one direction at one end (23) of the ingot in the off-angle direction, to delaminate a wafer precursor (26), which is a portion of the ingot between the surface and the delamination layer, from the ingot at the delamination layer. A major surface of a planar delaminated body obtained by delaminating the wafer precursor from the ingot is planarized to obtain a wafer.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on Japanese Patent Application No. 2021-199578, filed on December 8, 2021, the contents of which are incorporated herein by reference. [Technical Field]

[0002] The present disclosure relates to wafer manufacturing methods. [Background technology]

[0003] Patent Document 1 provides a wafer production method that can efficiently produce wafers from an ingot. Specifically, the wafer production method described in Patent Document 1 includes a separation origin formation step and a wafer peeling step. In the separation origin formation step, a focal point of a laser beam having a wavelength that is transparent to the hexagonal single crystal ingot is positioned at a depth from the surface corresponding to the thickness of the wafer to be produced, and the focal point and the ingot are moved relative to each other to irradiate the laser beam onto the surface. This forms a modified layer parallel to the surface and a crack extending from this modified layer along the C-plane, thereby forming a separation origin. In the wafer peeling step, the hexagonal single crystal ingot with the separation origin formed is immersed in water and ultrasonic vibration is applied to peel off the plate-shaped object from the hexagonal single crystal ingot. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6391471 Summary of the Invention

[0005] In the wafer production method described in Patent Document 1, when cleavage progresses due to ultrasonic vibration, the separation position varies in the height direction of the ingot, which results in unevenness on the surface formed after separation, which increases the amount of grinding and polishing required, or causes separation defects, resulting in poor manufacturing efficiency. The present disclosure has been made in consideration of the above-mentioned circumstances, etc. That is, the present disclosure provides, for example, a wafer manufacturing method with higher manufacturing efficiency than conventional methods.

[0006] According to one aspect of the present disclosure, a wafer manufacturing method is a method for obtaining wafers from an ingot, the method comprising the steps of: forming a peeling layer by irradiating a surface of one end side in a height direction of the ingot with a laser beam having transparency to form a peeling layer from the surface to a depth corresponding to a thickness of the wafer; wafer peeling, which peels off a portion of a wafer precursor between the surface and the peeling layer from the ingot at the peeling layer; wafer planarization, which flattens the main surface of the plate-shaped peeled body obtained by the wafer peeling; Including, a c-axis of the ingot is inclined at an off-angle exceeding 0 degrees in an off-angle direction with respect to a central axis perpendicular to the surface; The wafer peeling is performed by The Si plane is the top surface, and the orientation of the ingot is set so that the top surface is the upper surface, and the facet region is positioned on the higher side of the C plane. This is the higher end of the C plane. One end of the ingot in the off-angle direction a facet region that does not overlap with the facet region in a circumferential direction along a circumference surrounding the central axis, This is done by applying a load in the direction.

[0007] According to another aspect of the present disclosure, a wafer manufacturing method is a method for obtaining wafers from an ingot, the method comprising the steps of: forming a peeling layer by irradiating a top surface, which is a surface on one end side in a height direction of the ingot, with a laser beam having transparency to form a peeling layer from the top surface to a depth corresponding to a thickness of the wafer; wafer delamination, which delaminates a portion of the wafer precursor between the top surface and the delamination layer from the ingot at the delamination layer; wafer planarization, which flattens the main surface of the plate-shaped peeled body obtained by the wafer peeling; Including, The c-axis of the ingot is inclined at an off-angle exceeding 0 degrees in an off-angle direction with respect to a central axis perpendicular to the top surface, The wafer peeling is performed by joining the top surface of the ingot to a top surface fixing member; a bottom surface, which is a surface on the other end side in the height direction of the ingot, and a bottom surface fixing member; A load is applied to the top surface fixing member and / or the bottom surface fixing member so that a load is applied in one direction on one end side of the ingot in the off-angle direction. This is done by The joining of the top surface and the top surface fixing member, and the joining of the bottom surface and the bottom surface fixing member are carried out by spreading an adhesive placed inside the outer diameter of the ingot using heat and / or pressure until it overflows and adheres to the side surface between the top surface and the bottom surface of the ingot.

[0008] mosquito According to this wafer manufacturing method, it is possible to provide a wafer manufacturing method with higher manufacturing efficiency than conventional methods.

[0009] In addition, in each section of the application documents, each element may be assigned a reference symbol in parentheses. In this case, the reference symbol merely indicates an example of the correspondence between the element and the specific configuration described in the embodiment described below. Therefore, the present disclosure is not limited in any way by the description of the reference symbol. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a side view showing a schematic configuration of a wafer, an ingot, and a peeled body in a wafer manufacturing method according to an embodiment of the present disclosure. [Figure 2] 1A to 1C are process diagrams illustrating an outline of a wafer manufacturing method according to an embodiment of the present disclosure. [Figure 3A] 3 is a side view showing a schematic configuration of the ingot that has undergone the peeling layer forming step shown in FIG. 2. FIG. [Figure 3B] FIG. 3B is a plan view of the ingot shown in FIG. 3A. [Figure 4A] 3 is a side view showing an outline of the release layer forming step shown in FIG. 2 and a release layer forming device used in this step. [Figure 4B] 3 is a front view showing an outline of the release layer forming step shown in FIG. 2 and a release layer forming device used in this step. FIG. [Figure 5A] FIG. 4C is an enlarged plan view showing an outline of the release layer forming step shown in FIGS. 4A and 4B. [Figure 5B] 4C is an enlarged view showing an outline of the peeling layer forming step shown in FIG. 4A and FIG. 4B near the focal point. [Figure 6] FIG. 5 is a plan view illustrating an outline of the release layer forming step illustrated in FIGS. 4A and 4B. [Figure 7] FIG. 5 is an enlarged view of the laser beams shown in FIGS. 4A and 4B near the focal point. [Figure 8A] FIG. 4B is a side view illustrating an outline of the release layer forming step shown in FIG. 4A. [Figure 8B] FIG. 10 is a side view illustrating an outline of a release layer forming step in another example. [Figure 9]FIG. 4B is a side view illustrating an outline of the release layer forming step shown in FIG. 4A. [Figure 10] FIG. 4B is a side view illustrating an outline of the release layer forming step shown in FIG. 4A. [Figure 11] 3 is a side view showing an outline of the wafer peeling step shown in FIG. 2 and a peeling device used in the step. [Figure 12] 10A and 10B are diagrams illustrating an outline of a delamination device according to a modified example and a wafer delamination process using the same. [Figure 13] FIG. 10 is a partial cross-sectional side view for explaining an outline of a wafer peeling step according to a modified example. [Figure 14] 10A to 10C are process diagrams illustrating an outline of a wafer peeling process according to a modified example. [Figure 15A] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15B] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15C] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15D] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15E] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15F] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15G] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 15H] FIG. 10 is a side view illustrating an outline of a wafer peeling step according to a modified example. [Figure 16] FIG. 10 is a plan view illustrating an outline of a wafer peeling step according to a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0011] (Embodiment) Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0012] (Wafer and ingot structure) Referring to FIG. 1, the wafer 1 manufactured by the wafer manufacturing method according to the present embodiment is obtained by slicing an ingot 2 having a substantially cylindrical shape in a side view, and is formed in a substantially circular thin plate shape in a plan view. That is, the wafer 1 and the ingot 2 have a substantially cylindrical side surface or end surface surrounding the central axis L. The central axis L is a virtual straight line parallel to the substantially cylindrical side surface or end surface of the wafer 1 or the ingot 2 and passing through the axis center of the wafer 1 or the ingot 2. For the sake of simplicity of illustration and description, the so-called orientation flat usually provided on the wafer 1 or the ingot 2 is omitted in this specification.

[0013] In the present embodiment, the ingot 2 is a single crystal SiC ingot having a c-axis Lc and a (0001) plane Pc that are orthogonal to each other, and has an off-angle θ exceeding 0 degrees. The c-axis Lc is a crystal axis indicated as

[0001] by direction indices. The (0001) plane Pc is a crystal plane that is orthogonal to the c-axis Lc and is called the "C plane" in a strictly crystallographic sense. The off-angle θ is the angle formed between the central axis L of the wafer 1 or the ingot 2 and the c-axis Lc, and is, for example, about 1 to 4 degrees. That is, the c-axis Lc in the wafer 1 and the ingot 2 is provided in a state where the central axis L is inclined by an off-angle θ exceeding 0 degrees in the off-angle direction Dθ. The off-angle direction Dθ is the moving direction of a point on the central axis L located on the laser irradiation surface (that is, the upper surface or the top surface in the figure) of the wafer 1 or the ingot 2 when the central axis L is rotated toward the c-axis Lc around the intersection of the central axis L and the c-axis Lc, mapped onto the laser irradiation surface.

[0014] For the sake of simplicity of description, a right-handed XYZ coordinate system is set as shown in FIG. 1. In such a right-handed XYZ coordinate system, it is assumed that the off-angle direction Dθ and the positive X-axis direction are in the same direction. Also, the X-axis and the Y-axis are parallel to the main surfaces of the wafer 1 and the ingot 2. The "main surface" is a surface orthogonal to the plate thickness direction in a plate-shaped object, and can also be referred to as the "top surface", "bottom surface", or "plate surface". Alternatively, the "main surface" is a surface orthogonal to the height direction in a columnar object such as the ingot 2, and can also be referred to as the "top surface" or "bottom surface". Note that when the position or orientation in the vertical direction, i.e., the direction of the gravitational force, is a problem for the top surface or the bottom surface, it may be referred to as the "upper surface" or "lower surface". The "upper surface" refers to the upper surface when a pair of main surfaces facing opposite directions are arranged in the vertical direction. In contrast, the "lower surface" is the surface opposite to the "upper surface", and refers to the lower surface when a pair of main surfaces facing opposite directions are arranged in the vertical direction. Further, it is assumed that the thickness direction of the wafer 1 and the height direction of the ingot 2 are parallel to the Z-axis. Any direction orthogonal to the Z-axis may be hereinafter referred to as the "in-plane direction".

[0015] The wafer 1 has a wafer C plane 11 and a wafer Si plane 12 which are a pair of main surfaces. In the present embodiment, the wafer 1 is formed such that the wafer C plane 11, which is the top surface, is inclined by an off-angle θ with respect to the (0001) plane Pc. Similarly, the ingot 2 has an ingot side surface 20 in a substantially cylindrical surface shape, and an ingot C plane 21 and an ingot Si plane 22 which are a pair of main surfaces. The ingot 2 is formed such that the ingot C plane 21, which is the top surface, is inclined by an off-angle θ with respect to the (0001) plane Pc. Hereinafter, one end, i.e., the upstream end, in the off-angle direction Dθ of the ingot 2 is referred to as the first end 23, and the other end, i.e., the downstream end, is referred to as the second end 24. In FIG. 1, the direction in which the wafer C plane 11 and the ingot C plane 21 face is shown as the positive Z-axis direction.

[0016] Further, the ingot 2 has a facet region RF. The facet region RF may also be referred to as a "facet portion". The portion of the ingot 2 other than the facet region RF is hereinafter referred to as a non-facet region RN. Similarly, the non-facet region RN may also be referred to as a "non-facet portion".

[0017] (Outline of Wafer Manufacturing Method) The wafer manufacturing method according to this embodiment is a method for obtaining the wafer 1 from the ingot 2 and includes the following steps.

[0018] (1) Release layer formation step: By irradiating the ingot C surface 21, which is one end main surface in the height direction of the ingot 2, with a laser beam having a predetermined degree of permeability with respect to the ingot 2, a release layer 25 is formed at a depth corresponding to the thickness of the wafer 1 from the ingot C surface 21. Here, the "predetermined degree of permeability" means a degree of permeability that enables the formation of a laser beam focus point at a depth corresponding to the thickness of the wafer 1 inside the ingot 2. Further, the "depth corresponding to the thickness of the wafer 1" is a dimension obtained by adding a thickness corresponding to a predetermined processing allowance in a wafer flattening step or the like to the thickness of the wafer 1 as a finished product (i.e., the target value of the thickness), and may also be referred to as the "depth corresponding to the thickness of the wafer 1".

[0019] (2) Wafer peeling step: A wafer precursor 26, which is a portion between the ingot C surface 21, which is the laser irradiation surface, and the release layer 25, is peeled from the ingot 2 at the release layer 25. Here, as in the above expression of the "wafer peeling step", the plate-like object obtained by peeling the wafer precursor 26 from the ingot 2 may be referred to as a "wafer" in the common social concept. However, in order to distinguish it from the final wafer 1 after manufacturing having a main surface mirror-finished by epitaxy, such a plate-like object is hereinafter referred to as a "peeled body 30". The release body 30 has a pair of main surfaces, a non-release surface 31 and a release surface 32. The non-release surface 31 is the surface on the side that did not form the release layer 25 before the wafer release process, and it corresponds to the ingot C surface 21 before performing the release layer formation process and the wafer release process. The release surface 32 formed the release layer 25 before the wafer release process and is a newly generated surface by the wafer release process. The release surface 32 has rough (i.e., to the extent that grinding or polishing is required) irregularities due to the release layer 25 and the release by the wafer release process.

[0020] (3) Wafer planarization process: By planarizing at least the release surface 32 among the non-release surface 31 and the release surface 32 which are the main surfaces of the release body 30, the final wafer 1 after manufacturing is obtained. In the wafer planarization process, in addition to general grinding wheel polishing and CMP, ECMG and ECMP can be used. Note that CMP is the abbreviation of Chemical Mechanical Polishing. ECMG is the abbreviation of Electro-Chemical Mechanical Grinding. ECMP is the abbreviation of Electro-Chemical Mechanical Polishing. The wafer planarization process can be performed by using these multiple types of planarization processes alone or in an appropriate combination.

[0021] (4) Ingot planarization process: After peeling the wafer precursor 26, the newly generated top surface of the ingot 2, that is, the ingot C surface 21, is planarized or mirror-finished so that it can be reused in the release layer formation process. In the ingot planarization process, in addition to general grinding wheel polishing and CMP, ECMG and ECMP can be used. The ingot planarization process can also be performed by using these multiple types of planarization processes alone or in an appropriate combination.

[0022] 2 is a process diagram showing a typical example of a wafer manufacturing method according to this embodiment. As shown in FIG. 2, a peeled body 30, which has been peeled from an ingot 2 through a peeling layer forming process and a wafer peeling process, is finished into an epi-ready wafer 1 through the following processes. Rough grinding of the peeled surface 32 that will become the wafer Si surface 12 ECMG grinding of the peeled surface 32 after rough grinding ECMP polishing of the peeled surface 32 after ECMG grinding Cleaning

[0023] Furthermore, the ingot 2 remaining after the separation body 30 has been separated from the ingot 2 through the separation layer forming step and the wafer separation step can be subjected to the separation layer forming step again through the following steps. Rough grinding of the ingot C-face 21 newly generated by the wafer peeling process Finish grinding of the C surface 21 of the ingot after rough grinding Cleaning

[0024] Each step will be described in detail below with reference to other figures in addition to FIGS.

[0025] (Release layer formation process) 3A and 3B show a schematic configuration of ingot 2 in a state in which separation layer 25 and wafer precursor 26 have been formed by the separation layer formation process. FIGS. 4A and 4B show an overview of the separation layer formation process and a schematic configuration of separation layer formation apparatus 40 used in this process. The right-handed XYZ coordinates shown in FIGS. 3A to 4B are displayed to match the right-handed XYZ coordinates shown in FIG. 1.

[0026] Referring to FIGS. 3A and 3B, a plurality of scanning lines Ls, which are irradiation marks of a linear laser beam along the X-axis, are formed in the Y-axis direction, thereby forming the release layer 25. The scanning line Ls is formed by linearly forming an irradiation mark RM of the laser beam on the ingot 2. In the present embodiment, the scanning line Ls is provided along the off-angle direction Dθ. And the plurality of scanning lines Ls are arranged in the line feed direction Df. The line feed direction Df is an in-plane direction orthogonal to the off-angle direction Dθ. That is, the line feed direction Df is orthogonal to the off-angle direction Dθ and orthogonal to the height direction of the ingot 2.

[0027] Referring to FIGS. 4A and 4B, the release layer forming apparatus 40 includes a chuck table 41 and a condensing device 42. The chuck table 41 is configured to hold the ingot 2 on the side of the ingot Si surface 22 which is its bottom surface. Specifically, for example, the chuck table 41 includes a suction mechanism or the like that sucks the ingot Si surface 22 by air pressure or the like. Note that, as will be described later, the fixing method of the ingot 2 to the chuck table 41 is not limited to such a mode. The condensing device 42 is provided to irradiate the ingot 2, which is a workpiece, with a laser beam B oscillated by a pulse laser oscillator (not shown). Specifically, the condensing device 42 is configured to form a condensing point BP of the laser beam B at a depth corresponding to the thickness of the wafer 1 from the ingot C surface 21 inside the ingot 2. That is, the condensing device 42 is provided to irradiate the ingot 2 with the laser beam B from the side of the ingot C surface 21 which is the top surface of the ingot 2. The release layer forming apparatus 40 is configured such that the condensing point BP of the laser beam B can be relatively moved with respect to the ingot 2 at least in the in-plane direction, that is, the XY direction in the drawing. Here, the "in-plane direction" refers to a direction parallel to the ingot C surface 21 which is the top surface of the ingot 2.

[0028] The peeling layer forming apparatus 40 forms a scan line Ls along the scanning direction Ds (i.e., the first direction) by "laser scanning," which scans the ingot C-face 21 with the laser beam B in the scanning direction Ds. That is, "laser scanning" is the process of irradiating the ingot C-face 21 with the laser beam B while moving the irradiation position PR of the laser beam B on the ingot C-face 21, which is the laser irradiation surface, in the scanning direction Ds. In this embodiment, the scanning direction Ds is a direction along the off-angle direction Dθ, specifically, the same direction as or opposite to the off-angle direction Dθ. The peeling layer forming apparatus 40 then performs laser scanning multiple times while changing the position in the line feed direction Df (i.e., the second direction), thereby forming multiple scan lines Ls in the line feed direction Df, thereby forming the peeling layer 25. The line feed direction Df and the scanning direction Ds are both in-plane directions (i.e., directions along the ingot C-face 21) and are perpendicular to each other.

[0029] Specifically, in this embodiment, the peeling layer forming apparatus 40 moves the chuck table 41, on which the ingot 2 is placed, relative to the focusing device 42 in the scanning direction Ds, and scans the ingot C surface 21 with the laser beam B, thereby forming a scan line Ls along the scanning direction Ds. After performing one laser scan, the peeling layer forming apparatus 40 moves the chuck table 41 relative to the focusing device 42 by a predetermined amount in the line feed direction Df. The peeling layer forming apparatus 40 then again moves the chuck table 41 relative to the focusing device 42 in the scanning direction Ds (i.e., the same or opposite direction as in the previous laser scan) and scans the laser beam B, thereby forming the scan line Ls. In this manner, the peeling layer forming apparatus 40 scans the laser beam B over substantially the entire width in the line feed direction Df, thereby forming a plurality of scan lines Ls in the line feed direction Df. As a result, the peeling layer 25 is formed by the plurality of scan lines Ls provided along the line feed direction Df. Furthermore, a wafer precursor 26, which will eventually become the wafer 1, is formed on the ingot C-plane 21 side of the peeling layer 25. As described above, in this embodiment, the focusing device 42 is fixed in the in-plane direction, while the chuck table 41 supporting the ingot 2 is movable at least in the in-plane direction by a scanning device such as an electric stage device (not shown). However, as will be described later, the present disclosure is not limited to this embodiment. For example, there may be an embodiment in which the chuck table 41 supporting the ingot 2 is fixed in the in-plane direction, while the focusing device 42 is movable in the in-plane direction by a scanning device (not shown). However, in any of these embodiments, the laser beam B and its irradiation position PR appear to move in the in-plane direction on the surface of the ingot 2, or the laser beam B and its focal point BP appear to move in the in-plane direction within the ingot 2. Therefore, for simplicity of explanation, the following description may be given assuming that the laser beam B and its irradiation position PR move in an in-plane direction on the surface of the ingot 2, or that the laser beam B and its focal point BP move in an in-plane direction within the ingot 2. However, as will be described later, the present disclosure is not limited to such an embodiment.

[0030] In this embodiment, as shown in FIGS. 4A and 4B, a plurality of laser beams B with different irradiation positions PR in the scanning direction Ds and the line feed direction Df are irradiated onto the ingot C surface 21 in one laser scan. Specifically, as shown in FIG. 5A, a plurality of laser beams B (i.e., the first beam B1, etc.) whose irradiation positions PR on the ingot C surface 21 are arranged obliquely with respect to both the scanning direction Ds and the line feed direction Df in plan view move along the scanning direction Ds. Thereby, a plurality of scanning lines Ls are formed in one laser scan. Therefore, the cycle time in the peeling layer forming process can be favorably shortened.

[0031] In FIGS. 4A, 4B, and 5A, an example in which there are three laser beams B is illustrated as the plurality of laser beams B. However, this is due to the convenience of simplifying the illustration, and there is no particular limitation on the number of laser beams B. However, for the sake of simplicity of explanation, hereinafter, the explanation will continue assuming that at least the first beam B1, the second beam B2, and the third beam B3 are included as the plurality of laser beams B. Among the first beam B1, the second beam B2, and the third beam B3, the first beam B1 is the most leading, that is, the one located on the most scanning direction Ds side. On the other hand, the third beam B3 is the most trailing. And the second beam B2 is located between the first beam B1 and the third beam B3 with respect to the scanning direction Ds and the line feed direction Df.

[0032] As shown in FIG. 5A, the first beam B1 advances in the scanning direction Ds ahead of the second beam B2. FIG. 5B shows a state in which the subsequent second beam B2 is irradiated in the vicinity of the irradiation mark RM of the first beam B1. As shown in FIG. 5B, an irradiation affected region RA is generated to a predetermined depth by the irradiation of the first beam B1. The irradiation affected region RA includes an irradiation mark RM composed of a modified region formed by separating SiC into Si and C by the irradiation of the laser beam B, and a crack C extending from such an irradiation mark RM along the (0001) plane Pc to its periphery. Therefore, the irradiation position PR of the second beam B2 may overlap at least with the crack C and in the in-plane direction in the irradiation affected region RA formed by the preceding first beam B1. When the irradiation mark RM and the crack C included in the irradiation affected region RA by the preceding first beam B1 are present at the irradiation position PR of the subsequent second beam B2, the absorption rate of the second beam B2 is increased by such an irradiation affected region RA. For this reason, an irradiation mark RM by the subsequent second beam B2 is likely to be generated at a depth substantially the same as the depth of the irradiation affected region RA by the preceding first beam B1. The same applies to the relationship between the second beam B2 and the third beam B3. Therefore, three irradiation marks RM or irradiation affected regions RA adjacent to each other in the line feed direction Df, which are formed at once by the first beam B1 to the third beam B3 in one laser scan, are likely to be generated at substantially the same depth. That is, a plurality of scanning lines Ls constituting the peeling layer 25 adjacent to each other in the line feed direction Df are likely to be generated at substantially the same depth.

[0033] Thus, in the present embodiment, the variation in the depth of the irradiation mark RM, that is, the scanning line Ls, constituting the peeling layer 25 can be suppressed as much as possible. Thereby, the step and unevenness on the peeling surface 32 generated by the peeling in the peeling layer 25 are favorably suppressed, the processing cost of grinding and polishing on the peeling surface 32 is reduced, and the occurrence of peeling failure is favorably suppressed. In addition, the cycle time for forming the peeling layer 25 can be shortened. Therefore, according to the present embodiment, it is possible to improve the manufacturing efficiency as compared with the prior art.

[0034] FIG. 6 shows the locus of the relative movement of the central position in the in-plane direction of the light condensing device 42 with respect to the ingot 2. The "central position in the in-plane direction of the light condensing device 42" is typically, for example, the central position in the array of a plurality of laser beams B. As shown in FIGS. 4A and 6, in the present embodiment, the attitude of the ingot 2 is set such that the facet region RF is located on the "low off-angle side", and the laser beam B is irradiated from the side of the ingot C plane 21 (that is, so-called C-plane irradiation) to perform the peeling layer forming step. The "low off-angle side" means the side with a low inclination on the C plane, that is, the (0001) plane Pc, when the attitude of the ingot 2 is set such that the ingot C plane 21, which is one main surface, becomes the upper surface. On the other hand, the "high off-angle side" means the side with a high inclination on the C plane, that is, the (0001) plane Pc, when the attitude of the ingot 2 is set such that the ingot C plane 21 becomes the upper surface.

[0035] As will be described later, by applying a unidirectional load for peeling the wafer precursor 26 from the ingot 2 at one end on the "high off-angle side" of the ingot 2, extremely good wafer peeling is achieved. Here, an assumed example will be considered in which, after setting the attitude of the ingot 2 such that the facet region RF is located on the "high off-angle side" and irradiating the laser beam B from the side of the ingot Si plane 22 (that is, so-called Si-plane irradiation), a unidirectional load is applied at one end on the "high off-angle side" of the ingot 2. In this regard, the end portion of the ingot 2 close to the facet region RF is less likely to crack in the first place. Therefore, in such an assumed example, since the peeling start position becomes the end portion close to the facet region RF where cracks are less likely to occur, the success rate of the wafer peeling process may be low. On the other hand, in the present embodiment, after setting the attitude of the ingot 2 such that the facet region RF is located on the "low off-angle side" and irradiating the laser beam B from the side of the ingot C plane 21, a unidirectional load is applied at one end on the "high off-angle side" of the ingot 2. In this case, the peeling start position becomes a portion far from the facet region RF where cracks are relatively likely to occur. Therefore, according to the present embodiment, the success rate of the wafer peeling process is improved.

[0036] It is known that the intensity of the laser beam B reaching the focal point BP is stronger in the non-facet region RN than in the facet region RF. Therefore, in this embodiment, in the peeling layer formation process, the laser beam B is irradiated onto the main surface of the ingot 2 so that the energy application density due to the irradiation of the laser beam B is higher in the facet region RF than in the non-facet region RN. The "energy application density" here refers to the energy application density in a plane along the main surface of the ingot 2. The following measures may be used alone or in combination. Specifically, for example, the output of the laser beam B is increased in the facet region RF compared to the non-facet region RN. Alternatively, for example, the laser beam B is irradiated onto the main surface of the ingot 2 so that the irradiation frequency of the laser beam B is higher in the facet region RF than in the non-facet region RN. More specifically, for example, the repetition frequency of the laser beam B is increased in the facet region RF compared to the non-facet region RN, thereby narrowing the irradiation interval in the scanning direction Ds. When increasing the output power in the facet region RF compared to the non-facet region RN, the output power in the facet region RF is preferably about 1.5 times that in the non-facet region RN. When decreasing the irradiation interval in the scanning direction Ds or the line feed direction Df in the facet region RF compared to the non-facet region RN, the irradiation interval in the facet region RF is preferably about 2 / 5 of the irradiation interval in the non-facet region RN. Alternatively, for example, it is possible to irradiate the facet region RF with the laser beam B separately from irradiating the entire region including the facet region RF and the non-facet region RN with the laser beam B. Note that when irradiating the facet region RF with the laser beam B, the laser beam B may also be irradiated onto a region of the non-facet region RN adjacent to the facet region RF.

[0037] According to the separation layer forming process of this embodiment, it is possible to form a separation layer 25 satisfactorily in the entire region including the facet region RF and the non-facet region RN. In particular, the separation layer 25 can be formed in the facet region RF in the same manner as in the non-facet region RN without adjusting the distance in the Z-axis direction between the focusing device 42 on the irradiation side of the laser beam B and the chuck table 41 that supports the ingot 2. Therefore, according to this embodiment, it is possible to improve manufacturing efficiency compared to conventional methods.

[0038] 4A and 6, the peeling layer formation process involves forward scanning Sc1, in which the irradiation position PR when irradiated with laser beam B moves on the surface of the ingot 2 in the same direction as the off-angle direction Dθ, and backward scanning Sc2, in which the irradiation position PR when irradiated with laser beam B moves on the surface of the ingot 2 in the opposite direction to the off-angle direction Dθ. That is, in forward scanning Sc1, the scanning direction Ds is the same as the off-angle direction Dθ. In contrast, in backward scanning Sc2, the scanning direction Ds is the opposite direction to the off-angle direction Dθ. Forward scanning Sc1 and backward scanning Sc2 are performed alternately.

[0039] Between the end of one forward scan Sc1 and the start of the next forward scan Sc1, the relative position of the focusing device 42 with respect to the ingot 2 moves a predetermined amount in the line feed direction Df. However, between the end of one forward scan Sc1 and the start of the return scan Sc2 performed immediately thereafter, the relative position of the focusing device 42 in the line feed direction Df may or may not move. The same applies to the period between the end of one return scan Sc2 and the start of the subsequent forward scan Sc1. The amount of relative movement in the line feed direction Df at each stage can be set appropriately depending on the irradiation conditions of the laser beam B, etc.

[0040] In the forward path scanning Sc1, the laser beam B is irradiated over the entire width of the ingot 2 in the scanning direction Ds. That is, in the forward path scanning Sc1, while moving the irradiation position PR in the scanning direction Ds, which is the same direction as the off-angle direction Dθ on the surface of the ingot 2, the laser beam B is irradiated onto the main surface of the ingot 2, thereby forming a scanning line Ls across both ends of the main surface of the ingot 2 in the scanning direction Ds. On the other hand, in the return path scanning Sc2, the laser beam B may or may not be irradiated over the entire width of the ingot 2 in the scanning direction Ds. Alternatively, in the return path scanning Sc2, the laser beam B may be irradiated on only a part of the ingot 2, rather than the entire width in the scanning direction Ds.

[0041] Specifically, for example, in the return path scanning Sc2, the laser beam B may be irradiated only on the facet region RF and its peripheral portion. Thereby, it becomes possible to favorably form the release layer 25 over the entire region including the facet region RF and the non-facet region RN. Alternatively, for example, in the return path scanning Sc2, the laser beam B may be irradiated only on the end portion of the ingot 2 in the scanning direction Ds. In this case, in the return path scanning Sc2, while the irradiation position PR moves in the scanning direction Ds, which is the opposite direction to the off-angle direction Dθ on the surface of the ingot 2, a scanning line Ls is formed at the end portion of the main surface of the ingot 2 in the scanning direction Ds. Thereby, the start of peeling in the wafer peeling process is favorably promoted, and the success rate of the wafer peeling process is improved. Also, in the return path scanning Sc2, the laser beam B may be irradiated only on the facet region RF and its peripheral portion and the end portion of the ingot 2 in the scanning direction Ds.

[0042] As shown in FIG. 7, in the present embodiment, the laser beam B has an intensity distribution such that the intensity is higher at the peripheral edge portion outside the central portion in the beam diameter direction, which is the direction extending radially from the axis center thereof. Specifically, the laser beam B has a beam shape that is annular, i.e., hollow, on the front side of the condensing point BP and condenses into a dot shape at the condensing point BP. At the condensing point BP, the laser beam B has a condensing diameter dc, which is the minimum beam diameter. The intersection range RX shown in FIG. 7 is a predetermined range centered on the condensing point BP in the beam axis direction, which is the irradiation direction of the laser beam B, where the peripheral edge portions with high intensity in the laser beam B overlap each other.

[0043] Thus, the separation layer forming device 40 irradiates the ingot 2 with the annular laser beam B. Such an annular laser beam B and a device for generating such a laser beam B and irradiating a workpiece therewith are already known or well-known at the time of filing of the present application (see, for example, Japanese Patent Application Laid-Open No. 2006-130691, Japanese Patent Application Laid-Open No. 2014-147946, etc.). Therefore, details of the generating device and generating method of such a laser beam B are omitted in this specification.

[0044] FIG. 8A shows a state in which an irradiation affected area RA including an irradiation mark RM is formed by the annular laser beam B according to the present embodiment. FIG. 8B shows a state in which an irradiation affected area RA including an irradiation mark RM is formed by a non-annular, i.e., solid, laser beam B as another example different from the present embodiment.

[0045] As shown in FIG. 8B , when a solid laser beam B is used, an irradiation mark RM, which is a modified region formed by the separation of SiC into Si and C due to irradiation with the laser beam B, may occur at a depth different from that of the focal point BP. Therefore, the depth of the irradiation-affected region RA, which is composed of the irradiation mark RM and the crack C propagating from the irradiation mark RM, may also be different from that of the focal point BP. Specifically, for example, the applied energy density of the laser beam B may be high enough to generate the irradiation mark RM at a position shallower than the focal point BP. In this case, the irradiation mark RM may occur at a position shallower than the focal point BP. The depth of the irradiation mark RM may vary due to variations in the irradiation energy of the laser beam B, variations in the refractive index of the ingot 2, variations in the optical system of the focusing device 42, and so on. The region where the irradiation mark RM may occur is shown as a modifiable range RC in the figure. The irradiation mark RM corresponds to the “modified layer” in Patent Document 1.

[0046] In contrast, as shown in FIG. 8A , when an annular laser beam B is used, the energy application density due to irradiation of the laser beam B increases to a level sufficient to generate an irradiation mark RM only at a depth near the focal point BP. That is, as in the case of FIG. 8B , where a solid laser beam B is used, it is difficult for the energy application density due to irradiation of the laser beam B to increase to a level sufficient to generate an irradiation mark RM at a position shallower than the focal point BP. Therefore, the irradiation mark RM is stably generated at a depth near the focal point BP. Unlike the case of using a solid laser beam B, the modifiable range RC is limited to a narrow depth range centered on the focal point BP. Therefore, variation in the depth at which the irradiation mark RM occurs can be effectively suppressed. In other words, the peeling layer 25 can be formed as thin as possible, effectively reducing the processing cost during grinding and polishing after peeling. Therefore, this embodiment enables improved manufacturing efficiency compared to conventional methods.

[0047] By the way, in the method described in Patent Document 1, the laser scanning direction is orthogonal to the "direction in which the off-angle θ is formed (i.e., the off-angle direction Dθ in FIGS. 1, 3A, etc.)". For this reason, cracking is unstable and material loss increases. On the other hand, in the present embodiment, as shown in FIG. 4A, the scanning direction Ds, which is the moving direction of the laser beam B inside the ingot 2, is parallel to the off-angle direction Dθ. That is, in the peeling layer forming step, the irradiation position PR is moved in the scanning direction Ds along the off-angle direction Dθ by laser scanning. In other words, the peeling layer forming apparatus 40 forms the scanning line Ls along the off-angle direction Dθ by relatively moving the condensing device 42 with respect to the ingot 2 in the scanning direction Ds parallel to the off-angle direction Dθ and scanning the laser beam B. Then, as shown in FIGS. 9 and 10, the irradiation marks RM and the cracks C are formed along the (0001) plane Pc. As a result, cracking in the peeling layer 25 during the wafer peeling step can be stabilized, and material loss is favorably reduced. Also, the processing cost in the wafer flattening step is favorably reduced, and thus the process time can be shortened as much as possible. Therefore, according to the present embodiment, it is possible to provide a wafer manufacturing method with higher manufacturing efficiency than in the past.

[0048] FIG. 9 shows an example in which the scanning direction Ds is the same as the off-angle direction Dθ. FIG. 10 shows an example in which the scanning direction Ds is opposite to the off-angle direction Dθ. That is, in the example shown in FIG. 9, when the posture of the ingot 2 is set such that the C plane 21 of the ingot becomes the upper surface as shown in FIG. 4A, the irradiation position PR is moved from the higher side to the lower side in the (0001) plane Pc by laser scanning. On the other hand, in the example shown in FIG. 10, when the posture of the ingot 2 is set such that the C plane 21 of the ingot becomes the upper surface, the irradiation position PR is moved from the lower side to the higher side in the (0001) plane Pc by laser scanning.

[0049] For example, assume a situation where there is no irradiation influence region RA, i.e., irradiation mark RM or crack C, around the irradiation position PR in the in-plane direction. In such a situation, due to the irradiation of the laser beam B, the irradiation mark RM is likely to occur at a depth near the beam waist BP. On the other hand, in reality, the laser beam B moves in the scanning direction Ds while successively generating irradiation marks RM and cracks C. Therefore, the above situation mainly occurs when forming the irradiation mark RM corresponding to the starting point of the scanning line Ls, which is initially formed in one laser scan. Thus, in most scenes during laser scanning, a situation occurs where there is an irradiation influence region RA around the irradiation position PR in the in-plane direction.

[0050] That is, as shown in FIGS. 9 and 10, there is usually an irradiation influence region RA that was previously (e.g., immediately before) formed at the current irradiation position PR. Then, in such an irradiation influence region RA, the absorption rate of the laser beam B increases. Also, such an irradiation influence region RA is formed along the (0001) plane Pc. For this reason, due to laser scanning, the irradiation mark RM is likely to progress along the (0001) plane Pc.

[0051] Here, in the example shown in FIG. 9, due to laser scanning, when the irradiation mark RM progresses in the scanning direction Ds along the (0001) plane Pc, it is formed at a gradually deeper position and thus gradually moves away from the beam waist BP. Then, at approximately the same depth as the irradiation mark RM formed immediately before, the energy application density by the laser beam B irradiated this time may not be increased to a level that can generate a new irradiation mark RM. In this case, the irradiation mark RM can no longer progress along the (0001) plane Pc. Then, as shown in FIG. 9, the newly formed irradiation mark RM is formed at a depth near the beam waist BP in the laser beam B irradiated this time. That is, a step is generated between the irradiation mark RM formed immediately before and the irradiation mark RM formed this time.

[0052] On the one hand, in the example shown in FIG. 10, when the irradiation mark RM progresses in the scanning direction Ds along the (0001) plane Pc by laser scanning, it is formed at a gradually shallower position, thus gradually moving away from the beam waist BP. And when the energy application density by the laser beam B irradiated this time cannot be increased to a level capable of generating a new irradiation mark RM at approximately the same depth as the irradiation mark RM formed immediately before, the irradiation mark RM can no longer progress along the (0001) plane Pc. Then, as shown in FIG. 10, the newly formed irradiation mark RM is formed at a depth near the beam waist BP in the laser beam B irradiated this time. However, in the example shown in FIG. 10, unlike the example shown in FIG. 9, the progress direction of the irradiation mark RM is a direction approaching the light source side of the laser beam B, that is, the irradiation surface side of the ingot 2. Therefore, in the example shown in FIG. 10, the irradiation mark RM is more likely to progress longer than in the example shown in FIG. 9. Therefore, in the example shown in FIG. 10, the step generated between the irradiation mark RM formed immediately before and the irradiation mark RM formed this time becomes larger than in the example shown in FIG. 9.

[0053] <- In this way, by setting the scanning direction Ds to the same direction as the off-angle direction Dθ and moving the irradiation position PR in the laser scanning from the higher side to the lower side on the C plane, the step between the irradiation mark RM formed immediately before and the irradiation mark RM formed this time can be reduced. As a result, the peeling layer 25 can be formed as thin as possible, and thus the processing cost in grinding and polishing after peeling can be favorably reduced. Therefore, according to such an aspect, it is possible to further improve the manufacturing efficiency compared to the conventional case.

[0054] (Wafer peeling process) FIG. 11 shows an overview of the wafer peeling process and the peeling apparatus 50 used in such a process. It should be noted that the right-handed XYZ coordinates shown in FIG. 11 are shown to be consistent with the right-handed XYZ coordinates shown in FIG. 1.

[0055] The peeling device 50 is configured to peel the wafer precursor 26 from the ingot 2 at the peeling layer 25 by applying a load in one direction at the first end 23, which is one end of the ingot 2 in the in-plane direction parallel to the C-plane 21 of the ingot, that is, the off-angle direction Dθ. The first end 23 is the end on the "higher off-angle side", that is, the higher side end on the C-plane, that is, the (0001) plane Pc when the posture of the ingot 2 is set such that the C-plane 21 of the ingot becomes the upper surface. In the present embodiment, the peeling device 50 is configured to apply a static and / or dynamic load in the Z-axis direction in the drawing in a manner that separates the C-plane 21 of the ingot from the Si-plane 22 of the ingot at the first end 23 to the ingot 2. Specifically, in the present embodiment, the peeling device 50 includes a support table 51, a peeling pad 52, and a drive member 53.

[0056] The support table 51 is provided to support the ingot 2 from below. Specifically, the support table 51 is, for example, joined to the Si-plane 22 of the ingot, which is the bottom surface of the ingot, via an adhesive such as wax at the support fixing surface 51a, which is the upper surface thereof. The support table 51 has a first table end portion 51b and a second table end portion 51c, which are both end portions in the off-angle direction Dθ. The second table end portion 51c, which is the end portion on one side (that is, the left side in the drawing) in the off-angle direction Dθ, has a table base end surface 51d. The table base end surface 51d is formed in an inclined surface shape that rises as it goes in the off-angle direction Dθ. That is, as shown in FIG. 11, the support table 51 is formed in a trapezoidal shape in which the lower base is longer than the upper base in a side view.

[0057] The stripping pad 52 is disposed above the support table 51 and is movable toward and away from the support table 51 along the Z-axis in the figure. In other words, the stripping apparatus 50 is configured such that the support table 51 and the stripping pad 52 are relatively movable in the height direction of the ingot 2. The stripping pad 52 has a pad fixing surface 52a, which is its bottom surface, bonded to the ingot C-face 21, which is the top surface of the ingot 2, via an adhesive such as wax. The stripping pad 52 has a first pad end 52b and a second pad end 52c, which are opposite ends in the off-angle direction Dθ. The second pad end 52c, which is the end on one side (i.e., the left side in the figure) in the off-angle direction Dθ, has a pad end surface 52d. The pad end surface 52d is formed as an inclined surface that slopes downward toward the off-angle direction Dθ. In other words, as shown in FIG. 11 , the stripping pad 52 is formed in a trapezoidal shape in side view, with its lower base shorter than its upper base. The pad end surface 52d is provided at a position corresponding to the table base end surface 51d (i.e., directly above in the drawing). The state in which the ingot C-face 21 is fixed to the peeling pad 52 and the ingot Si-face 22 is fixed to the support table 51, and the ingot 2 is sandwiched between the support table 51 and the peeling pad 52 is hereinafter referred to as the "clamped state."

[0058] The driving member 53 is provided to apply an external force that relatively moves the support table 51 and the peeling pad 52 along the height direction of the ingot 2 in the clamped state to at least one of the support table 51 and the peeling pad 52. Specifically, the driving member 53 has a first driving end face 53a and a second driving end face 53b. The first driving end face 53a is formed in an inclined surface shape that descends as it goes in the off-angle direction Dθ. More specifically, the first driving end face 53a is provided parallel to the pad end face 52d. The second driving end face 53b is formed in an inclined surface shape that ascends as it goes in the off-angle direction Dθ. More specifically, the second driving end face 53b is provided parallel to the table base end face 51d. Further, the driving member 53 is provided such that, in the clamped state, the first driving end face 53a abuts on the pad end face 52d and the second driving end face 53b abuts on the table base end face 51d. That is, as shown in FIG. 11, the driving member 53 is formed in a shape obtained by rotating a trapezoid whose lower base is longer than the upper base 90 degrees clockwise in a side view. Then, the driving member 53 is configured to be driven in the upward direction along the height direction of the ingot 2 and / or in the off-angle direction Dθ which is the direction approaching the ingot 2 by a driving means (not shown). That is, the driving member 53 is provided such that, by being driven in the upward direction and / or the off-angle direction Dθ, a moment with the second pad end 52c as the force point FP and the first end 23 as the fulcrum PP and the action point WP acts on the ingot 2.

[0059] The wafer peeling process of peeling the wafer precursor 26 from the ingot 2 includes a table fixing process, a clamping process, and a peeling force application process. The table fixing process is a process of bonding the Si surface 22 of the ingot to the support fixing surface 51a to fix the ingot 2 to the support table 51. The clamping process is a process of forming a clamped state by bonding the C surface 21 of the ingot to the pad fixing surface 52a to fix the ingot 2 to the peeling pad 52. The peeling force application process is a process of applying a static or dynamic load with the second pad end 52c, which is an end of one of the peeling pads 52 in the off-angle direction Dθ, as the force point FP so that a moment with the first end 23 as the fulcrum PP and the action point WP acts on the ingot 2 in the clamped state. Specifically, the peeling force application process is a process of driving the driving member 53 upward and / or in the off-angle direction Dθ in the clamped state to press the second pad end 52c upward along the height direction of the ingot 2. Thereby, the wafer precursor 26, which is a part of the ingot 2, can be peeled from the ingot 2 with the peeling layer 25 as the interface.

[0060] Thus, in this embodiment, the wafer peeling process is performed by applying a load in one direction at the first end 23, which is one end of the ingot 2 in the in-plane direction parallel to the top surface of the ingot 2 (that is, the C surface 21 of the ingot in the example of FIG. 11). Then, a moment with the first end 23 as the fulcrum PP and the action point WP acts on the ingot 2.

[0061] In this regard, for example, in the wafer peeling process described in Japanese Patent No. 6678522, the action point WP and the fulcrum PP were provided inside the ingot 2, that is, inside the outer edge of the peeling layer 25 in the in-plane direction. In such a comparative example, in order to generate good peeling with the peeling layer 25 as the interface, a much larger load was required than in the present embodiment. Further, since a load was applied over a wide area of the peeling layer 25, the peeling crack position was not determined, and partial non-peeled portions or breakage in the taken-out wafer 1 sometimes occurred. Furthermore, there was a problem that the surface formed after peeling became rough and the processing cost of grinding and polishing increased. Therefore, in the comparative example, there was room for improvement in terms of reducing the load, yield, etc. Also, in the method for producing a wafer described in Patent Document 1, when cleavage progresses due to ultrasonic vibration, variations occur in the peeling position in the height direction of the ingot 2. Then, steps are generated on the surface formed after peeling, increasing the processing cost of grinding and polishing, or peeling failure occurs, deteriorating the manufacturing efficiency.

[0062] In contrast, in the wafer peeling process according to the present embodiment, in order to peel the wafer precursor 26 from the ingot 2 with the peeling layer 25, a load is applied in one direction at one end of the ingot 2 in the off-angle direction Dθ. That is, the load is concentrated at one end of the peeling layer 25 in the off-angle direction Dθ. Then, a moment with this one end as the fulcrum PP and the point of application WP acts on the ingot 2. As a result, peeling proceeds from a crack formed at one end side of the ingot 2 in the off-angle direction Dθ, so that breakage can proceed stably over the entire surface of the peeling layer 25 while reducing the applied load. Further, by stably setting the location where breakage occurs, it is possible to reduce the surface roughness of the peeling surface 32 and the ingot C surface 21 in the peeled body 30 generated after peeling. In particular, by setting the first end 23, which is the starting point of breakage occurrence, to one end on the "higher off-angle side" in the off-angle direction Dθ, breakage occurs smoothly and splitting becomes even more stable. For this reason, the defect occurrence rate in the wafer peeling process and the processing cost in grinding and polishing the ingot 2 and the peeled body 30 after the wafer peeling process can be favorably reduced. Therefore, according to the present embodiment, it is possible to provide a wafer manufacturing method with higher manufacturing efficiency than in the past.

[0063] (Modification example) The present disclosure is not limited to the above-described embodiment. Therefore, the above-described embodiment can be appropriately modified. Hereinafter, typical modification examples will be described. In the description of the following modification examples, the differences from the above-described embodiment will be mainly described. Also, in the above-described embodiment and the modification examples, parts that are identical or equivalent to each other are given the same reference numerals. Therefore, in the description of the following modification examples, with respect to the components having the same reference numerals as those in the above-described embodiment, the description in the above-described embodiment can be appropriately incorporated unless there is a technical contradiction or special additional explanation.

[0064] The present disclosure is not limited to the specific configuration shown in the above-described embodiment. That is, for example, there are no particular limitations on the outer diameter and planar shape of the wafer 1, that is, the ingot 2 (for example, the presence or absence of a so-called orientation flat).

[0065] There is no particular limitation on the magnitude of the off-angle θ. Also, in the above embodiment, the wafer C-plane 11 and the ingot C-plane 21 do not coincide with the C-plane in the strict crystallographic sense, that is, the (0001) plane Pc. However, even in such a case, since it is socially acceptable to refer to it as the "C-plane", the expression "C-plane" is used. The same applies to the "Si-plane". However, the present disclosure is not limited to such an aspect. That is, the wafer C-plane 11 and the ingot C-plane 21 may coincide with the C-plane in the strict crystallographic sense, that is, the (0001) plane Pc. Similarly, the wafer Si-plane 12 and the ingot Si-plane 22 may coincide with the Si-plane in the strict crystallographic sense. In other words, the off-angle θ may be 0 degrees.

[0066] Depending on the irradiation conditions and scanning conditions of the laser beam B, the peeling surface 32 may have a surface state and an edge state such that it can be satisfactorily ground or polished even when directly subjected to the ECMG process or the ECMP process. For this reason, the rough grinding process of the peeling surface 32 shown in FIG. 2 and the edge grinding process that is usually performed may be omitted. The same applies to the rough grinding of the top surface of the ingot 2 after the wafer peeling process.

[0067] The peeling layer forming apparatus 40 shown in FIGS. 4A and 4B is a simplified schematic diagram for briefly explaining the outline of the peeling layer forming process according to the present disclosure. Therefore, the specific configuration of the peeling layer forming apparatus 40 actually realized industrially does not necessarily coincide with the exemplary configuration shown in FIGS. 4A and 4B. Specifically, for example, in the peeling layer forming process shown in FIGS. 4A and 4B, the positive Z-axis direction in the figure is typically vertically upward, but the present disclosure is not limited to such an aspect. That is, for example, the positive Z-axis direction in FIGS. 4A and 4B may be the horizontal direction. In this case, although the laser irradiation surface is the "top surface" of the ingot 2, it is not the "upper surface". Also, the chuck table 41 may be configured to hold the ingot 2 by a method other than the air pressure adsorption mechanism.

[0068] In the above-described embodiment, in the peeling layer forming apparatus 40, the chuck table 41 that supports the ingot 2 is configured to be movable at least in the in-plane direction, while the condensing device 42 is fixedly provided in the in-plane direction. However, the present disclosure is not limited to such an aspect. That is, for example, in the peeling layer forming apparatus 40, the chuck table 41 that supports the ingot 2 may be fixedly provided in the in-plane direction, while the condensing device 42 may be configured to move in the in-plane direction by a scanning device (not shown). Specifically, for example, the peeling layer forming apparatus 40 may include a scanning device configured to relatively move the condensing point BP of the laser beam B with respect to the ingot 2 in the XYZ directions in the drawing. Further, depending on whether it is the facet region RF or the non-facet region RN, or regardless of this, performing distance adjustment in the Z-axis direction between the condensing device 42 on the irradiation side of the laser beam B and the chuck table 41 that supports the ingot 2 is an optional matter in the present disclosure. In addition, the specific configuration of the peeling layer forming apparatus 40 actually realized industrially can be appropriately changed from the exemplary configurations shown in FIGS. 4A and 4B.

[0069] The peeling device 50 shown in FIG. 11 is a simplified schematic diagram for briefly explaining the outline of the wafer peeling process according to the present disclosure. Therefore, the specific configuration of the peeling device 50 actually realized industrially does not necessarily coincide with the exemplary configuration shown in FIG. 11. Specifically, for example, the positive Z-axis direction in FIG. 11 is typically vertically upward, but the present disclosure is not limited to such an aspect. That is, for example, the positive Z-axis direction in FIG. 11 may be the horizontal direction. Hereinafter, various modified forms of the peeling device 50 shown in FIG. 11 will be described more specifically.

[0070] For example, if the applied load is excessive with respect to the load required for peeling, there is a concern that excessive stress may be applied to the wafer precursor 26 during peeling, resulting in cracking. Here, due to the cleavage state and variations in the material, the load required for peeling has variations. Therefore, in the wafer peeling process, it is preferable to continuously or intermittently increase the load until peeling occurs. That is, by sequentially increasing the applied load and stopping the load application when peeling can be achieved, the deviation between the load required for peeling and the applied load can be reduced, and the occurrence of cracking can be favorably suppressed.

[0071] If only a constant static load is continuously applied, the load may not be sufficiently concentrated at the end. On the other hand, if only an impact load, which is a dynamic load, is applied, although the load can be concentrated at the end, in the peeling layer 25, the peeling of the wafer precursor 26 from the ingot 2 may only proceed partway. In this case, if the impact load is applied again, there is a concern that the wafer precursor 26 may crack. Therefore, it is preferable to perform the wafer peeling process by superimposing a static load and an impact load. Specifically, for example, as shown in FIG. 12, while applying a static load F1 to the support table 51, an impact load F2 is applied by dropping a weight W onto the second table end 51c, which is one end thereof. According to such a load application method, the impact load F2 can surely apply a load from the end to establish peeling. Further, even if the peeling caused by the impact load F2 only proceeds partway, the entire surface can surely be peeled by continuously applying a constant static load F1.

[0072] The method for holding the ingot 2 in the wafer peeling process is not particularly limited. That is, for example, such a holding method may be one that uses air pressure. Specifically, for example, the support table 51 may have a number of suction holes (not shown) that open in the support fixing surface 51a, and may be configured to adsorb the ingot Si face 22 to the support fixing surface 51a by air pressure, i.e., negative pressure, at the suction holes. Similarly, for example, the peeling pad 52 may have a number of suction holes (not shown) that open in the pad fixing surface 52a, and may be configured to adsorb the ingot C face 21 to the pad fixing surface 52a by air pressure, i.e., negative pressure, at the suction holes.

[0073] However, taking into consideration the strength required to withstand the load required for peeling and the cost of the equipment, it is preferable that the ingot 2 be held in place using an adhesive 54 such as wax as shown in Figure 13. That is, it is preferable that the wafer peeling step be performed as follows. The ingot C-face 21, which is the surface at one end in the height direction of the ingot 2, that is, the top face, is joined to a peeling pad 52, which serves as a top face fixing member, using an adhesive 54. The surface on the other end side in the height direction of the ingot 2, that is, the ingot Si surface 22 which is the bottom surface, is bonded to a support table 51 which serves as a bottom surface fixing member using an adhesive 54. A load is applied to the support table 51 and / or the stripping pad 52 .

[0074] 13, it is preferable to spread the adhesive 54 until it overflows from the ingot side surface 20 provided between the ingot C-face 21 and the ingot Si-face 22 and adheres to the ingot side surface 20. By spreading the adhesive 54 over the ingot side surface 20 in this way, it is possible to improve the holding force.

[0075] When using the adhesive 54, the bonding layer formed by the adhesive 54 must be uniform. Furthermore, if air bubbles are trapped inside the bonding layer, defects may occur during the wafer peeling process. Specifically, if a peeling load is applied when air bubbles are trapped inside the bonding layer, the air bubble portion that is not fixed to the ingot 2 and the peeling mechanism may not be peeled properly, potentially resulting in cracking of the wafer precursor 26. Therefore, it is preferable to position the adhesive 54 inside the outer diameter of the ingot 2 and spread it by applying heat and / or pressure. This ensures that the bonding layer formed by the adhesive 54 after the bonding process is uniform, and effectively prevents air bubbles from being trapped inside the layer.

[0076] Fig. 14 shows an outline of a wafer peeling process using a method of holding an ingot 2 with an adhesive 54. Figs. 15A to 15H show an outline of a process of bonding an ingot 2 to a support table 51 and a peeling pad 52 by spreading the adhesive 54 with heat and pressure. Hereinafter, this wafer peeling process will be described with reference to Fig. 14 and Figs. 15A to 15H.

[0077] First, as shown in FIG. 15A, an adhesive 54 is applied to a support fixing surface 51a which is the upper surface of the support table 51. Specifically, a tablet-shaped adhesive 54 is placed on the support fixing surface 51a. At this time, the adhesive 54 is provided with a smaller diameter than the outer diameter of the ingot 2. Incidentally, it is preferable to heat the support table 51 at this time. Next, as shown in FIG. 15B, the ingot 2 is placed on the adhesive 54. At this time, the adhesive 54 is disposed inside the outer diameter of the ingot 2. Incidentally, at this time, the ingot 2 is preferably heated or pre-heated. The laminate of the support table 51, the adhesive 54, and the ingot 2 formed in this way is placed between a press lower mold 55 and a press upper mold 56 as shown in FIG. 15C. Then, such a laminate is pressed between the press lower mold 55 and the press upper mold 56. At this time, such a laminate is pre-heated. Alternatively, such a laminate is heated simultaneously with the pressing. As a result, as shown in FIG. 15D, the adhesive 54 spreads to the outside of the ingot side surface 20 by heat and pressure. Thereby, the ingot 2 and the support table 51 are adhered to each other.

[0078] The ingot 2 and the peeling pad 52 are bonded in a similar manner. First, as shown in FIG. 15E, adhesive 54 is applied to the pad fixing surface 52a, which is the bottom surface of the peeling pad 52. The size of the adhesive 54 to be applied is the same as above. Preferably, the peeling pad 52 is heated or preheated. Next, as shown in FIG. 15F, the pad fixing surface 52a, to which adhesive 54 is applied, is faced toward the ingot 2, and the peeling pad 52 is superimposed on the bonded assembly of the ingot 2 and the support table 51. The resulting laminate of the ingot 2, the support table 51, the peeling pad 52, and the adhesive 54 is placed between the lower press mold 55 and the upper press mold 56, as shown in FIG. 15G. Then, the laminate is pressed between the lower press mold 55 and the upper press mold 56. At this time, the laminate is preheated. Alternatively, the laminate is heated simultaneously with the pressing. 15H, the adhesive 54 spreads outward from the ingot side surface 20 due to heat and pressure, thereby bonding the ingot 2 and the peel pad 52 together.

[0079] 14 again, after the two bonding steps as described above, peeling is performed by applying a load to the support table 51 and / or the peeling pad 52. After peeling, the peeled body 30 attached to the peeling pad 52 side is detached from the peeling pad 52 while being heated, undergoes rough cleaning and finish cleaning using ethanol and ultrasonic waves, and is rinsed with pure water. Similarly, the ingot 2 attached to the support table 51 side is detached from the support table 51 while being heated, undergoes rough cleaning and finish cleaning using ethanol and ultrasonic waves, and is rinsed with pure water. The support table 51 and the peeling pad 52 also undergo rough cleaning and finish cleaning using ethanol and ultrasonic waves, and are rinsed with pure water.

[0080] Note that the support table 51 shown in FIGS. 15A to 15H may be a part of the support table 51, that is, a jig for joining with the ingot 2 in the support table 51. Specifically, the fixing of the ingot 2 and the support table 51 may be performed by adhering the ingot 2 to the joining jig by the above method and then fixing such a joining jig to the main body of the support table 51 by means such as screwing. The same applies to the release pad 52.

[0081] In the above embodiment, the release layer 25 is formed on the ingot C surface 21 side by "C surface side irradiation" in which the laser beam B is irradiated onto the ingot C surface 21. However, the present disclosure is not limited to such a mode. That is, the present disclosure is also applicable to "Si surface side irradiation" in which the laser beam B is irradiated onto the ingot Si surface 22 to form the release layer 25 on the ingot Si surface 22 side. In this case, the release pad 52 in the release device 50 described in FIG. 11 and the like is joined to the ingot Si surface 22. That is, the wafer peeling process is performed with the ingot Si surface 2, which is the main surface on the Si surface side of the ingot 2, as the upper surface.

[0082] When performing the wafer peeling process with the ingot Si surface 22 as the top surface and setting the posture of the ingot 2 such that the top surface in question becomes the upper surface, as shown in FIG. 16, the facet region RF is located on the higher side, i.e., the "side with a high off-angle," in the (0001) plane Pc, which is indicated by the dashed arc in the figure. In this case, it is necessary to consider that cracking is less likely to occur in the facet region RF and its vicinity, and it is difficult for cracking to extend to the end. Therefore, in this case, it is preferable to perform the wafer peeling process by applying a peeling load F in a region that does not overlap with the facet region RF in the circumferential direction along the circumference surrounding the central axis L, i.e., the region indicated by the solid arc in the figure. Such a peeling load F is directed in the direction of "pulling up" the ingot Si surface 22, i.e., the direction of separating the ingot Si surface 22 from the ingot C surface 21 on its back side. In this way, by applying a load and performing peeling from the region excluding the facet region RF among the ends on the "side with a high off-angle," good peeling of the wafer precursor 26 formed by irradiation on the Si surface side becomes possible.

[0083] It goes without saying that the elements constituting the above-described embodiments are not necessarily essential, except in cases where it is explicitly stated that they are particularly essential and cases where they are considered to be clearly essential in principle. Also, when numerical values such as the number, quantity, range, etc. of the components are mentioned, the present disclosure is not limited to those specific numerical values, except in cases where it is explicitly stated that they are particularly essential and cases where they are clearly limited to specific numerical values in principle. Similarly, when the shape, direction, positional relationship, etc. of the components are mentioned, the present disclosure is not limited to those specific shape, direction, positional relationship, etc., except in cases where it is explicitly stated that they are particularly essential and cases where they are clearly limited to specific shape, direction, positional relationship, etc. in principle.

[0084] The modification examples are not limited to the above examples. That is, for example, in addition to those exemplified above, multiple embodiments can be combined with each other as long as they do not technically conflict. Similarly, multiple modification examples can be combined with each other as long as they do not technically conflict.

Claims

1. A wafer manufacturing method for obtaining a wafer (1) from an ingot (2) made of single-crystalline SiC and having a c-axis (Lc) and a C-plane (Pc) orthogonal to each other, comprising: forming a release layer (25) at a depth corresponding to the thickness of the wafer from the top surface by irradiating a laser beam having permeability to the top surface (21) which is one end surface in the height direction of the ingot, the release layer formation; peeling the wafer precursor (26) which is a portion between the top surface and the release layer from the ingot at the release layer, the wafer peeling; planarizing the main surface (32) of the plate-like peeled body (30) obtained by the wafer peeling, the wafer planarization; including: the c-axis in the ingot is provided in a state where the central axis (L) orthogonal to the top surface is inclined by an off-angle (θ) exceeding 0 degrees in the off-angle direction (Dθ); the wafer peeling is performed by applying a load in one direction in a region that does not overlap with the facet region along the circumferential direction around the central axis, on one end (23) side of the ingot in the off-angle direction which is the higher side end on the C-plane when the attitude of the ingot is set such that the top surface becomes the upper surface with the Si-plane as the top surface and the facet region (RF) is positioned on the higher side of the C-plane; A wafer manufacturing method.

2. The wafer peeling is: joining the top surface of the ingot and a top surface fixing member (52); joining the bottom surface (22) which is the other end surface in the height direction of the ingot and a bottom surface fixing member (51); performing by applying a load to the top surface fixing member and / or the bottom surface fixing member; the joining of the top surface and the top surface fixing member and the joining of the bottom surface and the bottom surface fixing member are performed by spreading an adhesive (54) arranged inside the outer diameter of the ingot by heat and / or pressure; The wafer manufacturing method according to Claim 1.

3. In the joining of the top surface and the top surface fixing member and the joining of the bottom surface and the bottom surface fixing member, the adhesive is spread until it protrudes from the side surface (20) provided between the top surface and the bottom surface in the ingot and adheres to the side surface. The wafer manufacturing method according to Claim 2.

4. A wafer manufacturing method for obtaining a wafer (1) from an ingot (2) made of single-crystalline SiC and having a c-axis (Lc) and a C-plane (Pc) orthogonal to each other, comprising: By irradiating a laser beam having permeability with respect to the top surface (21) which is one end surface in the height direction of the ingot, a release layer (25) is formed at a depth corresponding to the thickness of the wafer from the top surface. Release layer formation, The wafer precursor (26) which is a portion between the top surface and the release layer is peeled from the ingot at the release layer. Wafer peeling, The main surface (32) of the plate-like peeled body (30) obtained by the wafer peeling is planarized. Wafer planarization, including The c-axis in the ingot is provided in a state where the central axis (L) orthogonal to the top surface is inclined by an off-angle (θ) exceeding 0 degrees in the off-angle direction (Dθ). The wafer peeling is joining the top surface of the ingot and the top surface fixing member (52), joining the bottom surface (22) which is the other end surface in the height direction of the ingot and the bottom surface fixing member (51), applying a load to the top surface fixing member and / or the bottom surface fixing member so that a load is applied in one direction on one end (23) side of the ingot in the off-angle direction. This is done by The joining of the top surface and the top surface fixing member, and the joining of the bottom surface and the bottom surface fixing member are performed by spreading an adhesive (54) arranged inside the outer diameter of the ingot by heat and / or pressure until it protrudes from the side surface (20) provided between the top surface and the bottom surface of the ingot and adheres to the side surface. Wafer manufacturing method.

5. The one end of the ingot in the off-angle direction is the higher side end on the C plane when the posture of the ingot is set such that the top surface is the upper surface. The wafer manufacturing method according to claim 4.

6. The release layer formation is performed such that the facet region (RF) is located on the lower side of the C plane when the posture of the ingot is set such that the top surface is the upper surface. The wafer manufacturing method according to claim 4 or 5.

7. The peeling layer formation is performed by irradiating the top surface with a laser beam while moving the irradiation position (PR) of the laser beam on the top surface along a first direction (Ds) along the top surface, and performing laser scanning a plurality of times while changing the position in a second direction (Df) orthogonal to the first direction on the top surface and along the top surface. By forming a plurality of scanning lines (Ls), which are irradiation marks of the linear laser beam along the first direction, along the second direction, the peeling layer is formed. When moving the irradiation position in the first direction, the scanning lines are formed across both ends of the top surface in the first direction. When moving the irradiation position in the direction opposite to the first direction, the irradiation marks are formed at the ends of the top surface in the first direction. The wafer manufacturing method according to any one of claims 4 to 6.

8. The wafer peeling is performed by continuously or intermittently increasing the load until peeling occurs. The wafer manufacturing method according to any one of claims 1 to 7.

9. The wafer peeling is performed by superimposing a static load and an impact load. The wafer manufacturing method according to any one of claims 1 to 8.

Images