Method for manufacturing a wafer

Abstract

To provide a wafer manufacturing method capable of improving yield.SOLUTION: In a wafer manufacturing method in which a wafer 20 is obtained by cutting an ingot 2 with a laser beam L, a plurality of locations of the ingot 2 are irradiated with a laser beam L from a direction intersecting a planned cutting surface 21 to form a plurality of modified portions, and cracks are extended starting from the modified portions, thereby cutting out the wafer 20 from the ingot 2. The energy density, which is the energy per unit area of the laser light L in the ingot 2, exceeds a modification threshold on the planned cutting surface 21. At least one of the following conditions 1, 2, and 3 is satisfied with respect to the energy density. The condition 1 is that the peak value of the energy density is 44 J / cm2 or less, the condition 2 is that the energy density increase rate is 1000 J / cm3 or more, and the condition 3 is that the depth range in which the energy density exceeds the modification threshold is 30 μm or less.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a method for manufacturing a wafer.

Background Art

[0002] As a method for generating a wafer from a semiconductor ingot, there is a method of irradiating the ingot with a laser beam and slicing it. That is, a laser beam is irradiated onto the ingot, and modified portions are formed at a large number of locations along the planned cutting plane in the ingot. Then, cracks are generated starting from these modified portions to separate the wafer from the ingot.

[0003] However, when forming these numerous modified portions, there may occur a phenomenon in which modified portions are also formed at positions shallower than the target position. That is, while modified portions are formed at the targeted depth position, modified portions may also be formed at other depth positions. This multi-stage formation of modified portions leads to an increase in material loss of the ingot and is a factor in reducing the yield.

[0004] Patent Document 1 discloses a technique for accurately forming a fine modified region in a predetermined portion by condensing an annular portion of a laser beam on the predetermined portion of an object to be processed.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, simply irradiating the annular portion of the laser beam alone makes it difficult to suppress the multi-stage formation of modified portions. Therefore, the problem of improving the yield of the wafer remains.

[0007] The present invention has been made in view of such problems, and aims to provide a method for manufacturing a wafer that can improve the yield.

Means for Solving the Problems

[0008] One aspect of the present invention is a method for manufacturing a wafer, in which a transparent or translucent ingot (2) is cut by a laser beam (L) to obtain a wafer (20), from a direction intersecting the planned cutting surface (21) where the ingot is to be cut, the laser beam is irradiated at a plurality of locations on the ingot, and a plurality of modified portions (31) are formed at the depth position of the planned cutting surface inside the ingot. By extending cracks (32) starting from the modified portions, the wafer is cut out from the ingot, The energy density, which is the energy per unit area of the laser beam in the ingot when the laser beam is irradiated on the ingot, exceeds a modification threshold value (Eth) that can modify a part of the ingot at the planned cutting surface, Regarding the energy density, further, the following Conditions 1, 2, and 3 All are satisfied, Condition 1 is that the peak value (Ep) of the energy density is 44 J / cm 2 or less, Condition 2 is that the energy density increase rate (α), which is the increase width of the energy density per unit depth at the shallowest depth position where the energy density reaches the modification threshold value, is 1000 J / cm 3 or more, Condition 3 is that the depth range (W) where the energy density exceeds the modification threshold value is 30 μm or less. This is the method for manufacturing a wafer.

Advantages of the Invention

[0009] In the method for manufacturing a wafer as described above, the energy density of the laser beam irradiated onto the ingot is controlled as described above. As a result, while suppressing the multi-staging of the modified portions, the modified portions can be formed at desired depth positions. Therefore, it is possible to suppress the material loss of the ingot and improve the yield of the wafer.

[0010] As described above, according to the above aspect, it is possible to provide a method for manufacturing a wafer capable of improving the yield. Note that the reference numerals in parentheses described in the claims and the means for solving the problems indicate the correspondence with the specific means described in the embodiments described later, and do not limit the technical scope of the present invention.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Mode for Carrying Out the Invention

[0012] (Embodiment 1) Embodiments of the method for manufacturing a wafer will be described with reference to FIGS. 1 to 11. The method for manufacturing a wafer according to this embodiment is a method of cutting an ingot 2 with a laser beam L to obtain a wafer 20 as shown in FIGS. 1 to 9. Here, the ingot 2 is transparent or translucent.

[0013] First, as shown in FIGS. 1 and 2, laser light L is irradiated onto a plurality of locations of the ingot 2 from a direction intersecting the planned cutting surface 21 where the ingot 2 is to be cut. As a result, as shown in FIGS. 2 and 3, a plurality of modified portions 31 are formed at the depth position of the planned cutting surface 21 inside the ingot 2. Then, as shown in FIG. 5, a crack 32 is extended starting from the modified portion 31. As a result, as shown in FIGS. 7 and 8, the wafer 20 is cut out from the ingot 2.

[0014] The energy density, which is the energy per unit area of the laser light L in the ingot 2 when the laser light L is irradiated onto the ingot 2, satisfies the following conditions. That is, the energy density of the laser light L exceeds the modification threshold value Eth at the planned cutting surface 21. Here, the modification threshold value Eth is the threshold value of the energy density capable of modifying a part of the ingot 2.

[0015] Regarding the energy density, further, at least one of the following Conditions 1, 2, and 3 is satisfied. Condition 1: The peak value of the energy density is 44 J / cm 2 or less. Condition 2: The energy density increase rate at the shallowest depth position where the energy density reaches the modification threshold value Eth is 1000 J / cm 3 or more. Here, the energy density increase rate is the increase width of the energy density per unit depth. Condition 3: The depth range in which the energy density exceeds the above modification threshold value Eth is 30 μm or less.

[0016] Regarding the above conditions related to the energy density, an explanation will be given with reference to FIG. 10. Curve M in the figure schematically shows an example of the relationship between the depth from the surface of ingot 2 and the energy density of laser beam L when the ingot 2 is irradiated with the laser beam L. Here, the surface of the ingot 2 means the surface that serves as the incident surface of the laser beam L. The vertical axis in the figure represents the depth from the surface of the ingot 2, indicating that it becomes deeper as it goes downward. The horizontal axis represents the energy density of the laser beam, indicating that the energy density increases as it goes to the right.

[0017] As shown in the figure, the ingot 2 will be irradiated with the laser beam L so that the energy density increases at a certain depth position. Here, the energy density is made to be at least equal to or higher than the modification threshold value Eth at the depth position of the planned cutting surface 21. On the other hand, at least one of the above Conditions 1 to 3 is satisfied. More preferably, all of the above Conditions 1 to 3 are satisfied.

[0018] As Condition 1, the peak value Ep of the energy density is made to be 44 J / cm 2 or less. In the following, the peak value Ep of the energy density will also be simply referred to as "peak value Ep".

[0019] Further, as Condition 2, the energy density increase rate at the shallowest depth position where the energy density reaches the modification threshold value Eth is made to be 1000 J / cm 3 or more. That is, in FIG. 10, the slope of the tangent line T of curve M at point P1 is made to be 1000 J / cm 3 or more. At point P1, the increase width of the energy density that increases for each unit depth increase is 1000 J / cm 3 or more. That is, the energy density increase rate at the shallowest depth position where the energy density reaches the modification threshold value Eth is 1000 J / cm 3 or more. In the following, the energy density increase rate at the shallowest depth position where the energy density reaches the modification threshold value Eth will also be simply referred to as "energy density increase rate α".

[0020] Also, as Condition 3, the depth range in which the energy density exceeds the modification threshold Eth is set to 30 μm or less. In the figure, this means that the length of the range indicated by the symbol W is 30 μm or less. Hereinafter, the depth range in which the energy density exceeds the modification threshold Eth is also simply referred to as the "depth range W".

[0021] In this embodiment, the ingot 2 is made of SiC (that is, silicon carbide). And as shown in FIG. 1, the ingot 2 has a substantially cylindrical shape. The laser beam L is irradiated from a surface corresponding to one of the pair of substantially cylindrical bottom surfaces onto the ingot 2 of such a material and shape. The surface of the ingot 2 on which the laser beam L is incident is conveniently referred to as the upper surface 23. The upper surface 23 of the ingot 2 is a flat surface. In this embodiment, the laser beam irradiated onto the ingot 2 is a pulsed laser beam with a pulse width of 250 fs to 10 nm.

[0022] As shown in FIGS. 1 and 2, the planned cutting surface 21 is parallel to the upper surface 23. Then, the planned cutting surface 21 is set at a depth position from the upper surface 23 according to the thickness of the wafer 20 to be obtained. The laser beam L is irradiated from the upper surface 23 of the ingot 2 so that a large number of modified portions 31 are formed on this planned cutting surface 21. The laser beam L is irradiated through the condensing lens 41 so as to have an energy density equal to or higher than the modification threshold Eth at the depth position of the planned cutting surface 21. The modified portion 31 is formed when a part of the ingot 2 made of a single crystal of SiC is separated into amorphous Si (silicon) and amorphous C (carbon) by the energy of the laser beam L.

[0023] As shown in FIGS. 1 to 3, the laser beam L is focused on a number of locations on the planned cutting surface 21 in the ingot 2 to form the modified portions 31. Here, as shown in FIGS. 3(a) and (b), a number of modified portions 31 are formed along each of a plurality of virtual straight lines VL parallel to each other on the planned cutting surface 21. That is, the laser beam L is scanned along the virtual straight line VL. This virtual straight line VL can be set in a direction orthogonal to the formation direction of the off-angle θ in the ingot 2 when viewed from the axial direction of the ingot 2. In FIG. 3, only a part of the number of modified portions 31 is shown. Actually, the modified portions 31 on the virtual straight line VL are densely formed so as to partially overlap each other. Also, in FIG. 3(b), the broken line indicated by the reference numeral 2c represents the c-plane inclined at the off-angle θ with respect to the upper surface 23 of the ingot 2. The same applies to FIG. 4.

[0024] However, as shown in FIGS. 4(a) and (b), the virtual straight line VL can also be set as a straight line parallel to the formation direction of the off-angle θ in the ingot 2 when viewed from the axial direction of the ingot 2. In this case, the inclination of the off-angle θ appears in a cross section parallel to both the axial direction of the ingot 2 and the virtual straight line LV. Along such a virtual straight line LV, the pulsed laser beam L is scanned. Thereby, a number of modified portions 31 are formed on each of the number of virtual straight lines VL.

[0025] When the modified portions 31 are formed in the ingot 2, as shown in FIG. 5, cracks 32 occur starting from the modified portions 31. These cracks 32 extend along the c-plane of the ingot 2. The c-plane is a plane inclined at a certain off-angle θ with respect to the upper surface 23 and is a plane existing innumerably in the ingot 2. In this embodiment, for example, the off-angle θ can be set to 4°.

[0026] Therefore, the crack 32 is formed at an inclination of, for example, 4° with respect to the upper surface 23. As described above, since a large number of modified portions 31 are formed on the planned cutting surface 21, by appropriately setting the arrangement of the large number of modified portions 31, the cracks 32 are continuously connected. As a result, on a surface substantially along the planned cutting surface 21, as shown in FIGS. 7 and 8, the wafer 20 is separated from the ingot 2.

[0027] However, as described above, the crack 32 is inclined with respect to the upper surface 23 and is also inclined with respect to the planned cutting surface 21. Therefore, as shown in FIG. 8, the cut surfaces 321 and 322 formed by connecting the cracks 32 are uneven surfaces with a large number of irregularities formed thereon. The uneven cut surfaces 321 and 322 generated on the wafer 20 and the ingot 2 are polished using a grindstone or the like. As a result, as shown in FIG. 9, the wafer 20 is obtained and the upper surface 23 of the ingot 2 after cutting is flattened.

[0028] Note that, as described above, when the virtual straight line VL is parallel to the formation direction of the off-angle θ (see FIGS. 4(a) and 4(b)), the irregularities can be made minute. That is, in this case, as shown in FIG. 6, the modified portions 31 adjacent to each other on the virtual straight line VL are continuous. These modified portions 31 are partially arranged along the formation direction of the off-angle θ, but are within the above-described depth range W. Therefore, the irregularities can be made extremely small.

[0029] In this embodiment, when forming the modified portion 31, as shown in FIG. 11, the ingot 2 is simultaneously irradiated with laser light from a plurality of directions. The laser light from the plurality of directions overlaps with each other in a part of the depth region including the depth position of the planned cutting surface 21. The overlapping portion Lc has a length h in the depth direction of 5 to 50 μm. Here, the depth direction coincides with the normal direction with respect to the upper surface 23 of the ingot 2.

[0030] Note that "simultaneously irradiating the ingot 2 with laser light from a plurality of directions" includes both the case where one laser light is irradiated from a plurality of directions and the case where a plurality of laser lights are irradiated from a plurality of directions.

[0031] In this embodiment, the laser beam is an annular laser beam. Here, the annular laser beam refers to a beam whose cross-sectional shape perpendicular to the optical axis outside the overlapping portion Lc is annular as shown in Fig. 12. That is, the annular laser beam has an annular intensity distribution outside the overlapping portion Lc. Such an annular laser beam also becomes a laser beam that irradiates the ingot 2 from a plurality of directions simultaneously.

[0032] Particularly in this embodiment, the cross-sectional shape (i.e., the intensity distribution) is annular. Such an annular laser beam has an annular cross-sectional shape before being incident on the condenser lens 41. This annular laser beam is refracted by the condenser lens 41 as shown in Fig. 11 and forms the overlapping portion Lc inside the ingot 2. The overlapping portion Lc has a length h in the depth direction of 5 to 50 μm. Also, the overlapping portion Lc is formed in a depth region including the depth position of the planned cutting surface 21.

[0033] By irradiating the laser beam so as to provide the overlapping portion Lc as described above, control of the energy density of the laser beam irradiating the ingot 2 is realized.

[0034] Next, the operation and effect of this embodiment will be described. In the above method for manufacturing a wafer, the energy density of the laser beam L irradiating the ingot 2 is controlled as described above. That is, after ensuring that the energy density of the laser beam L is at least equal to or higher than the modification threshold value Eth at the depth position of the planned cutting surface 21, the energy density of the laser beam L is controlled to satisfy at least one of the above conditions 1, 2, and 3. That is, at least one of the peak value Ep of the energy density of the laser beam L, the energy density increase rate α, and the depth range W is controlled to be within a predetermined range (see Fig. 10). Thereby, while suppressing the multi-stage formation of the modified portion 31, the modified portion 31 can be formed at a desired depth position. Therefore, the material loss of the ingot 2 can be suppressed and the yield of the wafer 20 can be improved.

[0035] That is, by controlling the energy density of the laser beam L irradiated on the ingot 2 so as to satisfy at least one of Condition 1, Condition 2, and Condition 3, it is possible to suppress the multi-stage formation of the modified portion 31 formed in the ingot 2. If the energy density of the laser beam L does not satisfy any of the above Condition 1, Condition 2, and Condition 3, there is a high possibility that multi-stage formation of the modified portion will occur significantly (see FIG. 13(b) described later). When multi-stage formation of the modified portion 31 occurs due to laser beam irradiation at a large number of locations, the material of the ingot 2 becomes wasted by the thickness at which the multi-stage formation has occurred. Therefore, by controlling each parameter regarding the energy density of the laser beam as described above as in this embodiment, material loss can be suppressed and the yield of the wafer 20 can be improved.

[0036] Here, the mechanism of multi-stage formation of the modified portion and its suppression will be described. For example, when a laser beam having a Gaussian distribution as shown in FIG. 13(a) is condensed and irradiated on the ingot 2, as shown in FIG. 13(b), the inventors of the present application have found that modified portions 311 and 312 may be formed at a plurality of locations in the depth direction. This is because the light in the portion of the laser beam close to the optical axis is condensed at a relatively shallow position, and when the energy density exceeds the modification threshold value Eth, the modified portion 311 is formed. At the same time, the light in the portion of the laser beam far from the optical axis wraps around from the outer peripheral side of the modified portion 311 and is condensed at a relatively deep position, and when the energy density exceeds the modification threshold value Eth, the modified portion 312 is formed. In this way, as shown in FIG. 14, the energy density exceeds the modification threshold value Eth at a plurality of locations in the depth direction in the ingot 2. As a result, as shown in FIG. 13(b), it is considered that the modified portions 311 and 312 are formed in multiple stages.

[0037] In contrast, in the present embodiment, for example, as shown in Fig. 15(a), an annular laser beam having no energy distribution near the optical axis is condensed, and the ingot 2 is irradiated so as to form an overlapping portion at the target depth position as shown in Fig. 15(b). In this case, at the target depth position, that is, at the depth position of the planned cutting surface 21, as shown in Fig. 16, the energy density exceeds the modification threshold value Eth, and as shown in Fig. 15(b), the modified portion 31 is formed. At this time, as shown in Fig. 16, at depths other than the depth position including the depth position of the planned cutting surface 21, there is no depth at which the energy density of the laser beam exceeds the threshold value and becomes high. Therefore, in this case, the multi-stage formation of the modified portion can be suppressed.

[0038] Under such a mechanism, by controlling to satisfy at least one of Condition 1, Condition 2, and Condition 3, the multi-stage formation of the modified portion can be suppressed. Also, regarding the energy density, by controlling to satisfy all of Condition 1, Condition 2, and Condition 3, the multi-stage formation of the modified portion can be further suppressed.

[0039] Also, the laser beam L irradiated to the ingot 2 is a pulsed laser beam with a pulse width of 250 fs to 10 n s This makes it possible to form a sufficient modified portion at a desired depth position while suppressing the multi-stage formation of the modified portion.

[0040] Also, the length h of the overlapping portion Lc of the laser beams from a plurality of directions with respect to the ingot 2 is 5 to 50 μm (see Fig. 11). This makes it possible to easily and surely control each parameter of the energy density within a predetermined range. As a result, the yield of the wafer 20 can be easily and surely improved.

[0041] As described above, according to the present embodiment, it is possible to provide a method for manufacturing a wafer that can improve the yield.

[0042] In addition, when the virtual straight line VL is set as a straight line parallel to the formation direction of the off-angle θ in the ingot 2 as viewed from the axial direction of the ingot 2, as described above, the height of the unevenness of the cut surfaces 321 and 322 can be made extremely small. As a result, the material loss of the ingot 2 can be further suppressed.

[0043] (Experimental Example) In this example, the effect of preventing the multi-staging of the modified portion by controlling the peak value Ep, the energy density increase rate α, and the depth range W within the ranges shown in the above-described Embodiment 1 was confirmed.

[0044] That is, as described above, the inventors have found that by controlling the peak value Ep, the energy density increase rate α, and the depth range W within predetermined ranges, respectively, the multi-staging of the modified portion can be suppressed. The predetermined ranges are reorganized as follows. [Condition 1]: The peak value Ep is 44 J / cm 2 or less. [Condition 2]: The energy density increase rate α is 1000 J / cm 3 or more. [Condition 3]: The depth range W is 30 μm or less.

[0045] Therefore, in this example, a sample A in which a laser beam was irradiated onto the ingot so as to satisfy all of Conditions 1 to 3, and samples B, C, and D in which a laser beam was irradiated onto the ingot so as to satisfy any one or more of Conditions 1 to 3 were produced. Further, a sample E in which a laser beam was irradiated onto the ingot so that none of Conditions 1 to 3 was satisfied was produced. Specifically, it will be described below. Note that methods and the like not particularly shown in this example are the same as those in Embodiment 1. Note that the scanning direction of the laser beam (that is, the direction of the virtual straight line VL) was set as the direction orthogonal to the formation direction of the off-angle (see Fig. 3(b)).

[0046] In each of Samples A to E, the ingot was irradiated with pulsed laser light at multiple locations. At this time, the pulsed laser light was continuously irradiated in a linear manner along the virtual straight line VL (see Fig. 3) at multiple locations. The irradiation pitch was set to 0.5 μm. Also, a large number of virtual straight lines VL were provided, and the arrangement pitch of the virtual straight lines VL was set to 100 μm.

[0047] Also, the wavelength of the pulsed laser light to be irradiated was 1030 nm, the pulse width was 10 ps, and the oscillation frequency was 10 kHz. Below, each condition variously changed by Samples A to E will be described.

[0048] <Sample A> As described above, in order to satisfy all of Conditions 1 to 3, the irradiation conditions of the laser light were set as follows. That is, as the condenser lens, a single axicon lens was used to condense the annular laser light and irradiate the ingot. The outer diameter of the irradiated laser light before being condensed by the condenser lens was 2 mm, and the inner diameter was 1 mm. Also, the pulse energy was set to 3 μJ.

[0049] The state of the energy density of the laser light irradiated into the ingot at this time (that is, the relationship between the depth from the upper surface of the ingot and the energy density) is shown in Fig. 17. And the actual data for Conditions 1 to 3 at this time are shown in Table 1.

[0050]

Table 1

[0051] In Table 1, the experimental values are the values read from the graph of Fig. 17 above, and are values calculated based on the outer diameter of the laser light incident on the condenser lens, etc., the NA value of the condenser lens, the distance between the condenser lens and the ingot, the depth from the upper surface of the ingot, the refractive index of the ingot, etc. Also, in the "Pass / Fail" column, "〇" indicates that the target value of each condition is satisfied, and "×" indicates that the target value of each condition is not satisfied. The same applies to Tables 2 to 5 described later.

[0052] Then, regarding the obtained sample A, the ingot was cut and its interior was observed. A metallographic photograph (magnification of about 100 times) thereof is shown in Fig. 18. Two photographs are shown in the figure, and these are cross-sectional photographs of two different locations in sample A. The same applies to the following similar photographs. In each photograph, the part indicated by arrow 31 is the modified part. As can be seen from these photographs, no multi-staging of the modified part was observed in sample A.

[0053] <Sample B> When producing sample B, the pulse energy of the laser beam irradiated on the ingot was set to 6 μJ. Other irradiation conditions were the same as those for sample A.

[0054] The state of the energy density of the laser beam irradiated into the ingot at this time (that is, the relationship between the depth from the upper surface of the ingot and the energy density) is shown in Fig. 19. Note that the graph in the figure is based on the premise that the laser beam is irradiated once into the ingot. The same applies to Figs. 21, 23, 25, and further Fig. 17 described later. Then, the actual data for conditions 1 to 3 at this time are shown in Table 2.

[0055]

Table 2

[0056] As can be seen from Table 2, sample B does not satisfy condition 1, but satisfies conditions 2 and 3. That is, the energy density increase rate α is sufficiently large and the depth range W is sufficiently small.

[0057] Then, for the obtained sample B, the ingot was cut and the inside was observed. The metal micrographs (magnification of about 100 times) are shown in Fig. 20. As can be seen from these photographs, multi-staging of the modified portions was observed in sample B. That is, it was confirmed that the modified portions 311 and 312 were formed at two positions in the depth direction. However, the scale of the multi-staging of the modified portions was small and the modified range was also small. Here, the modified range refers to the length range in the depth direction in which the modified portions were formed in the ingot.

[0058] <Sample C> When producing sample C, the laser beam irradiated on the ingot was a Gaussian beam. That is, a laser beam having a Gaussian distribution with the intensity centered on the optical axis was used. The outer diameter of the irradiated laser beam before being focused by the condenser lens was 2.2 mm. Also, the pulse energy was 3 μJ.

[0059] The state of the energy density of the laser beam irradiated into the ingot at this time (that is, the relationship between the depth from the upper surface of the ingot and the energy density) is shown in Fig. 21. And the actual data for conditions 1 to 3 at this time are shown in Table 3.

[0060]

Table 3

[0061] As can be seen from Table 3, sample C does not satisfy conditions 2 and 3, but satisfies condition 1. That is, the peak value Ep of the energy density is sufficiently suppressed.

[0062] Then, for the obtained sample C, the ingot was cut and the inside was observed. The metal micrographs (magnification of about 100 times) are shown in Fig. 22. As can be seen from these photographs, multi-staging of the modified portions was observed in sample C. Also, in this sample C, it was confirmed that the modified portions 311, 312, and 313 were formed at three positions in the depth direction. However, compared with sample E described later, the scale of the multi-staging of the modified portions was small and the modified range was also relatively small.

[0063] <Sample D> When preparing Sample D, for the laser beam irradiated on the ingot, the outer diameter before being condensed by the condenser lens was set to 1.8 mm and the inner diameter was set to 0.2 mm. Other irradiation conditions were the same as those for Sample A.

[0064] The state of the energy density of the laser beam irradiated into the ingot at this time (that is, the relationship between the depth from the upper surface of the ingot and the energy density) is shown in Fig. 23. And the actual data for Conditions 1 to 3 at this time are shown in Table 4.

[0065]

Table 4

[0066] As can be seen from Table 4, Sample D does not satisfy Condition 2, but satisfies Condition 1 and Condition 3. That is, the peak value Ep of the energy density is sufficiently suppressed and the depth range W is also sufficiently small.

[0067] Then, for the obtained Sample D, the ingot was cut and the inside was observed. The metallographic micrograph (magnification of about 100 times) is shown in Fig. 24. As can also be seen from these photographs, multi-stage modification was observed in Sample D. That is, it was confirmed that the modified parts 311 and 312 were formed at two positions in the depth direction. However, the scale of the multi-stage modification of the modified part is small and the modified range is also small.

[0068] <Sample E> When preparing Sample E, the laser beam irradiated on the ingot was a Gaussian beam. That is, a laser beam having a Gaussian distribution with the optical axis as the center of intensity was used. The outer diameter of the irradiated laser beam before being condensed by the condenser lens is 2.2 mm. Also, the pulse energy was set to 9 μJ.

[0069] The state of the energy density of the laser beam irradiated into the ingot at this time (that is, the relationship between the depth from the upper surface of the ingot and the energy density) is shown in Fig. 25. And the actual data for Conditions 1 to 3 at this time are shown in Table 5.

[0070]

Table 5

[0071] As can be seen from Table 5, Sample E does not satisfy any of Conditions 1, 2, and 3. That is, the peak value Ep of the energy density is too large, the energy density increase rate α is too small, and the depth range W is too large.

[0072] Then, for the obtained Sample E, the ingot was cut and the inside was observed. The metallographic micrograph (magnification of about 100 times) is shown in Fig. 26. As can also be seen from this photograph, in Sample E, multi-stage formation of the modified portion has occurred significantly. In this Sample E, modified portions 311, 312, 313, and 314 are formed at three or more locations in the depth direction, and the modified range is also large.

[0073] Summarizing the conditions satisfied by the irradiated laser beam and the cross-sectional observation results during the production of Samples A, B, C, D, and E, it becomes as shown in Table 6.

[0074]

Table 6

[0075] In Table 6, "〇" means that each condition is satisfied, and "×" means that each condition is not satisfied. As can be seen from the table, for those that do not satisfy any of Conditions 1 to 3, significant multi-stage formation of the modified portion occurred. In contrast, for those that satisfy at least one of Conditions 1 to 3, multi-stage formation of the modified portion is suppressed. Among them, for those that satisfy two or more of Conditions 1 to 3, multi-stage formation of the modified portion is more suppressed. Furthermore, for those that satisfy all of Conditions 1 to 3, multi-stage formation of the modified portion did not occur. From these results, it can be seen that by satisfying at least one of Conditions 1 to 3, multi-stage formation of the modified portion can be effectively suppressed. Also, it can be seen that by satisfying all of Conditions 1 to 3, multi-stage formation of the modified portion can be effectively prevented. Also, from the results of Sample A, Sample C, and Sample D above, at least under the irradiation conditions of this experimental example, the peak value Ep of the energy density is 30 J / cm 2 It is considered effective to satisfy the following.

[0076] (Embodiment 2) This embodiment is a form in which a plurality of laser beams L1 and L2 having non-parallel optical axes with respect to each other are used, as shown in FIG. 27. The plurality of laser beams L1 and L2 are simultaneously irradiated onto the ingot 2 so as to overlap on the planned cutting surface 21.

[0077] In this embodiment, a plurality of laser beams L1 and L2 having non-parallel optical axes with respect to each other are focused by condenser lenses 41 respectively, and are overlapped on the planned cutting surface 21. The optical axes of the respective laser beams L1 and L2 are inclined with respect to the normal direction of the planned cutting surface 21. In the overlapping portion Lc of the plurality of laser beams L1 and L2, the energy density becomes large. Thereby, a modified portion 31 is formed on the planned cutting surface 21.

[0078] Before being focused by the condenser lens 41, each of the laser beams L1 and L2 can be, for example, a Gaussian beam or an annular beam. Also, the plurality of laser beams L1 and L2 can be obtained by branching laser light oscillated from the same laser oscillator and making them travel different optical paths.

[0079] The rest is the same as in Embodiment 1. Among the reference numerals used in Embodiments 2 and later, those that are the same as the reference numerals used in the previously described embodiments represent the same components and the like as those in the previously described embodiments unless otherwise specified.

[0080] In the case of this embodiment, by arranging the optical axes of the plurality of laser beams L1 and L2 on a plane orthogonal to the scanning direction of the laser beam with respect to the ingot 2, even at the edge portion of the planned cutting surface 21, it is easy to make the entire laser beams L1 and L2 incident from the upper surface 23 of the ingot 2. As a result, a decrease in the energy density of the laser beam on the planned cutting surface 21 can be suppressed. In addition, it has the same operational effects as in Embodiment 1. Note that in this embodiment, laser beams having three or more non-parallel optical axes to each other can also be used.

[0081] (Embodiment 3) In this embodiment, as shown in FIG. 28, the intensity distribution of each laser beam is made to spread in a direction orthogonal to the virtual straight line VL rather than in the direction along the virtual straight line VL. As described in Embodiment 1, the ingot 2 has a substantially cylindrical shape (see FIG. 1). Then, a large number of modified portions 31 are formed along each of a plurality of virtual straight lines VL that are parallel to each other and orthogonal to the axial direction of the ingot 2 on the planned cutting surface 21 (see FIG. 3).

[0082] In this embodiment, as shown in FIG. 28, the intensity distribution of the laser beam for forming each modified portion 31 as viewed from the axial direction spreads more in the direction orthogonal to the virtual straight line VL than in the direction along the virtual straight line VL. The elliptical contour Ld21 shown in the figure indicates the contour of the laser beam on the planned cutting surface 21, and the elliptical contour Ld23 indicates the contour of the intensity distribution of the laser beam on the upper surface 23 of the ingot 2. Here, the contour of the intensity distribution is a closed curve surrounding the portion with a predetermined intensity or higher. And, as shown in the figure, if the length in the direction orthogonal to the virtual straight line VL in the contour Ld23 is d1 and the length in the direction along the virtual straight line VL is d2, then d1>d2. Other aspects are the same as those in the first embodiment.

[0083] In this embodiment, it is easy to prevent the energy density of the laser beam irradiated near the edge of the planned cutting surface 21 of the ingot 2 from becoming too small compared to the energy density of the laser beam at other locations.

[0084] Here, as a comparative form for this embodiment, as shown in FIG. 29, consider the case where the intensity distribution of the laser beam is a circular distribution with d1 = d2 or an elliptical distribution with d1 < d2. In this case, a part of the laser beam irradiated near the edge of the ingot 2 is likely to enter from the side surface of the ingot 2. Because if d2 is large, as shown in FIG. 29, a part of the contour Ld23 of the laser beam on the plane including the upper surface 23 of the ingot 2 is likely to protrude outside the contour of the edge of the ingot 2. The protruding part of the laser beam will enter from the side surface of the ingot 2.

[0085] As a result, the energy density of the laser beam on the planned cutting surface 21 may decrease. That is, the energy density of the laser beam irradiated near the edge part may become smaller than the energy density of the laser beam irradiated on the inner side thereof. Then, the modification state near the edge part may become insufficient.

[0086] On the other hand, as in this embodiment, by making the intensity distribution of the laser beam elliptical with d1 > d2, it is possible to suppress the incidence of the laser beam irradiated near the edge of the planned cutting surface 21 from the side surface of the ingot 2. That is, even if the optical axis of the laser beam is brought closer to the edge than in the case of an annular laser beam, it is less likely to cause a decrease in the energy density on the planned cutting surface 21. As a result, the above problems can be suppressed. In addition, it has the same operational effects as those of the first embodiment.

[0087] (Embodiment 4) In this embodiment, as shown in FIG. 30, the pitch of the irradiation points Lp of the adjacent laser beams along the virtual straight line VL is made shorter at the edge of the ingot 2 than at other parts.

[0088] As described in the third embodiment, part of the laser beam irradiated on the edge of the ingot 2 is likely to enter from the side surface of the ingot 2. Therefore, the energy density on the planned cutting surface 21 is likely to decrease more than at other locations. Thus, in this embodiment, at the edge of the ingot 2, the pitch of the irradiation points Lp of the laser beam is made shorter. Thereby, a sufficient modified portion 31 can be formed even at the edge of the planned cutting surface 21. Other configurations and operational effects are the same as those of the first embodiment.

[0089] (Embodiment 5) In this embodiment, as shown in FIG. 31, the energy of the laser beam irradiated on the edge of the ingot 2 is made larger than the energy of the laser beam irradiated on other parts. Note that the size of the dot Lq in FIG. 31 represents the magnitude of the energy of each laser beam. In this embodiment, as described above, a laser beam with high energy is irradiated on the edge of the ingot 2 where the energy density on the planned cutting surface 21 is likely to decrease. Thereby, a sufficient modified portion 31 can be formed even at the edge of the planned cutting surface 21. Other configurations and operational effects are the same as those of the first embodiment.

[0090] (Embodiment 6) In this embodiment, as shown in FIG. 32, the virtual straight line VL at a position far from the central axis 2C of the ingot 2 has a narrower arrangement interval than the virtual straight line VL at a position closer to the central axis 2C. In this embodiment, as described above, the virtual straight line VL is concentrated on the edge portion of the ingot 2 where the energy density on the planned cutting surface 21 is likely to decrease. Then, laser light is irradiated along this virtual straight line VL. As a result, a sufficient modified portion 31 can be formed even at the edge portion of the planned cutting surface 21. Other configurations and operational effects are the same as those in Embodiment 1.

[0091] The present invention is not limited to the above-described embodiments, and can be applied to various embodiments without departing from the gist thereof.

Explanation of Reference Numerals

[0092] 2 Ingot 20 Wafer 21 Planned cutting surface 31 Modified portion 32 Crack Eth Modification threshold Ep Peak value α Energy density increase rate W Depth range

Claims

1. A method for manufacturing a wafer, comprising cutting a transparent or translucent ingot (2) with a laser beam (L) to obtain a wafer (20), wherein: the laser beam is irradiated onto a plurality of locations of the ingot from a direction intersecting a planned cutting surface (21) where the ingot is to be cut, so as to form a plurality of modified portions (31) at the depth position of the planned cutting surface inside the ingot, and by extending cracks (32) starting from the modified portions, the wafer is cut out from the ingot; the energy density, which is the energy per unit area of the laser beam inside the ingot when irradiating the ingot with the laser beam, exceeds a modification threshold value (Eth) capable of modifying a part of the ingot at the planned cutting surface; with respect to the energy density, further satisfying all of the following Condition 1, Condition 2, and Condition 3: Condition 1 is that the peak value (Ep) of the energy density is 44 J / cm 2 or less. Condition 2 is that the energy density increase rate (α), which is the increase width of the energy density per unit depth at the shallowest depth position where the energy density reaches the reforming threshold value, is 1000 J / cm 3 or more. Condition 3 is that the depth range (W) in which the energy density exceeds the modification threshold value is 30 μm or less. A method for manufacturing a wafer.

2. The method for manufacturing a wafer according to Claim 1, wherein the laser beam irradiated onto the ingot is a pulsed laser beam having a pulse width of 250 fs to 10 ns.

3. The laser beam is simultaneously irradiated onto the ingot from a plurality of directions, and the laser beams from the plurality of directions overlap with each other in a part of the depth region including the depth position of the planned cutting surface, and the overlapping portion (Lc) has a length (h) in the depth direction of 5 to 50 μm. The method for manufacturing a wafer according to Claim 1 or 2.

4. The method for manufacturing a wafer according to Claim 1 or 2, wherein a plurality of laser beams having non-parallel optical axes to each other are simultaneously irradiated onto the ingot so as to overlap on the planned cutting surface.

5. The ingot has a substantially cylindrical shape, and a large number of the modified portions are formed along each of a plurality of virtual straight lines (VL) parallel to each other and orthogonal to the axial direction of the ingot at the planned cutting surface; the intensity distribution of the laser beam for forming each of the modified portions as viewed from the axial direction spreads in a direction orthogonal to the virtual straight line rather than in the direction along the virtual straight line. The method for manufacturing a wafer according to any one of Claims 1 to 4.

6. The ingot has a substantially cylindrical shape, and a large number of the modified portions are formed along each of a plurality of virtual straight lines (VL) parallel to each other and orthogonal to the axial direction of the ingot at the planned cutting surface; The pitch of the irradiation points of the adjacent laser beams along the virtual straight line is made shorter at the edge of the ingot than at other parts, according to the method for manufacturing a wafer according to any one of claims 1 to 5.

7. The ingot has a substantially cylindrical shape, and a plurality of the modified parts are formed along each of a plurality of virtual straight lines (VL) that are parallel to each other and orthogonal to the axial direction of the ingot on the planned cutting surface. The energy of the laser beam irradiated on the edge of the ingot is made larger than the energy of the laser beam irradiated on other parts, according to the method for manufacturing a wafer according to any one of claims 1 to 6.

8. The ingot has a substantially cylindrical shape, and a plurality of the modified parts are formed along each of a plurality of virtual straight lines (VL) that are parallel to each other and orthogonal to the axial direction of the ingot on the planned cutting surface. The virtual straight line at a position far from the central axis of the ingot has a narrower arrangement interval than the virtual straight line at a position close to the central axis, according to the method for manufacturing a wafer according to any one of claims 1 to 7.

9. A plurality of the modified parts are formed along each of a plurality of virtual straight lines (VL) that are parallel to each other on the planned cutting surface, and the virtual straight line is a straight line parallel to the formation direction of the off-angle (θ) in the ingot as viewed from the axial direction of the ingot, according to the method for manufacturing a wafer according to any one of claims 1 to 8.

Images