β-Ga2O3 SINGLE CRYSTAL SUBSTRATE, β-Ga2O3 SINGLE-CRYSTAL INGOT, MELT SUPPLY MEMBER, β-Ga2O3 SINGLE CRYSTAL SUBSTRATE PRODUCTION METHOD, AND β-Ga2O3 SINGLE-CRYSTAL INGOT PRODUCTION METHOD

Abstract

Provided is a melt supply member for producing a β-Ga2O3 single crystal using the edge-defined film-fed growth (EFG) method. The melt supply member comprises: an upper surface that has a circular or polygonal shape; and a through hole that reaches the upper surface, that has an opening having a planar shape which does not extend in one direction, and that is for supplying melt to the upper surface.

Description

β-Ga2O3 single crystal substrate, β-Ga2O3 single crystal ingot, melt supply member, method for manufacturing β-Ga2O3 single crystal substrate, and method for manufacturing β-Ga2O3 single crystal ingot

[0001] β-Ga ２ O ３ Single crystal substrate, β-Ga ２ O ３ Single crystal ingot, melt supply member, β-Ga ２ O ３ Method for manufacturing single crystal substrate and β-Ga ２ O ３ The present invention relates to a method for producing a single crystal ingot.

[0002] β-Ga ２ O ３ As methods for producing single crystals, the Edge-defined Film-fed Growth (EFG) method, the Vertical Bridgman (VB) method, the Czochralski (CZ) method, and the like have been proposed.

[0003] In the EFG method, a die is typically placed in a crucible, and the melt in the crucible is supplied to the top surface of the die, where it is brought into contact with a seed crystal to grow the crystal. A typical die has linear slits on the top, bottom, side, etc. When the linear slit comes into contact with the melt in the crucible, capillary action occurs, and the melt is supplied along the linear slit to the top surface of the die. In the EFG method, it is possible to pull crystals into shapes such as plates or rods, depending on the shape of the die and linear slit.

[0004] A related technique is disclosed in Patent Document 1. Patent Document 1 discloses a method for producing β-Ga by the EFG method. ２ O ３ This document discloses a method for producing a single crystal, in which a die having one or more linear slits is used to rotate and pull up a seed crystal, thereby constantly uniformizing the relative positional relationship between the linear slits and the crystal, thereby enlarging the diameter while maintaining the cross section of the crystal in a cylindrical shape.

[0005] JP 2018-076193 A

[0006] As a result of investigations by the inventors, when a die having a linear slit on its upper surface is used, grain boundaries may occur at the position of the linear slit. An example of an object of the present disclosure is to provide a technology for suppressing the occurrence of grain boundaries.

[0007] According to the present disclosure, β-Ga ２ O ３ There is provided a melt supplying member for producing a single crystal, the melt supplying member having a circular or polygonal upper surface, and a through hole that reaches the upper surface, the opening of which does not extend in one direction in plan view, and through which melt is supplied to the upper surface.

[0008] According to the present disclosure, there is provided a β-Ga alloy crystal growth method comprising: a first step of supplying a melt to an upper surface of a melt supply member having a circular or polygonal upper surface and a through-hole that reaches the upper surface, the opening of which does not extend in one direction in plan view, and supplies the melt to the upper surface; a second step of bringing a seed crystal into contact with the melt supplied to the upper surface, and forming a meniscus between the melt and the seed crystal that are in contact with each other; and a third step of growing a crystal while pulling up the seed crystal. ２ O ３ A method for producing a single crystal ingot is provided.

[0009] Further, according to the present disclosure, the β-Ga ２ O ３ β-Ga produced by the single crystal ingot manufacturing method ２ O ３ β-Ga substrates cut from single crystal ingots ２ O ３ A method for manufacturing a single crystal substrate is provided.

[0010] According to the present disclosure, after immersion in phosphoric acid at 140°C for 80 minutes, the average defect density, which is the average of defect densities measured in nine observation areas of a square with a side of 200 μm and a center at each of nine points including each vertex, center of each side, and intersection of the diagonal of a square with a side of 4 mm, is 1.00 × 10 ３ pieces / cm ２ β-Ga ２ O ３ A single crystal substrate is provided.

[0011] According to the present disclosure, the cut substrate is immersed in phosphoric acid at 140° C. for 80 minutes, and then the average defect density, which is the average of the defect densities measured in nine observation areas of a square with a side of 200 μm and a center, including each of the nine points including the vertices, the center of each side, and the intersection of the diagonal lines of a square with a side of 4 mm, is 1.00×10 ３ pieces / cm ２ β-Ga ２ O ３ A single crystal ingot is provided.

[0012] According to one aspect of the present disclosure, a technique for suppressing the occurrence of grain boundaries is provided.

[0013] FIG. 1 is a diagram for explaining a method for measuring the average defect density. FIG. 2 is a perspective view of a melt supply member of a first example. FIG. 3 is a perspective view of the melt supply member of the first example, observed from another angle. FIG. 4 is a perspective view of a melt supply member of a second example. FIG. 5 is a perspective view of a melt supply member of a third example. FIG. 6 is a perspective view of another example of the melt supply member of the fourth example. FIG. 7 is a perspective view of a melt supply member of a fifth example. FIG. 8 is a perspective view of the melt supply member of the fifth example, observed from another angle. FIG. 9 is a cross-sectional schematic view of an example of a manufacturing apparatus. FIG. 10 is a cross-sectional schematic view of another example of the manufacturing apparatus. FIG. 11 is a flowchart showing an example of a manufacturing method. FIG. 12 is a diagram showing the results of Example 1. FIG. 13 is another diagram showing the results of Example 1. FIG. 14 is another diagram showing the results of Example 1. FIG. 15 is a diagram for explaining the effects. FIG. 16 is another diagram for explaining the effects. FIG. 17 is a diagram showing the results of Comparative Example 1.

[0014] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In this disclosure, the drawings relate to one or more embodiments. In all drawings, similar components are designated by similar reference numerals, and descriptions thereof will be omitted as appropriate.

[0015] <<β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ Manufacturing method of single crystal ingot >> β-Ga of this embodiment ２ O ３Single crystal substrate and β-Ga ２ O ３ In the manufacturing method of single crystal ingots, a characteristic melt supplying member (die) is used to produce β-Ga ２ O ３ By growing a single crystal, the occurrence of grain boundaries is suppressed. ２ O ３ Single crystal substrate and β-Ga ２ O ３ According to the method for producing a single crystal ingot, β-Ga having the excellent characteristics described below can be obtained. ２ O ３ Single crystal substrate and β-Ga ２ O ３ Single crystal ingots can be produced.

[0016] <Melt Supply Member> First, the configuration of the melt supply member will be described. ２ O ３ A melt supply member for producing a single crystal.

[0017] The melt supply member has a circular or polygonal upper surface. In this specification, the term "circular" is a concept that encompasses a perfect circle, an ellipse, and a substantially circular shape.

[0018] The "top surface" of the melt supply member is the surface that comes into contact with the seed crystal when growing a crystal. The melt supply member has a bottom surface that is the reverse of the top surface. The "bottom surface" of the melt supply member is the surface that comes into contact with the melt in the crucible when growing a crystal.

[0019] When the upper surface of the melt supply member is polygonal, the upper surface may be quadrilateral or pentagonal. However, it is preferable to adopt a shape with hexagons or more. For example, the upper surface of the melt supply member may be heptagonal, octagonal, or a polygon with more than two corners. By adopting a shape with hexagons or more, each interior angle can be made relatively wide. When the upper surface of the melt supply member is polygonal, it is preferable to adopt a polygon with all interior angles of 100 degrees or more.

[0020] When the melt supply member is observed from above, the aspect ratio (maximum diameter / minimum diameter) is preferably 5 or less, 2 or less, 1.5 or less, or 1.1 or less. By doing so, columnar (cylindrical) or pyramidal (bell-shaped) crystals, preferably crystals close to cylinders, can be produced.

[0021] The melt supply member has a through hole that reaches the upper surface, and supplies the melt to the upper surface through this through hole.

[0022] The planar shape of the opening of the through hole in the upper surface of the melt supply member does not extend in one direction like a linear slit. Such a through hole can also be described as being dot-shaped on the upper surface of the melt supply member. Furthermore, the opening of the through hole that appears in the upper surface of the melt supply member can be configured not to reach the outer periphery of the upper surface of the melt supply member.

[0023] The "opening planar shape" refers to the planar shape of the opening of the through-hole that appears on the upper surface of the melt supply member. The opening planar shape can also be said to be the shape of the through-hole observed when the upper surface of the melt supply member is viewed from above.

[0024] The "opening planar shape not extending in one direction" is, for example, an opening planar shape in which the maximum diameter is 5 times or less, preferably 2 times or less, and more preferably 1.5 times or less, of the minimum diameter.

[0025] The "maximum diameter" here is the length of the longest line among the lines passing through two points on the periphery of the opening planar shape and the center of the opening planar shape, and the "minimum diameter" is the length of the shortest line among the lines passing through two points on the periphery of the opening planar shape and the center of the opening planar shape.

[0026] The planar opening shape of the through hole can be any shape that does not extend in one direction and that allows the melt to be supplied to the upper surface by utilizing capillary action. Examples of the planar opening shape of the through hole include, but are not limited to, a relatively small circle or polygon. For example, if the planar opening shape of the through hole is circular, its diameter may be 0.1 mm to 3.0 mm, preferably 0.5 mm to 2.0 mm, and more preferably 0.8 mm to 1.5 mm. Furthermore, if the planar opening shape of the through hole is polygonal, the diameter of the circumscribed circle of the polygon may be 0.1 mm to 3.0 mm, preferably 0.5 mm to 2.0 mm, and more preferably 0.8 mm to 1.5 mm. When the planar opening shape of the through hole is polygonal, it may be a square, an elongated rectangle, or any other polygon. However, to transport the melt uniformly, the planar opening shape of the through hole is preferably circular.

[0027] By using a melt supply member in which the planar shape of the opening of the through hole does not extend in one direction like a linear slit, the occurrence of grain boundaries and defects can be suppressed, for reasons that will be described later.

[0028] Specific examples of the melt supply member as described above will be described below with reference to the drawings, but the configuration of the melt supply member of this embodiment is not limited to the examples shown here.

[0029] 2 and 3 show a first example of the melt supply member 10. The melt supply member 10 shown in Fig. 2 has a circular upper surface 11. However, as described above, the upper surface 11 may also be polygonal.

[0030] As shown in FIG. 2, a first through-hole 13 and a second through-hole 14 are formed in the upper surface 11 of the melt supply member 10 .

[0031] The opening planar shapes of the first through hole 13 and the second through hole 14 do not extend in one direction. In the illustrated example, the opening planar shapes of the first through hole 13 and the second through hole 14 are circles with a relatively small diameter. The first through hole 13 and the second through hole 14 extend from the top surface 11 toward the bottom surface 17. However, the first through hole 13 and the second through hole 14 do not reach the bottom surface 17.

[0032] 3, two intersecting linear slits 16 are provided in the bottom surface 17 of the melt supplying member 10. The slits 16 extend to the outer periphery 15. The slits 16 extend from the bottom surface 17 toward the top surface 11. However, the slits 16 do not reach the top surface 11 of the melt supplying member 10.

[0033] The first through hole 13 and the second through hole 14 are connected to the slit 16. Therefore, when the melt supplying member 10 is placed in a crucible filled with melt, the melt is supplied along the slit 16 toward the upper surface 11 by capillary action, and then the melt is supplied to the upper surface 11 along the first through hole 13 and the second through hole 14 connected to the slit 16.

[0034] 2, a recessed portion 12 is provided on the upper surface 11. The recessed portion 12 is located at the center or approximately the center of the upper surface 11. A first through hole 13 is present in this recessed portion 12. The second through hole 14 is located in a portion other than the recessed portion 12. That is, in this example, the first through hole 13 is located in a recessed portion (recessed portion 12) of the upper surface 11. The second through hole 14 is located in a non-recessed portion of the upper surface 11. The first through hole 13 and the second through hole 14 differ in this respect.

[0035] As will be described later, the recessed portion 12 has the function of forcibly correcting the outer shape of the crystal during crystal growth in the early stage of crystal growth so that the outer shape of the crystal conforms to the outer peripheral shape of the recessed portion 12 .

[0036] The area of ​​the depression 12 is preferably larger than the bottom area of ​​the seed crystal used in producing the crystal. When the area of ​​the depression 12 is larger than the bottom area of ​​the seed crystal, the crystal shape correction effect described above using the depression 12 is more easily achieved.

[0037] The planar shape of the recess 12 is a circle or a polygon having hexagons or more sides. The planar shape of the recess 12 is the shape of the recess 12 observed when the upper surface 11 of the melt supply member 10 is viewed from above.

[0038] In order to smooth the angles between the facets that appear in the crystal when the diameter is enlarged, it is desirable that the planar shape of the recess 12 be a perfect circle or a circle close to a perfect circle. In this way, the planar shape of the crystal in the initial stage of crystal growth (β-Ga ２ O ３ The shape of the single crystal when observed from the drawing direction (pulling direction) is corrected to a perfect circle or a circle close to it, and the planar shape (β-Ga ２ O ３ The single crystal has a shape in the drawing direction (when viewed from the pulling direction) that is a circle or a circle close to a circle that is enlarged in a substantially similar manner on the upper surface 11 of the melt supply member 10 .

[0039] The above-described effect can be achieved to a certain extent even if the planar shape of the recessed portion 12 is a shape other than a perfect circle or a nearly perfect circle, as long as it is a circle or a polygon having hexagons or more.

[0040] For example, β-Ga ２ O ３ Since the growth anisotropy is strong, the planar shape of the depression 12 can be made elliptical depending on the shape of the crystal to be grown.

[0041] The planar shape of the recess 12 may also be a polygon inscribed in a perfect circle. To smooth the ridges between facets that appear when the diameter is increased (i.e., to increase the angle between the facets) and effectively reduce the occurrence of grain boundaries originating from the ridges between facets, it is preferable to make the interior angle larger than that of a regular square. A polygon with an interior angle of at least 100 degrees or more is expected to suppress the occurrence of grain boundaries originating from the ridges between facets. For example, the planar shape of the recess 12 may be a regular hexagon (with an interior angle of 120 degrees), a regular octagon (with an interior angle of 135 degrees), a regular dodecagon (with an interior angle of 150 degrees), or the like.

[0042] The depth of the recess 12 is preferably 0.3 mm or more and 5.0 mm or less. If the depth of the recess 12 is 0.3 mm or more, the disadvantage of the crystal quickly growing beyond the depth of the recess 12 can be suppressed. As a result, the crystal shape correction effect described above using the recess 12 can be fully exerted. Furthermore, if the depth of the recess 12 is 5.0 mm or less, the disadvantage of a temperature distribution occurring between the upper surface 11 of the melt supply member 10 and the interior of the recess 12, which is unsuitable for crystal growth, can be suppressed. As a result, grain boundaries caused by an inappropriate temperature distribution can be suppressed.

[0043] 2, one first through hole 13 is provided in the recess 12, but two or more first through holes 13 may be provided in the recess 12. In this case, it is preferable to provide two or more first through holes 13 symmetrically with respect to the center of the recess 12. The first through hole 13 may be provided in the center of the recess 12, and two or more further first through holes 13 may be provided symmetrically with respect to the center of the recess 12.

[0044] 2, four second through holes 14 are provided, but the number of second through holes 14 is not limited to this. The number of second through holes 14 may be three or less, or five or more. When providing multiple second through holes 14, it is preferable to provide the multiple second through holes 14 symmetrically with respect to the center of the recessed portion 12.

[0045] If the second through-holes 14 are not present and the melt is supplied to the upper surface 11 only through the first through-holes 13, it may become difficult to supply the melt toward the edge of the crystal if the diameter of the crystal increases (for example, if it increases to exceed the diameter of the recessed portion 12, or if it increases to 3 inches or 4 inches). If the second through-holes 14 are present, the melt can be easily supplied to the entire surface of the crystal, even if the diameter of the crystal increases to be larger than the diameter of the recessed portion 12.

[0046] 3, two slits 16 are provided, but the number of slits 16 may be one, or three or more. Also, in the example of FIG. 3, two slits 16 intersect, but the layout of the multiple slits 16 may be other layouts.

[0047] Second Example Fig. 4 shows a second example of melt supplying member 10. Melt supplying member 10 of the second example differs from melt supplying member 10 of the first example in that it does not have second through-hole 14. The other configurations of melt supplying member 10 of the second example are similar to those of melt supplying member 10 of the first example.

[0048] Third Example Fig. 5 shows a third example of melt supply member 10. Melt supply member 10 of the third example differs from melt supply member 10 of the first example in that it does not have recessed portion 12. In this example, there is no recessed portion in upper surface 11, and first through-hole 13 is located in a non-recessed portion of upper surface 11. The other configuration of melt supply member 10 of the third example is similar to that of melt supply member 10 of the first example.

[0049] Fourth Example Fig. 6 shows a fourth example of melt supply member 10. Melt supply member 10 of the fourth example differs from melt supply member 10 of the first example in that it does not have second through-hole 14 or recessed portion 12. In this example, there is no recessed portion in upper surface 11, and first through-hole 13 is located in a non-recessed portion of upper surface 11. The other configuration of melt supply member 10 of the fourth example is similar to that of melt supply member 10 of the first example.

[0050] 7 and 8 show a fifth example of melt supplying member 10. Melt supplying member 10 of the fifth example differs from melt supplying member 10 of the first example in that melt supplying member 10 of the fifth example does not have slit 16 and first through hole 13 and second through hole 14 penetrate from top surface 11 to bottom surface 17. The other configuration of melt supplying member 10 of the fifth example is similar to that of melt supplying member 10 of the first example.

[0051] The melt supply member 10 of the fifth example may be configured such that the melt supply member 10 of the third example does not have the slit 16, and the first through hole 13 and the second through hole 14 penetrate from the top surface 11 to the bottom surface 17.

[0052] Furthermore, the melt supply member 10 of the fifth example can be configured such that, in the melt supply member 10 of the second or fourth example, the slit 16 is not provided, and the first through hole 13 penetrates from the top surface 11 to the bottom surface 17.

[0053] <Production Apparatus> Next, the β-Ga２ O ３ The configuration of the manufacturing apparatus used in the method for manufacturing a single crystal will be described below. The manufacturing apparatus of this embodiment can have a widely known configuration except for the melt supply member 10.

[0054] 9 and 10 show the β-Ga ２ O ３ 9 and 10 show an example of a cross-sectional view of a manufacturing apparatus used in the method for manufacturing a single crystal, which includes a melt supplying member 10, a seed shaft 20, a seed crystal 30, a crucible 40, an insulating material 60, an insulating material 70, a support member 80, and a high-frequency heating coil 90.

[0055] The melt supplying member 10 in FIGS. 9 and 10 is the melt supplying member 10 of the fifth example described above, but the same device configuration can also be achieved when using the melt supplying member 10 of other examples.

[0056] The heat insulating material 70 is a container. The container is made of, for example, quartz. ２ O ３ and / or ZrO ２ The container (insulating material 70) is made of ceramics containing fluorine. A high frequency heating coil 90 is provided around the outer periphery of the container (insulating material 70).

[0057] An insulating container (insulating material 60) is located inside the container (insulating material 70). Inside the insulating container (insulating material 60), a crucible 40 is located, supported by a support member 80. The support member 80 may be configured to be rotatable. The crucible 40 is made of, for example, iridium. Ga is placed inside the crucible 40. ２ O ３ The melt 50 is stored in the crucible 40. A melt supplying member 10 is positioned in the crucible 40. The melt supplying member 10 is disposed so that the bottom surface 17 thereof comes into contact with the melt 50.

[0058] The seed shaft 20 is made of β-Ga ２ O ３ The seed shaft 20 holds the seed crystal 30. The seed shaft 20 is configured to be rotatable and movable up and down.

[0059] <Production Method> Next, the β-Ga ２ O ３ Single crystal substrate and β-Ga２ O ３ A method for producing a single crystal ingot will be described. ２ O ３ Single crystal substrate and β-Ga ２ O ３ The method for producing a single crystal ingot can be realized using widely known techniques, except for the use of the characteristic melt supply member 10 described above (such as the production conditions). In an example in which a melt supply member 10 having a recess 12 is used, some ingenuity is required to obtain the above-mentioned crystal shape correction effect. This will be described later.

[0060] First, referring to FIGS. 9 to 11, the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ The overall method for manufacturing a single crystal ingot will now be explained.

[0061] As shown in the flowchart of FIG. 11, the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ The method for producing a single crystal ingot includes at least first to third steps. ２ O ３ Single crystal substrate and β-Ga ２ O ３ The method for producing a single crystal ingot may include other steps.

[0062] In the first step, as shown in Fig. 9 or 10 , for example, melt supplying member 10 is placed in crucible 40 containing melt 50, and melt 50 is supplied to upper surface 11 of melt supplying member 10. When bottom surface 17 of melt supplying member 10 is brought into contact with melt 50, capillary action causes melt 50 to be supplied to upper surface 11 along first through hole 13. Note that when the melt supplying member 10 of the other examples described above is used, melt 50 is supplied to upper surface 11 through at least one of second through hole 14 and slit 16 in addition to first through hole 13.

[0063] In the second step, the seed crystal 30 is brought into contact with the melt 50 supplied to the upper surface 11 (seed touch), and a meniscus is formed between the melt 50 and the seed crystal 30 which are in contact with each other.

[0064] In the third step, the seed crystal 30 is pulled up to grow the crystal.

[0065] When using the melt supply member 10 having the recessed portion 12 as in the example of FIG. 9, the process is as follows.

[0066] In a first step S10, melt 50 is supplied to recess 12 on upper surface 11. In a second step S11, seed crystal 30 is brought into contact with melt 50 supplied to recess 12 on upper surface 11 (seed touch) while melt 50 supplied to recess 12 remains inside recess 12 and does not leak out of recess 12.

[0067] The third step includes a step 3-1 and a step 3-2.

[0068] In step 3-1, the seed crystal 30 that has come into contact with the melt 50 supplied to the recess 12 is pulled up while the melt 50 supplied to the recess 12 remains inside the recess 12 and does not leak out to the outside of the recess 12, and the crystal is grown until the diameter of the crystal reaches the diameter of the recess 12.

[0069] In this step 3-1, first, melt 50 supplied to depression 12 is brought into contact with seed crystal 30, and a meniscus is formed between melt 50 and seed crystal 30 in depression 12. Then, while pulling up seed crystal 30, the crystal is gradually grown until its diameter becomes approximately the same as the diameter of the opening of depression 12. After that, a meniscus is formed between seed crystal 30 and the inner surface of depression 12, and the crystal grows so as to wet the inner surface of depression 12.

[0070] In step 3-2, after step 3-1, the seed crystal 30 (grown crystal) that has been in contact with the melt 50 supplied to the outside of the recess 12 is pulled up to grow the crystal.

[0071] In step 3-2, which follows step 3-1 described above, the seed crystal 30 (grown crystal) is further grown while being pulled up, gradually increasing the diameter of the crystal. Then, a meniscus is formed between the seed crystal 30 (grown crystal) and the melt 50 supplied to the outside of the recess 12 via the second through-hole 14, and the diameter of the crystal is increased on the upper surface 11 of the melt supply member 10 (second growth step). Note that depending on the diameter of the crystal to be grown, the supply of melt 50 from only the first through-hole 13 may be sufficient.

[0072] β-Ga of this embodiment ２ O ３ Single crystal substrate and β-Ga ２ O ３ In the method for producing a single crystal ingot, the growth step 3-1 is particularly important. In the growth step 3-1, the meniscus formed by the melt 50 at the bottom of the recess 12 and the surface of the seed crystal 30 in the seed touch step moves from the bottom of the recess 12 to the inner surface of the recess 12 as the crystal grows inside the recess 12, wetting the recess 12 as the crystal grows. That is, the growth step 3-1 has the effect of forcibly correcting the shape of the crystal (the shape of the cross section perpendicular to the stretching direction (pulling direction)) to conform to the recess 12. As a result, in the growth step 3-2, the diameter is expanded so that the shape corrected in the growth step 3-1 (the shape of the cross section perpendicular to the stretching direction (pulling direction)) expands similarly on the upper surface 11 of the melt supply member 10.

[0073] For example, if the planar shape of the recess 12 (the shape of the recess 12 observed when the upper surface 11 of the melt supply member 10 is viewed in plan) is circular, the initial shape of the crystal growth is forcibly corrected to a circular shape in the growth step 3-1. Then, in the growth step 3-2, the diameter is further enlarged so that the corrected circular shape expands similarly. As a result, the diameter is enlarged while the shape of the grown crystal (the shape of the cross section perpendicular to the stretching direction (pulling direction)) is maintained as circular as possible. This suppresses the appearance of strong facets and sharp ridges between facets during the diameter enlargement. Ideally, the grown crystal will be close to a cone with a circular base. In this case, the side surface of the cone is smooth, and the appearance of sharp ridges between facets is suppressed. This ultimately suppresses the occurrence of grain boundaries due to ridges.

[0074] By the above-mentioned first to third steps, β-Ga described below can be obtained. ２ O ３ A single crystal ingot is produced. ２ O ３ By cutting a substrate at a desired angle from a single crystal ingot, a β-Ga ２ O ３ A single crystal substrate is fabricated.

[0075] β-Ga ２ O ３ β-Ga from single crystal ingot ２ O ３ The method of cutting out the single crystal substrate is realized by using a widely known technique. For example, a cutting machine having a wheel saw or a fixed abrasive or loose abrasive wire saw is used to cut out the β-Ga substrate. ２ O ３ This may be achieved by slicing a single crystal ingot.

[0076] In addition, the extracted β-Ga ２ O ３The single crystal substrate may be subjected to a polishing process such as mechanical polishing or chemical polishing. In the chemical polishing, it is preferable to remove the process-affected layer so that latent scratches are not affected during the phosphoric acid treatment or the measurement of the average defect density. Furthermore, it is preferable to polish both sides of the substrate to avoid the inconvenience of scratches on the back surface of the substrate being affected during processing during the same measurement using an optical microscope.

[0077] <<β-Ga ２ O ３ Single crystal substrate >> As a result of the inventors' investigation, when using a die with a linear slit on the top surface, grain boundaries sometimes occur at the position of the linear slit. In contrast, as described above, the β-Ga ２ O ３ In the method for manufacturing a single crystal substrate, a die having a linear slit on the upper surface is not used. ２ O ３ In the single crystal substrate, grain boundaries that may occur at the positions of the linear slits are suppressed.

[0078] Furthermore, by using the melt supply member 10 having the recessed portion 12, the β-Ga melt of this embodiment having the following features can be obtained. ２ O ３ A single crystal substrate is obtained.

[0079] β-Ga of this embodiment ２ O ３ The single crystal substrate has an average defect density of 1.00 × 10 ３ pieces / cm ２ less than 9.0 x 10 ２ pieces / cm ２ less than 8.5 × 10 ２ pieces / cm ２ It has the characteristic of being less than

[0080] The "average defect density" is defined as calculated by the following method.

[0081] First, β-Ga ２ O ３ The single crystal substrate is immersed in phosphoric acid at 140°C for 80 minutes (phosphoric acid etching). This exposes defects on the surface so that they can be observed under an optical microscope.２ O ３ Nine observation points (points A to I) are determined on the surface of the single crystal substrate as shown in Figure 1. Specifically, the nine observation points are determined, including the vertices (points A, C, G, I) of a square with sides of 4 mm, the centers of the sides (points B, D, F, H), and the intersection of the diagonals (point E).

[0082] Next, an optical microscope equipped with a differential interference filter (Keyence Digital Microscope VHX-8000, high-resolution, high-magnification lens VHX-E500) is used to obtain differential interference optical images of nine observation areas (field of view areas), each of which is a square with a side length of 200 μm and centered on each of the nine observation points.

[0083] Then, the number of defects contained in the differential interference optical image is calculated for each observation area, and the average value is calculated as the average defect density.

[0084] The term "defects" includes both point defects and line defects. That is, when counting the number of defects contained in the differential interference optical image, both point defects and line defects are counted together.

[0085] β-Ga of this embodiment ２ O ３ The single crystal substrate has the average defect density thus defined reduced as described above.

[0086] In addition, the β-Ga ２ O ３ The single crystal substrate has a diameter of 10 mm or more. ２ O ３ Although the single crystal substrate is relatively large, with a diameter of 10 mm or more, the average defect density is reduced as described above. Furthermore, if the planar shape of the substrate is polygonal, the diameter of the circumscribed circle of the polygon may be 10 mm or more. The substrate may have an orientation flat or notch that indicates the orientation, and the diameter is measured without including these.

[0087] In addition, the β-Ga ２ O ３The single crystal substrate is characterized by a rhodium concentration in the substrate of less than 1 ppm by mass, preferably less than 0.8 ppm by mass, and more preferably less than 0.5 ppm by mass, and a platinum concentration in the substrate of less than 0.1 ppm by mass.

[0088] In addition, the β-Ga ２ O ３ The single crystal substrate is characterized in that the iridium concentration in the substrate is 0.1 mass ppm or more and 30 mass ppm or less, or may be 1.0 mass ppm or more and 10 mass ppm or less, or 2.0 mass ppm or more and 5.0 mass ppm or less.

[0089] Such characteristics of the rhodium concentration and the iridium concentration in the substrate are characteristics resulting from the production by the EFG method. ２ O ３ The single crystal substrate is β-Ga manufactured by the EFG method. ２ O ３ Although it is a single crystal substrate, the average defect density is reduced as described above.

[0090] Generally, β-Ga is detected by the EFG method. ２ O ３ When growing single crystals, crucibles made of iridium are used. Therefore, β-Ga is an inevitable impurity. ２ O ３ The concentration of iridium contained in the single crystal substrate is relatively high as described above. ２ O ３ When growing single crystals, crucibles made of rhodium are not used. ２ O ３ The concentration of rhodium contained in the single crystal substrate is relatively low as described above.

[0091] By the way, β-Ga is generally produced by the VB method. ２ O ３ When growing single crystals, a crucible made of an alloy containing rhodium (e.g., an alloy of platinum and rhodium) is used. ２ O ３The rhodium concentration in the single crystal substrate is about 2 to 20 mass ppm. ２ O ３ The iridium concentration in the single crystal substrate is less than 0.1 mass ppm.

[0092] Rhodium causes a change in the optical properties of the crystal (yellowish color). Furthermore, since the degree of high-temperature diffusion of rhodium is greater than that of other noble metal impurities, there is a concern that its inclusion as an impurity in the crystal may adversely affect the long-term reliability of semiconductor devices manufactured using the substrate. The β-Ga SiO 2 of this embodiment, in which the rhodium concentration is reduced, ２ O ３ A single crystal substrate can prevent such problems from occurring. For the same reason, it is desirable to reduce the platinum concentration as much as possible.

[0093] In addition, the β-Ga ２ O ３ The single crystal substrate has a feature that the plane orientation of the main surface is (100), (010), (001), or (-201). ２ O ３ The single crystal substrate has a principal surface with a plane orientation of (100), (010), (001), or (-201), and yet has a reduced average defect density as described above.

[0094] <<β-Ga ２ O ３ Single crystal ingot >> As described above, the β-Ga ２ O ３ In the method for manufacturing a single crystal ingot, a die having a linear slit on the upper surface is not used. ２ O ３ In the single crystal ingot, grain boundaries that may occur at the positions of the linear slits are suppressed.

[0095] By using the melt supply member 10 having the recessed portion 12, the β-Ga melt of this embodiment having the following features can be obtained. ２ O ３ A single crystal ingot is obtained.

[0096] β-Ga of this embodiment ２ O ３ The single crystal ingot is cut out from the β-Ga substrate. ２ O ３ It has the characteristics of a single crystal substrate. ２ O ３ The single crystal ingot is a β-Ga alloy having the above-mentioned excellent characteristics. ２ O ３ A single crystal substrate can be cut out.

[0097] β-Ga of this embodiment ２ O ３ The shape of the single crystal ingot is columnar (cylindrical) or conical (bell-shaped).

[0098] That is, the β-Ga ２ O ３ In the single crystal ingot, among multiple cross sections perpendicular to the elongation direction (also referred to as the long axis direction), there are cross sections in which the maximum diameter is 3 times or less, preferably 2 times or less, more preferably 1.5 times or less, and even more preferably 1.2 times or less of the minimum diameter.

[0099] The "maximum diameter" here is the length of the line that is the longest among the lines passing through two points on the periphery of the cross section and the center of the cross section, and the "minimum diameter" is the length of the shortest among the lines passing through two points on the periphery of the cross section and the center of the cross section.

[0100] In addition, the β-Ga ２ O ３ The single crystal ingot satisfies the above-mentioned relationship and has a portion where the minimum diameter of the cross section perpendicular to the elongation direction (also called the long axis direction) is 3 mm or more, preferably 5 mm or more, and more preferably 8 mm or more.

[0101] The stretching direction is β-Ga ２ O ３ This is the direction in which the crystal is pulled up during the production of a single crystal.

[0102] Such cylindrical or pyramidal β-Ga ２ O ３ In the case of a single crystal ingot, β-Ga is grown not only perpendicular to the elongation direction but also at various other angles. ２O ３ Single crystal substrates can be cut at any angle. ２ O ３ Even when the single crystal substrate is cut out, the surface area of ​​the main surface of the substrate is large enough to manufacture a device. ２ O ３ In the case of a single crystal ingot, by adjusting the cutting angle, it is possible to obtain β-Ga with a (100) orientation on the main surface. ２ O ３ Single crystal substrate, β-Ga with a (010) orientation on the main surface ２ O ３ Single crystal substrate, β-Ga with the main surface oriented in (001) ２ O ３ Single crystal substrate, β-Ga with a (-201) orientation on the main surface ２ O ３ Substrates of various plane orientations, such as single crystal substrates, can be cut out. Note that the plane orientation of the main surface includes variations in orientation (for example, ±1 degree from the just plane) that can inevitably occur during the cutting process.

[0103] <<Examples>> <Example 1> In Example 1, β-Ga was grown by the EFG method using a melt supplying member 10 as shown in FIG. ２ O ３ A single crystal was produced. The planar shape of the recess 12 was a circle with a diameter of 10 mm. The depth of the recess 12 was 1.5 mm. The opening planar shape of the through-hole was a circle, and its diameter was 1 mm.

[0104] The melt supply member 10 and the crucible 40 were made of pure iridium (product specification values ​​for impurity content: rhodium 100 mass ppm or less, platinum 100 mass ppm or less).

[0105] First, β-Ga was placed in the crucible 40. ２ O ３ The raw materials were placed in the crucible 40, and the temperature was raised to a crystal growth temperature range (1750 to 1800° C.). Melt 50, which was stored in the crucible 40 by melting the raw materials, was supplied into the recess 12 through the first through-hole 13 provided in the recess 12.

[0106] Next, β-Ga ２ O ３The seed crystal 30 (oriented so as to have a {100} plane and a {001} plane on the side surface, and having a bottom area of ​​3 mm square) was brought into contact with the melt 50 in the depression 12 (seed touch), and a meniscus was formed between the melt 50 present on the bottom surface of the depression 12 and the seed crystal 30 (seed touch process).

[0107] While visually checking the crystal shape through an internal observation camera in the crystal growth furnace, the crystal was grown at an average pulling rate of 1.8 mm / h until the crystal diameter became slightly larger than the diameter of the planar shape of the recess 12 (the diameter of the opening of the recess 12) (i.e., until the diameter became slightly larger than about 10 mm). Crystal growth continued, and visual observation confirmed that the cross-sectional shape of the crystal exceeded the size of the planar shape of the recess 12 and maintained the same nearly circular shape as the planar shape of the recess 12 (growth step 3-1). As the crystal grew, the edge of the meniscus originally formed between the bottom surface of the recess 12 and the seed crystal 30 also moved to the inner surface of the recess 12, forming a meniscus between the inner surface of the recess 12 and the crystal, which is thought to have grown so as to wet the recess 12.

[0108] Next, the seed shaft 20 was raised, and the melt 50 supplied to the depression 12 was supplied to the outside of the depression 12 (the periphery of the depression 12 on the upper surface 11), the meniscus was spread on the upper surface 11, and the crystal was grown while further expanding its diameter at an average pulling rate of 1.5 mm / h (growth step 3-2).

[0109] β-Ga of Example 1 ２ O ３ Single crystal (β-Ga ２ O ３ The β-Ga single crystal ingot was grown by about 11 mm in the pulling direction. ２ O ３ It was confirmed that no grain boundaries originating from the linear slits were generated in the single crystal.

[0110] As shown in Example 1, in the growth step 3-1, the shape of the crystal at the initial stage of growth (the shape of the cross section perpendicular to the stretching direction (pulling direction)) is corrected to a circular shape by the circular recess 12 formed in the melt supply member 10, and then the diameter is enlarged in the growth step 3-2, resulting in a β-Ga crystal whose diameter is enlarged while maintaining the circular shape. ２ O ３ By controlling the crystal shape (cross-sectional shape perpendicular to the drawing direction (pulling direction)), it was possible to obtain a single crystal of β-Ga. ２ O ３ The appearance of facets due to the strong growth anisotropy of β-Ga was reduced. ２ O ３ The size of the base of the single crystal was 14.0 x 10.4 mm.

[0111] FIG. 12 is a schematic diagram showing the external shape of the crystal. As indicated by the arrow in the figure, a characteristic crystal external shape (protrusion) is observed at a position corresponding to the recess 12 of the melt supply member 10. This is thought to support the idea that the meniscus moves to the inner surface of the recess 12 of the melt supply member 10 as the crystal grows, and the crystal grows by wetting the recess 12. By undergoing this growth step 3-1, the shape of the crystal in the initial growth stage (the shape of the cross section perpendicular to the stretching direction (pulling direction)) is corrected to a circular shape, reducing the appearance of strong facets due to strong growth anisotropy and, ultimately, suppressing the occurrence of grain boundaries due to the ridges between facets.

[0112] When the melt supply member 10 without the recessed portion 12 is used, the above-mentioned characteristic crystal shape (protrusion) is not observed. Therefore, it is considered that the β-Ga alloy having such a crystal shape is ２ O ３ It can be assumed that the single crystal was produced using a melt supply member 10 having a recess 12 .

[0113] Example 2 Example 2 used the same melt supply member 10 as Example 1. Also, Example 2 used the same seed crystal 30 as Example 1 (i.e., the orientation, size, etc. were the same).

[0114] β-Ga ２ O３ The process for producing the single crystal was the same as in Example 1, except for the growth rate, which is described below.

[0115] In Example 2, the crystal pulling speed in the growth step 3-1 was "6.2 mm / h on average," and the crystal pulling speed in the growth step 3-2 was "5.5 mm / h on average."

[0116] β-Ga of Example 2 ２ O ３ The single crystal was grown in the pulling direction by about 12 mm. The pulling rate in Example 2 was faster than that in Example 1, but the β-Ga ２ O ３ It was also confirmed that the single crystal did not have grain boundaries due to the appearance of strong facets (sharp ridges). ２ O ３ It was confirmed that no grain boundaries originating from the linear slits were generated in the single crystal.

[0117] Example 3 In Example 3, a melt supply member 10 without a recess 12 as shown in FIG. 5 was used to produce β-Ga by the EFG method. ２ O ３ A single crystal was produced. The other configurations of the melt supply member 10 of Example 3 were the same as those of Example 1. After a seed crystal identical to the seed crystal 30 used in Examples 1 and 2 was brought into contact with the melt, in the initial growth stage, the growth temperature was adjusted within the same temperature range as in Examples 1 and 2, and the crystal was grown in the pulling direction by approximately 33 mm at an average pulling rate of 6.1 mm / h until the crystal had the same diameter and shape as the seed crystal. In the subsequent diameter expansion stage, the growth temperature was adjusted within the same temperature range, and the crystal was grown in the pulling direction by approximately 24 mm at a uniform pulling rate of 5.0 mm / h.

[0118] The β-Ga melt of Example 3 was obtained by using a melt supply member 10 having no linear slits on the upper surface 11. ２ O ３ It was confirmed that no grain boundaries originating from the linear slits were generated in the single crystal.

[0119] Example 4 In Example 4, β-Ga was grown in the same manner as in Example 3, except that in the initial growth stage, the growth was performed at an average pulling rate of 4.4 mm / h, with a growth of approximately 30 mm in the pulling direction, and in the diameter expansion stage, the growth was performed at a uniform pulling rate of 3.0 mm / h, with a growth of approximately 22 mm in the pulling direction. ２ O ３ Single crystal (β-Ga ２ O ３ A single crystal ingot was produced.

[0120] The crystal growth rate of Example 4 is relatively slower than that of Example 3. However, the β-Ga of Example 4 using the melt supply member 10 without the linear slit on the upper surface 11 is ２ O ３ It was confirmed that no grain boundaries originating from the linear slits were generated in the single crystal.

[0121] Example 5 In Example 5, β-Ga was grown in the pulling direction at an average pulling rate of 1.0 mm / h in the initial growth stage (growth step 3-1), and in the diameter expansion stage (growth step 3-2), at an average pulling rate of 2.8 mm / h in the pulling direction, resulting in growth of approximately 30 mm in the pulling direction. ２ O ３ Single crystal (β-Ga ２ O ３ A single crystal ingot was produced.

[0122] β-Ga of Example 5 ２ O ３ It was also confirmed that no grain boundaries were generated due to the appearance of strong facets (sharp ridges) in the single crystal. Furthermore, it was confirmed that no grain boundaries were generated due to linear slits in the β-GaO single crystal of Example 5, which used a melt supply member 10 that did not have linear slits on its upper surface.

[0123] Comparative Example 1 In Comparative Example 1, a die having five parallel, evenly spaced linear slits on the upper surface was used. The slits had a width of 1 mm. After the same seed crystal as used in Examples 1 to 4 was brought into contact with the melt, in the initial growth stage, the crystal was grown in the pulling direction by approximately 45 mm at a uniform pulling rate of 5.0 mm / h while adjusting the growth temperature to the same temperature range as in Examples 1 to 4. The crystal was also grown in the radial direction along the linear slits by a maximum of approximately 49 mm and perpendicular to the linear slits by a maximum of approximately 23 mm.

[0124] FIG. 17 shows the β-Ga obtained in Comparative Example 1. ２ O ３ Single crystal (β-Ga ２ O ３ 1 is a photograph of a single crystal ingot. In the case of Comparative Example 1 having a linear slit on the top surface, multiple grain boundaries were generated near the position of the linear slit. In the case of Comparative Example 1 having a linear slit on the top surface, the linear slits cause the temperature distribution on the top surface to become non-uniform, which is thought to make it easier for grain boundaries to be generated near the position of the linear slit.

[0125] <Considerations on Using the Melt Supply Member 10 Without a Linear Slit on the Upper Surface 11> The β-Ga melts of Examples 1 to 4 using the melt supply member 10 without a linear slit on the upper surface 11 were ２ O ３ It was confirmed that no grain boundaries originating from the linear slits were generated in the single crystal. ２ O ３ In the single crystal, the occurrence of grain boundaries originating from linear slits was confirmed.

[0126] <Considerations on Using Melt Supply Member 10 Having Recess 12> A crystal shape correction effect is achieved when a melt supply member 10 having a recess 12 is used. The effects obtained by using melt supply member 10 having a recess 12 will be explained by comparing Examples 1 and 2 with Examples 3 and 4.

[0127] In both Examples 3 and 4, in which the diameter was enlarged on the melt supply member 10 without the recessed portion 12, the β-Ga ２ O ３Strong facets appeared due to the strong growth anisotropy (i.e., in the comparative example, the overall shape was similar to a square pyramid with a rectangular base), and slight grain boundaries were locally generated starting from the sharp ridges between the facets.

[0128] In contrast, in Examples 1 and 2, which used a melt supply member 10 (die) having a recess 12, the recess 12 was used to forcibly correct the crystal shape (the cross-sectional shape perpendicular to the drawing direction (pulling direction)) into a circular shape during diameter expansion. As a result, the circular shape was relatively maintained during diameter expansion, and the appearance of sharp ridges between facets was suppressed (i.e., in the Examples, the overall shape was close to a cone with a circular base). Furthermore, by making the ridges between facets more blunt, grain boundaries originating from the ridges between facets were suppressed. In other words, when a melt supply member 10 (die) having a recess 12 is used, in addition to the aforementioned "suppression of grain boundaries achieved by using a melt supply member 10 without a linear slit on its upper surface 11," the suppression of grain boundaries due to the above-mentioned effects is further achieved.

[0129] <Evaluation of Examples> "Evaluation of Impurity Concentration" β-Ga in Examples 1 and 5 ２ O ３ Trace amounts of precious metal elements contained in the single crystal were measured by glow discharge mass spectrometry (GD-MS method), and the analytical results shown in Table 1 were obtained (the matrix component Ga ２ O ３ (These are relative mass concentrations when the mass of the crystal is taken as 100%). These noble metal element impurities are thought to originate from the materials used in the crystal growth.

[0130]

[0131] "Evaluation of average defect density" β-Ga in Example 1 ２ O ３ Single crystal (β-Ga ２ O ３The single crystal ingot was sliced ​​with a free abrasive wire saw at a slicing speed of 0.025 mm / min to obtain a substrate with a (100) major surface. After slicing, the deviation (off-angle) of the cut surface from the (100) major surface was measured by XRD. As shown in Figure 13, the off-angle was 2 degrees from the (100) just plane to the [0-5-6] direction.

[0132] Next, the substrate was mechanically polished using a polishing cloth and diamond slurries of various sizes (2-4 μm, 0-2 μm, 0-1 μm) for 3 hours, 1 hour, and 1 hour, respectively.

[0133] Next, the substrate was polished for 10 hours using a polishing cloth and a CMP chemical (colloidal silica slurry, "COMPOL 80" manufactured by Fujimi Incorporated Co., Ltd.).

[0134] Next, the substrate was immersed in phosphoric acid (85% aqueous solution, Hayashi Pure Chemical Industries) at 140° C. for 80 minutes for etching to expose the defects, thereby obtaining a substrate for evaluating the average defect density.

[0135] Next, the average defect density of the obtained substrate was evaluated. First, nine observation points (points A to I) were determined on the surface of the obtained substrate, as shown in Figure 1. Specifically, the nine observation points were determined, including each vertex (points A, C, G, I) of a square with sides of 4 mm, the centers of each side (points B, D, F, H), and the intersection of the diagonals (point E).

[0136] Next, an optical microscope equipped with a differential interference filter (Keyence Digital Microscope VHX-8000, high-resolution, high-magnification lens VHX-E500) was used to obtain differential interference images of nine observation areas (field of view areas) each of which was a square with a side of 200 μm and centered on each of the nine observation points, and the number of defects contained therein was counted. The total number of defects counted in the nine observation areas was then divided by 9 to calculate the number per observation area, and the resulting number was then expressed in units of cm. ２ converted to the number of pieces per

[0137] 14 shows the differential interference contrast images and defect observation results for three of the nine observation areas (observation areas corresponding to points A, B, and D). No defects were observed in the observation area corresponding to point A. On the other hand, defects were observed in the observation areas corresponding to points B and D (points indicated by arrows in the figure).

[0138] Table 2 shows the number of defects measured in each of the nine observation areas and the average defect density calculated based on the results. As shown in Table 2, the average defect density of the substrate in Example 1 was 8.3×10 ２ pieces / cm ２ It was.

[0139]

[0140] In Example 2, the average number of defects measured in each of the nine observation areas was 0.33, the same as in Example 1, and the average defect density was 8.3 × 10 ２ / cm ２ The positions where defects occurred in Example 2 were different from those in Example 1. Furthermore, although Example 1 was a substrate with a (100) plane orientation of the main surface, it is believed that similar results would be obtained with substrates with other plane orientations (for example, substrates with a (010), (001), or (-201) plane orientation of the main surface) that can be cut out from an ingot produced by the same method at different cutting angles.

[0141] "Evaluation of Example 5" The grown crystal of Example 5 was cut with a free abrasive wire saw at a slicing speed of 0.025 mm / min to obtain substrates with (100), (0-10), and (001) major surfaces. After slicing, the deviation (off angle) of the cut surface from each major surface was measured by XRD. The (100) substrate had an off angle of 2.4 degrees from the just plane to the

[017] direction, the (0-10) substrate had an off angle of 1.5 degrees from the just plane to the [-90-20] direction, and the (001) substrate had an off angle of 1 degree from the just plane to the [-5-20] direction.

[0142] Next, each wafer was subjected to mechanical polishing and CMP under the same conditions as those in Example 1 described above in "Evaluation of Average Defect Density."

[0143] Next, the substrates were immersed in phosphoric acid (85% aqueous solution, Hayashi Pure Chemical Industries) at 140°C for etching ((100) plane: 80 min, (0-10) plane: 18 h, (001) plane: 2 h) to expose the etch pits, and each substrate was obtained for dislocation density evaluation. Next, the average dislocation density was evaluated under the same conditions as in Example 1. Table 3 shows the evaluation results and average dislocation density at each point. The average defect density was 5.6 × 10 on the (100) plane. ２ pieces / cm ２ , (0-10) plane, 8.3 × 10 ２ pieces / cm ２ Compared with the previous literature, the dislocation density was reduced to an unprecedented level in each direction. ２ O ３ Single crystals were obtained.

[0144]

[0145] <<Considerations>> <β-Ga of this embodiment ２ O ３ Single crystal substrate and β-Ga ２ O ３ Consideration of the reason why the single crystal ingot can reduce the grain boundary and average defect density compared to the conventional technology> The β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ The following three points are thought to be the reasons why the single crystal ingot has reduced grain boundaries and average defect densities compared to conventional techniques.

[0146] The first reason is that the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ The single crystal ingot is produced using the above-described characteristic die (melt supply member 10) without using a die with a linear slit. This is thought to improve the temperature distribution on the upper surface of the die during production. Figure 15(A) shows the β-Ga single crystal ingot of this embodiment. ２ O ３ Single crystal substrate and β-Ga ２ O ３15(A) and 15(B) are schematic diagrams of the temperature distribution on the top surface of an example die used in the production of a single crystal ingot, and Fig. 15(B) is a schematic diagram of the temperature distribution on the top surface of a die having a linear slit. Fig. 15(A) is a qualitative conceptual diagram, and Fig. 15(B) shows the results of a simulation performed using software.

[0147] As shown in FIG. 15B, when the die has a linear slit, that is, when the die has an opening (linear slit) on the top surface of the die whose planar shape extends in one direction, the temperature distribution may become large at the position across the linear slit (for example, the four circle positions in FIG. 15B). ２ O ３ During the production of single crystals, when the meniscus crosses a linear slit, grain boundaries are generated starting from the linear slit, resulting in an increase in defects. In other words, when a die has a linear slit, the in-plane temperature distribution on the upper surface of the die becomes non-uniform, which can lead to the generation of grain boundaries and defects. In contrast, when the die does not have a linear slit, i.e., when the upper surface of the die does not have an opening whose planar shape extends in one direction, as shown in FIG. 15(A), the above-mentioned problem of in-plane temperature distribution on the upper surface of the die is alleviated. Therefore, it is believed that the generation of grain boundaries and defects is suppressed.

[0148] The second reason is that, in the conventional technology, crystals are grown in a plate shape, but the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ The single crystal ingot grows crystals more isotropically (in a shape closer to a cone or cylinder). ２ O ３ Single crystal substrate and β-Ga ２ O ３ It is believed that single crystal ingots have a more uniform temperature gradient within the grown crystal than conventional techniques, which is believed to suppress the occurrence of grain boundaries and defects.

[0149] The third reason is that the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３The single crystal ingot is grown so that the strong facets (having acute crystal angles) that occur when the diameter of the crystal is enlarged are suppressed and the crystal has a smoother side shape (obtuse corners). ２ O ３ Single crystal substrate and β-Ga ２ O ３ The side shape of a crystal during growth of a single crystal ingot is shown schematically in Figure 16(B), with an example of a crystal with an acute side shape. The example in Figure 16(A) has a smoother side shape (obtuse corners), so the temperature gradient inside the crystal during crystal growth is smaller (i.e., the temperature gradient from the top to the base of the crystal is smaller) compared to the comparative example in Figure 16(B). Therefore, the example in Figure 16(A) is thought to have a more uniform temperature distribution. This is thought to suppress the occurrence of grain boundaries and defects.

[0150] The melt supply member 10 having the recessed portion 12 is used to melt the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ When a single crystal ingot is produced, the above-mentioned crystal shape correction effect is exerted, which is thought to further suppress the occurrence of grain boundaries and defects.

[0151] β-Ga ２ O ３ Single crystals have strong growth anisotropy (the growth rate of the b-axis is particularly fast, followed by the c-axis, and the a-axis is slowest), so the crystal shape (the cross-sectional shape perpendicular to the stretching direction (pulling direction)) tends to be elongated in one direction. This growth anisotropy causes strong facets to appear in the crystal, and there is a risk of grain boundaries arising from sharp ridges between facets (where the angle between the facets is relatively small). Facet growth is particularly likely when the diameter of the crystal is enlarged, and the occurrence of facets with sharp ridges induces the occurrence of grain boundaries.

[0152] When the melt supply member 10 having the recessed portion 12 is used, β-Ga ２ O ３Regardless of the strong growth anisotropy of the single crystal, in the early stages of crystal growth, the external shape of the crystal can be forcibly corrected to fit the outer peripheral shape of the depression 12. By using such a melt supply member 10, the connections between facets can be made smoother (a state in which the angles between the surfaces of the facets are relatively large), and the occurrence of grain boundaries originating from the ridges between the facets can be reduced.

[0153] The melt supply member 10 without the recessed portion 12 was used to manufacture the β-Ga ２ O ３ Single crystal substrate and β-Ga ２ O ３ When a single crystal ingot is produced, the crystal shape correction effect shown as the third reason above is not exerted, but it is thought that the occurrence of grain boundaries and dislocation defects is suppressed due to the first and second reasons above.

[0154] <Considerations on the prior art using dies with linear slits> Generally, the density of defects that occur in a crystal is related to the temperature distribution near the crystal growth portion during crystal growth, and it is thought that the more uniform the temperature distribution in the crystal growth portion and inside the growing crystal, the lower the defect density.

[0155] In the case of conventional techniques for producing cylindrical single crystals using a die with a linear slit, previous studies by the present inventors have shown that the presence of the linear slit causes uneven temperature distribution within the die, resulting in the formation of grain boundaries near the slit position. It is believed that grain boundaries are formed when the atomic arrangement undergoes extreme slippage in order to relieve thermal strain within the crystal. In the case of such conventional techniques, the defect density can be relatively high.

[0156] In the case of conventional techniques for producing plate-shaped single crystals using dies with linear slits, it is believed that the temperature distribution becomes uneven near the crystal growth area of ​​the plate-shaped crystal and inside the crystal, making it easier for grain boundaries and defects to occur.

[0157] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0158] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes: 1. β-Ga using the EFG method ２ O ３ A melt supply member for producing a single crystal, the melt supply member having a circular or polygonal upper surface and a through-hole that reaches the upper surface, the opening of which does not extend in one direction in plan view, and supplies melt to the upper surface. 2. The melt supply member according to 1, wherein the upper surface has a recessed portion, the planar shape of the recessed portion is a circle or a polygon having 6 or more sides, and the through-hole is located in the recessed portion. 3. The melt supply member according to 2, wherein the through-hole is located in a portion of the upper surface other than the recessed portion. 4. The melt supply member according to 2 or 3, wherein the area of ​​the recessed portion is larger than the bottom area of ​​a seed crystal used for crystal growth. 5. The melt supply member according to any one of 2 to 4, wherein the planar shape of the recessed portion is a polygon with an interior angle of 100 degrees or more. 6. The melt supply member according to any one of 2 to 5, wherein the depth of the recessed portion is 0.3 mm or more and 5.0 mm or less. 7. A circular or polygonal upper surface and an opening extending to the upper surface, the opening plane shape of which does not extend in one direction, ２ O ３ a first step of supplying the melt to the upper surface of a melt supply member having a through hole for supplying the melt through the through hole; a second step of bringing a seed crystal into contact with the melt supplied to the upper surface to form a meniscus between the melt and the seed crystal which are in contact with each other; and a third step of growing a crystal while pulling up the seed crystal. ２ O ３8. The method for manufacturing a single crystal ingot according to 7, wherein the first step supplies the melt to the recessed portion on the upper surface of the melt supply member, the recessed portion having a planar shape of a circle or a polygon having hexagons or more sides, and the through hole is provided in the recessed portion. ２ O ３ 9. The method for producing a single crystal ingot according to 7 or 8, wherein the area of ​​the planar shape of the depression is larger than the bottom area of ​​the seed crystal. ２ O ３ 10. A method for producing a single crystal ingot according to any one of 7 to 9, wherein in the second step, the seed crystal is brought into contact with the melt supplied to the depression on the upper surface in a state where the melt supplied to the depression remains inside the depression and does not leak out to the outside of the depression. ２ O ３ 11. A method for manufacturing a single crystal ingot, comprising: a 3-1 step of growing a crystal until a diameter of the crystal reaches a diameter of the depression while pulling up the seed crystal that has been in contact with the melt supplied to the depression in a state where the melt supplied to the depression remains inside the depression and does not leak out of the depression; and a 3-2 step of growing a crystal while pulling up the seed crystal that has been in contact with the melt supplied to the outside of the depression after the 3-1 step. ２ O ３ 12. A method for producing a single crystal ingot. ２ O ３ β-Ga produced by the single crystal ingot manufacturing method ２ O ３ β-Ga substrates cut from single crystal ingots ２ O ３ 13. A method for manufacturing a single crystal substrate. After immersion in phosphoric acid at 140°C for 80 minutes, the average defect density, which is the average of defect densities measured in nine observation areas of a 200 μm square with sides of 4 mm each, centered on nine points including each vertex, the center of each side, and the intersection of the diagonal lines, is 1.00 × 10 ３ pieces / cm ２ β-Ga ２ O ３14. The β-Ga single crystal substrate according to 13, having a diameter of 10 mm or more. ２ O ３ 15. The β-Ga single crystal substrate according to 13 or 14, wherein the rhodium concentration in the substrate is less than 1 ppm by mass. ２ O ３ 16. A β-Ga single crystal substrate according to any one of 13 to 15, wherein the iridium concentration in the substrate is 0.1 mass ppm or more and 30 mass ppm or less. ２ O ３ 17. A β-Ga single crystal substrate according to any one of 13 to 16, wherein the principal surface has a plane orientation of (100), (010), (001), or (-201). ２ O ３ Single crystal substrate. 18. After immersing the cut substrate in phosphoric acid at 140°C for 80 minutes, the average defect density, which is the average of the defect densities measured in nine observation areas of a 200 μm square with sides of 4 mm each, centered on nine points including each vertex, the center of each side, and the intersection of the diagonal lines, is 1.00 × 10 ３ pieces / cm ２ β-Ga ２ O ３ 19. A β-Ga single crystal ingot according to 18, wherein the diameter of the cut substrate is 10 mm or more. ２ O ３ 20. A β-Ga single crystal ingot according to 18 or 19, wherein the rhodium concentration in the sliced ​​substrate is less than 1 ppm by mass. ２ O ３ 21. A β-Ga single crystal ingot according to any one of 18 to 20, wherein the iridium concentration in the sliced ​​substrate is 0.1 mass ppm or more and 30 mass ppm or less. ２ O ３ 22. A β-Ga single crystal ingot according to any one of 18 to 21, wherein the plane orientation of the main surface of the sliced ​​substrate is (100), (010), (001), or (-201). ２ O ３ Single crystal ingot.

[0159] This application claims priority based on Japanese Patent Application No. 2024-103897, filed June 27, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0160] REFERENCE SIGNS LIST 10 melt supply member 11 upper surface 12 recessed portion 13 first through-hole 14 second through-hole 15 outer periphery 16 slit 17 bottom surface 20 seed shaft 30 seed crystal 40 crucible 50 melt 60 heat insulating material 70 heat insulating material 80 support member 90 high frequency heating coil

Claims

1. β-Ga ２ O ３ A melt supply member for producing a single crystal, the melt supply member having a circular or polygonal upper surface, and a through hole that reaches the upper surface, the opening of which does not extend in one direction in plan view, and through which melt is supplied to the upper surface.

2. The melt supply member according to claim 1, wherein the upper surface has a recessed portion, the planar shape of the recessed portion is a circle or a polygon having hexagons or more sides, and the through hole is present in the recessed portion.

3. The melt supply member according to claim 2, wherein the through-holes are further present in a portion of the upper surface other than the recessed portion.

4. The melt supply member according to claim 2, wherein the area of ​​the recess is larger than the bottom area of ​​the seed crystal used for crystal growth.

5. The melt supply member according to claim 2, wherein the planar shape of the recess is a polygon with an interior angle of 100 degrees or more.

6. The melt supply member according to claim 2, wherein the depth of the recess is 0.3 mm or more and 5.0 mm or less.

7. A circular or polygonal upper surface and an opening that reaches the upper surface and has an opening plan shape that does not extend in one direction, and Ga is provided on the upper surface. ２ O ３ a first step of supplying the melt to the upper surface of a melt supply member having a through hole for supplying the melt through the through hole; a second step of bringing a seed crystal into contact with the melt supplied to the upper surface to form a meniscus between the melt and the seed crystal which are in contact with each other; and a third step of growing a crystal while pulling up the seed crystal. ２ O ３ A method for manufacturing a single crystal ingot.

8. The β-Ga melt supplying member according to claim 7, wherein in the first step, the melt is supplied to a recessed portion on the upper surface of the melt supplying member, the recessed portion having a planar shape of a circle or a polygon having hexagons or more sides, and the through-hole is present in the recessed portion. ２ O ３ A method for manufacturing a single crystal ingot.

9. The β-Ga as claimed in claim 7, wherein the area of ​​the planar shape of the recess is larger than the bottom area of ​​the seed crystal. ２ O ３ A method for manufacturing a single crystal ingot.

10. The β-Ga alloy according to claim 7, wherein in the second step, the seed crystal is brought into contact with the melt supplied to the recess on the upper surface in a state in which the melt supplied to the recess remains inside the recess and does not leak out to the outside of the recess. ２ O ３ A method for manufacturing a single crystal ingot.

11. The β-Ga alloy according to claim 7, wherein the third step comprises: a 3-1 step of growing a crystal until the diameter of the crystal reaches the diameter of the recess while pulling up the seed crystal that has been in contact with the melt supplied to the recess in a state where the melt supplied to the recess remains inside the recess and does not leak out of the recess; and a 3-2 step of growing a crystal while pulling up the seed crystal that has been in contact with the melt supplied to the outside of the recess after the 3-1 step. ２ O ３ A method for manufacturing a single crystal ingot.

12. β-Ga according to any one of claims 7 to 11 ２ O ３ β-Ga produced by the single crystal ingot manufacturing method ２ O ３ β-Ga substrates cut from single crystal ingots ２ O ３ A method for manufacturing a single crystal substrate.

13. After immersion in phosphoric acid at 140°C for 80 minutes, the average defect density, which is the average of the defect densities measured in nine observation areas of a 200 μm square with sides each, centered on nine points including each vertex, center of each side, and intersection of the diagonal of a 4 mm square, is 1.00 × 10 ３ pieces / cm ２ β-Ga ２ O ３ Single crystal substrate.

14. The β-Ga according to claim 13, having a diameter of 10 mm or more. ２ O ３ Single crystal substrate.

15. The β-Ga as claimed in claim 13, wherein the rhodium concentration in the substrate is less than 1 ppm by mass. ２ O ３ Single crystal substrate.

16. The β-Ga as claimed in claim 13, wherein the iridium concentration in the substrate is 0.1 mass ppm or more and 30 mass ppm or less. ２ O ３ Single crystal substrate.

17. The β-Ga alloy according to any one of claims 13 to 16, wherein the plane orientation of the primary surface is (100), (010), (001), or (-201). ２ O ３ Single crystal substrate.

18. After immersing the cut substrate in phosphoric acid at 140°C for 80 minutes, the average defect density, which is the average of the defect densities measured in nine observation areas of a 200 μm square with sides each, centered on nine points including each vertex, center of each side, and intersection of the diagonal of a 4 mm square, is 1.00 × 10 ３ pieces / cm ２ β-Ga ２ O ３ Single crystal ingot.

19. The β-Ga alloy according to claim 18, wherein the diameter of the cut substrate is 10 mm or more. ２ O ３ Single crystal ingot.

20. The β-Ga as claimed in claim 18, wherein the rhodium concentration in the cut substrate is less than 1 ppm by mass. ２ O ３ Single crystal ingot.

21. The β-Ga GaN crystal according to claim 18, wherein the iridium concentration in the cut-out substrate is 0.1 mass ppm or more and 30 mass ppm or less. ２ O ３ Single crystal ingot.

22. The β-Ga GaN crystal according to any one of claims 18 to 21, wherein the plane orientation of the main surface of the cut-out substrate is (100), (010), (001), or (-201). ２ O ３ Single crystal ingot.

Images