Method of manufacturing silicon carbide single crystal, regression model generating device, composition estimating device, program, recording medium, regression model generating method, and composition estimating method

Abstract

To provide a method for producing an SiC single crystal in which a used solution used for producing the SiC single crystal in a solution growth method is reused again as a raw material for producing the SiC single crystal.SOLUTION: The method for producing a silicon carbide single crystal includes (a) a step of preparing a solution 20 containing silicon and carbon and (b) a step of bringing a seed crystal 30 of silicon carbide into contact with the solution 20 and growing a silicon carbide single crystal on a crystal growth surface 30a of the seed crystal 30, wherein a solidified product of a solution used in a previously performed method for producing a silicon carbide single crystal is used as a raw material of the solution 20 in the step (a).SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for producing a single crystal of silicon carbide. [Background technology]

[0002] For example, inverter circuits are used as circuits to control motors included in automobiles, home appliances, etc. These inverter circuits use power semiconductor elements such as power MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) and IGBTs (Insulated Gate Bipolar Transistors).

[0003] Such power semiconductor devices are required to have, for example, low on-resistance and low switching loss in addition to high breakdown voltage. Currently, the mainstream of power semiconductor devices is a field effect transistor formed on a semiconductor substrate whose main component is silicon, but these power semiconductor devices are approaching their theoretical performance limits.

[0004] In this regard, semiconductor elements including field effect transistors formed on a semiconductor substrate whose main component is a semiconductor material with a wider band gap than silicon (hereinafter referred to as wide band gap power semiconductor elements) have attracted attention.

[0005] This is because a large band gap means that the material has high dielectric breakdown strength, making it easier to achieve high breakdown voltage.

[0006] Furthermore, if the semiconductor material itself has high dielectric breakdown strength, the withstand voltage can be ensured even if the drift layer that maintains the withstand voltage is made thin. Therefore, for example, by making the drift layer thinner and increasing the impurity concentration, the on-resistance of the power semiconductor element can be reduced.

[0007] That is, wide bandgap power semiconductor devices are superior in that they can achieve both improved breakdown voltage and reduced on-resistance, which are in a trade-off relationship with each other. Therefore, wide bandgap power semiconductor devices are expected to be semiconductor devices that can achieve high performance.

[0008] Examples of semiconductor materials with a band gap larger than that of silicon include silicon carbide (SiC), gallium nitride (GaN), gallium oxide (Ga2O3), diamond, etc. The following description focuses on silicon carbide.

[0009] A single crystal made of silicon carbide (hereinafter also referred to as a SiC single crystal) can be produced by, for example, a sublimation method, a high-temperature gas growth method, a solution growth method, or the like.

[0010] The sublimation method involves evaporating silicon and carbon from raw materials at temperatures above 2000°C, and then condensing SiC single crystals onto a silicon carbide seed crystal, which is at a lower temperature than the raw material. With this method, the growth interface is between the gas and solid phases, and the temperature difference between them is large, so the grown crystal has a high dislocation density, which poses a problem in terms of crystal quality.

[0011] High-temperature gas growth is a method in which gaseous material is introduced at a high temperature of over 2000°C and a chemical reaction is used to grow SiC single crystals on a silicon carbide seed crystal. As with sublimation, this method has a growth interface between the gas and solid phases, and the temperature difference is large, so the grown crystal has a high dislocation density, which poses a problem in terms of crystal quality.

[0012] In contrast to these methods, the solution growth method involves immersing a silicon carbide seed crystal in a solution containing silicon and carbon, creating a temperature gradient within the solution, and creating a carbon supersaturated state near the growth interface, thereby growing SiC single crystals. With this method, the growth interface is between the liquid and solid phases, and the temperature difference between them is smaller than in the above methods, resulting in a state closer to thermal equilibrium. As a result, the grown crystal has a low dislocation density and is of good quality.

[0013] However, the solution growth method has the problem of slower crystal growth rate than the sublimation method. To address this issue, it is considered to increase the carbon solubility, and one known method is to use a transition element such as chromium (Cr) in the solution.

[0014] Furthermore, it is known that adding aluminum (Al), which has a low surface tension, to a Si-Cr solution can suppress roughness of the crystal growth interface and improve crystal quality (see, for example, Non-Patent Document 1).

[0015] Furthermore, since yttrium (Y) is less easily incorporated into SiC single crystals than Cr or Ti, a method is known in which the incorporation of Y into a solution reduces the amount of metal incorporated, thereby enabling a reduction in the concentration of metal impurities in SiC single crystals (see, for example, Patent Document 1).

[0016] Furthermore, a method is known in which a rare earth element having a melting point lower than that of silicon is added to suppress the inclusion of polycrystals and grow a high-quality SiC single crystal (see, for example, Patent Document 2). [Prior art documents] [Patent documents]

[0017] [Patent Document 1] Japanese Patent Application Publication No. 2019-19037 [Patent Document 2] Japanese Patent Application Publication No. 2019-104661 [Patent Document 3] Japanese Patent Application Publication No. 2018-16498 [Non-patent literature]

[0018] [Non-Patent Document 1] T. Mitani, et al.: J. Cryst. Growth, 401 (2014) p681-685. Summary of the Invention [Problem to be solved by the invention]

[0019] In the production of SiC single crystals by the solution growth method described above, a crucible made of graphite (C) or silicon carbide (SiC) is generally used. This crucible serves as a carbon source, and a solution containing predetermined concentrations of carbon and silicon is held in the crucible.

[0020] After sufficient growth of the SiC single crystal, the SiC single crystal manufacturing equipment is cooled from a high-temperature state to room temperature, completing the manufacturing process. The solution solidifies during cooling, but the difference in thermal contraction rate between the solution and the crucible during cooling can cause deformation or collapse of the crucible. Furthermore, the crucible's internal shape changes as carbon (C) dissolves during the manufacturing of SiC single crystals. Silicon (Si) and carbon (C) are consumed in the solution used, and the composition changes during cooling due to the precipitation of miscellaneous crystals. These miscellaneous crystals are 4H-SiC and other polymorphic SiC that occur in addition to the SiC crystals grown from the seed crystals.

[0021] Therefore, the solution and crucible used in the production of SiC single crystals are discarded, and new solution and crucibles are prepared for the next production of SiC single crystals. As mentioned above, elements such as Cr and Al are added in the solution growth method to improve the solubility of carbon, but because a new solution is used for each production, these added elements are discarded along with the used solution.

[0022] However, additive elements such as Cr and Al are components that are not consumed during the production of SiC single crystals. If these additive elements could be reused as raw materials for the production of SiC single crystals, resources could be used more effectively and production costs could be reduced.

[0023] Therefore, an object of the present invention is to provide a method for producing a SiC single crystal in a solution growth method, in which a solution used in producing a SiC single crystal is reused as a raw material for producing a SiC single crystal.

[0024] In addition, in a method for producing a silicon carbide single crystal ingot using a sublimation recrystallization method, a method is known in which an ingot is first produced, and then a recycled raw material powder is obtained by recycling the raw material residue remaining in the crucible (see Patent Document 3). [Means for solving the problem]

[0025] In one embodiment, the method for producing a silicon carbide single crystal comprises the steps of: (a) preparing a solution containing silicon and carbon; and (b) contacting a silicon carbide seed crystal with the solution and growing a silicon carbide single crystal on the crystal growth surface of the seed crystal, wherein a solidified solution used in a previous method for producing a silicon carbide single crystal is used as the raw material for the solution in step (a).

[0026] In one embodiment, the regression model generation device is included in a composition estimation system for estimating the composition of a used solution used in single crystal growth by a solution growth method. The regression model generation device includes a regression model generation unit that generates a regression model that outputs an estimated value for the composition of the used solution when a pre-use composition value indicating the composition of the solution before use and manufacturing conditions for growing the single crystal are input.

[0027] In one embodiment, the program causes a computer to execute a process for estimating the composition of a used solution used in growing a single crystal by a solution growth method, and includes a regression model generation process for generating a regression model that outputs an estimate for the composition of the used solution when a pre-use composition value indicating the composition of the solution before use and manufacturing conditions for growing the single crystal are input.

[0028] The above-mentioned program is recorded on a computer-readable recording medium.

[0029] In one embodiment, the regression model generation method is a method for generating a regression model by a computer that estimates the composition of a used solution used in single crystal growth by a solution growth method, and includes a regression model generation step in which, when a pre-use composition value indicating the composition of the solution before use and manufacturing conditions for growing the single crystal are input, the computer generates the regression model that outputs an estimate for the composition of the used solution.

[0030] In one embodiment, the composition estimation device is included in a composition estimation system that estimates the composition of a used solution used in single crystal growth by a solution growth method. The composition estimation device includes an estimation unit that estimates the composition of the used solution based on a pre-use composition value indicating the composition of the solution before use, manufacturing conditions for growing the single crystal, and a regression model. Here, the regression model is a function that outputs an estimated value for the composition of the used solution when the pre-use composition value and manufacturing conditions are input.

[0031] In one embodiment, the program causes a computer to execute a process for estimating the composition of a used solution used in single crystal growth by a solution growth method. This program includes a process for estimating the composition of the used solution based on a pre-use composition value indicating the composition of the solution before use, manufacturing conditions for growing a single crystal, and a regression model. Here, the regression model is a function that outputs an estimated value for the composition of the used solution when the pre-use composition value and manufacturing conditions are input.

[0032] The above-mentioned program is recorded on a computer-readable recording medium.

[0033] In one embodiment, a composition estimation method is a method for estimating the composition of a used solution used in single crystal growth by a solution growth method using a computer. This composition estimation method includes a step of estimating the composition of the used solution using a pre-use composition value indicating the composition of the solution before use, manufacturing conditions for growing the single crystal, and a regression model. Here, the regression model is a function that outputs an estimated value for the composition of the used solution when the pre-use composition value and manufacturing conditions are input. [Effects of the Invention]

[0034] According to the method for producing a silicon carbide single crystal of one embodiment, the raw materials used for producing a SiC single crystal in a solution growth method are reused, thereby reducing the amount of raw materials used and enabling efficient production of a silicon carbide single crystal. [Brief explanation of the drawings]

[0035] [Figure 1] 1 is a schematic cross-sectional view illustrating an example of the configuration of a manufacturing apparatus used in a method for manufacturing a silicon carbide single crystal according to an embodiment. FIG. [Figure 2] 1 is a schematic cross-sectional view showing the crystal orientation of a silicon carbide seed crystal used in a method for producing a silicon carbide single crystal according to one embodiment. FIG. [Figure 3] FIG. 10 is a diagram illustrating an example of cutting out a solidified product of a used solution to be reused. [Figure 4] FIG. 2 is a diagram illustrating an example of a hardware configuration of a composition estimation device. [Figure 5] FIG. 2 is a functional block diagram showing functions of the composition estimation device. [Figure 6] 10 is a table showing an example of related data. [Figure 7] FIG. 10 is a diagram illustrating the generation of a regression model by machine learning using related data as training data. [Figure 8] 10 is a flowchart illustrating an operation for generating a regression model. [Figure 9] 10 is a flowchart illustrating an operation for estimating the composition of a used solution. DETAILED DESCRIPTION OF THE INVENTION

[0036] In all the drawings for explaining the embodiments, the same components are generally designated by the same reference numerals, and repeated explanations thereof will be omitted. In order to make the drawings easier to understand, hatching may be applied even to plan views, and hatching may be omitted even to cross-sectional views.

[0037] <Consideration of improvements> When growing silicon carbide single crystals (SiC single crystals) using the solution growth method, it is necessary to create a carbon-supersaturated state in the solution used to manufacture the crystal in order to precipitate the SiC single crystals. Therefore, in the solution growth method for growing SiC single crystals, a temperature gradient is created in the solution to create a carbon-supersaturated state.

[0038] In this case, a high-temperature region and a low-temperature region are formed in the solution, and a carbon-supersaturated state is realized in the low-temperature region. Therefore, by forming a temperature gradient in the solution so that the region of the solution in contact with the silicon carbide seed crystal is the low-temperature region where the carbon-supersaturated state is realized, it is possible to grow a crystal on the seed crystal.

[0039] The carbon supersaturated state in the low-temperature region is formed mainly on the surface of a solution containing carbon (C) and silicon (Si), and a seed crystal can be brought into contact with the surface of this solution to grow a SiC single crystal as described above. The SiC single crystal is produced inside an insulating member (insulating container) arranged to surround a crucible containing the solution, and the temperature inside this insulating member is maintained at a temperature suitable for single crystal production.

[0040] As described above, the solution used here is known to contain Cr or Al in order to suppress the growth rate of SiC single crystals and the roughness of the resulting crystal growth interface. However, when these elements are contained, a new solution is used each time production is carried out, and these elements are also discarded along with the used solution.

[0041] Therefore, the present inventors have discovered a method for producing a SiC single crystal that reuses these elements to make effective use of resources, and that allows the production of a new SiC single crystal by adjusting the composition so that the element concentrations are similar even when the elements are reused.

[0042] <Embodiment> An embodiment of the present invention will be described in detail below.

[0043] In describing this embodiment, first, a manufacturing apparatus used in the method for manufacturing a SiC single crystal will be described.

[0044] [Silicon carbide single crystal manufacturing equipment] FIG. 1 is a diagram showing a schematic configuration of silicon carbide single crystal manufacturing apparatus 101 (in use) for explaining the apparatus's configuration used in the method for manufacturing a SiC single crystal according to the present embodiment.

[0045] In FIG. 1, a single crystal manufacturing apparatus 101 is configured to include a crucible 10, a furnace 11, an induction coil 12, a crucible holding shaft 13, and a crystal holding shaft 14.

[0046] The crucible 10 is a container that contains a crystal production solution 20, which is a raw material for growing a SiC single crystal. The crucible 10 is made of, for example, graphite (C) or silicon carbide (SiC), so that when crystal growth is performed, the crucible 10 contains a high-temperature solution 20 containing silicon (Si) and carbon (C).

[0047] Furnace 11 is a component that can accommodate crucible 10 and forms a field for producing a single crystal. Furnace 11 is shaped to surround crucible 10, and is capable of maintaining its interior at a predetermined high temperature (hot zone). Furnace 11 is provided with a crucible-holding shaft 13 (described later) at the bottom, to which crucible 10 can be attached, and a crystal-holding shaft 14 (described above), to the tip of which a silicon carbide seed crystal 30 can be attached.

[0048] An induction coil 12 through which a high-frequency current flows is provided on the outer periphery of this furnace 11, and crucible 10 can be heated by induction heating based on the high-frequency current flowing through induction coil 12. Specifically, induction coil 12 is provided at a position facing the side surface of crucible 10 across furnace 11, and crucible 10 is heated by the induction heating phenomenon caused by passing a high-frequency current through induction coil 12. Although not shown in FIG. 1, induction coil 12 is configured so that cooling water can be flowed inside.

[0049] Although not shown, the single crystal manufacturing apparatus is provided with a housing surrounding the furnace 11 and induction coil 12, and the interior space of the housing can be filled with an inert gas such as argon gas for single crystal manufacturing. The housing is made of an iron-based material such as SUS.

[0050] As a structure for filling argon gas, for example, a quartz tube may be passed between the furnace 11 and the induction coil 12, and the upper and lower ends of the quartz tube may be sealed with flanges to fill the argon gas.

[0051] Crucible holding shaft 13 is a member that fixes and holds crucible 10 at its upper end, and is configured to be movable in the vertical direction. Furthermore, it may be configured to be rotatable either clockwise or counterclockwise. This allows crucible 10 attached to crucible holding shaft 13 to be moved in the vertical direction and, further, to be rotated as necessary. A pedestal may be attached to crucible holding shaft 13, and crucible 10 may be fixed to this pedestal.

[0052] In addition, the crucible holding shaft 13 may have a hollow structure inside, allowing a thermocouple to be inserted or used as a path for passing measurement light from a radiation thermometer, thereby enabling temperature measurement at the bottom of the crucible (near the crucible).

[0053] Crystal holding shaft 14 is a member that allows a silicon carbide seed crystal 30 to be attached to its lower end (tip), and that allows the tip to be positioned inside furnace 11. Crystal holding shaft 14 is also configured to be movable in the vertical direction. FIG. 1 shows a state in which seed crystal 30 is attached. By moving crystal holding shaft 14 in the vertical direction, seed crystal 30 attached to its tip also moves in the vertical direction in conjunction, and by adjusting its height position, a single crystal is produced.

[0054] This crystal holding shaft 14 may be configured to be rotatable clockwise or counterclockwise, similar to the crucible holding shaft 13. In other words, the crystal holding shaft 14 and the crucible holding shaft 13 may or may not have a rotation mechanism.

[0055] [Method for producing silicon carbide single crystals] Next, a method for producing a silicon carbide single crystal will be described using the silicon carbide single crystal production apparatus 101 described above and a crucible made of graphite (C) or silicon carbide (SiC) as the crucible 10.

[0056] (a) A step of preparing a solution As described above, the solution 20 is accommodated in the crucible 10. To this end, first, raw materials for the solution 20, for example, a silicon (Si) raw material, and, if necessary, a carbon (C) raw material and raw materials of additive elements such as chromium (Cr) and aluminum (Al), are weighed out to a desired composition ratio and placed in the crucible 10. Note that although carbon is an essential component of SiC single crystals, it does not necessarily have to be added as a raw material here.

[0057] Then, power is applied to the induction coil 12, and the raw material is heated to a crystal growth temperature by induction heating, and the raw material is melted. As this heating continues, the carbon that constitutes the crucible 10 dissolves into the raw material solution, and a carbon-containing solution 20 is obtained.

[0058] In preparing this solution, it is preferable to attach a silicon carbide seed crystal 30 to the tip of the crystal holding shaft 14 and introduce this seed crystal 30 into the furnace 11 (above the crucible 10).

[0059] The solution 20 thus obtained contains, in addition to Si and C, elements such as Cr and Al, which are added as needed. Furthermore, other elements may be added as needed. It is preferable that the solution is substantially free of elements that are not necessary for the production of SiC single crystals. Here, "substantially free" means that the solution 20 may contain other elements that are contained as impurities in the raw materials of the Si source, C source, Cr source, Al source, and other element sources that may be added, or other elements that are inevitably contained from the production equipment during the production of silicon carbide seed crystals.

[0060] Examples of other elements that are inevitably contained include N, B, Fe, etc. "Substantially free of these elements" means that the content of each of these elements is 1 atomic % or less.

[0061] In this embodiment, in step (a), a solidified solution used in a previous method for producing a silicon carbide single crystal is used as the raw material for the solution. By using a solidified solution of the used solution in this manner, it is possible to reuse unconsumed components such as Cr and Al, thereby achieving effective utilization of resources.

[0062] The SiC single crystal may be produced in the same manner as in the conventionally known method for producing a SiC single crystal, except that a solidified used solution is used as the raw material for this solution 20.

[0063] However, the composition of each component in the solidified product of this used solution differs from the composition required to produce the desired SiC crystal, and therefore, in the step of preparing this solution, it is necessary to adjust the content of each component so that the desired composition is obtained.

[0064] In the method for producing a SiC single crystal according to the present embodiment, the missing elements are added to the solidified product of the used solution before starting the growth and cultivation of the SiC single crystal. At this time, the amount of the missing elements to be added can be calculated based on the results of a composition analysis of the solidified product of the used solution.

[0065] The composition analysis is not particularly limited as long as it is a known composition analysis method. For this composition analysis, a composition analysis device such as an X-ray fluorescence analyzer (XRF) using X-ray irradiation, an analytical scanning electron microscope (SEM-EDS), or an analyzer using inductively coupled plasma atomic emission spectroscopy (ICP-AES) can be preferably used. These composition analysis devices are preferred because they can simultaneously analyze the elements contained.

[0066] The sample used for this composition analysis may be obtained by cutting out any part of the solidified product from the used solution, but it is preferable to use a sample near the center of the solidified product because the element concentrations are stable. That is, this solidified product is obtained by cooling to about room temperature after the production of SiC single crystals, and during the cooling process, the temperature drops relatively quickly at the interface between the solution and the crucible (especially the interface with the bottom of the crucible) and at the liquid surface of the solution, and at that time, miscellaneous SiC crystals may be generated, causing fluctuations in the element concentrations.

[0067] From the spectrum obtained by such composition analysis, the content ratios of elements such as Si, Cr, and Al contained in the solidified product of the used solution are calculated.

[0068] Although C is also contained in the solidified product of the used solution, it is difficult to detect C in the above composition analysis and its source is mainly the crucible, so it is not necessarily an element that needs to be added as a deficient element and its amount must be calculated. Therefore, the C content does not need to be taken into consideration in this composition analysis.

[0069] Then, based on the calculated content ratio, when reusing the solution, the missing elements are added to obtain a solution of the desired composition, for example, so that the composition is equivalent to that of a new solution used in a previous SiC single crystal production. The new solution corresponds to the solution immediately after the raw materials are dissolved in the crucible.

[0070] During the production of SiC single crystals, Si is consumed as the main component of the single crystal, and its amount decreases in the solution. On the other hand, Cr and Al do not constitute SiC single crystals and are not consumed components, so they remain in the solution. As mentioned above, Si is consumed, but Cr and Al are not, so these components remain concentrated in the used solution. Although Al is not consumed, because the production of SiC single crystals involves high temperatures of over 1800°C, it evaporates and its content decreases, and there are cases where the concentration fluctuation is small.

[0071] Here, the solidified material of the used solution to be reused may be reused in its entirety, or only a portion of it. For example, if the crucible does not collapse when the used solution is solidified and can be reused again for the production of SiC single crystals, the entire solidified material contained in the crucible can be reused as is, along with the crucible. Furthermore, since miscellaneous crystals are generated during the solidification process of the solution as described above, the solidified material containing the miscellaneous crystals can be removed before reuse. When removing the miscellaneous crystals, for example, the solidified material to be reused is cut out from the solidified material and placed in a new crucible for reuse.

[0072] Next, any elements that are lacking are added to the solidified material to be reused. For example, if only a portion of the material is reused, all elements except for C, such as Si, Cr, and Al, will be lacking. Therefore, the required amounts of each element are calculated, and the raw materials for the missing elements are added. These are then heated in a crucible to prepare a solution for producing SiC single crystals.

[0073] The amount of the missing element to be added can be calculated by performing the calculation of the following formula (1) for each component.

[0074] Mass of component added (A) = Mass of component in solution composition at the start of growth (A') - Mass of component in solution to be reused (A R ) ···(1) Here, the component mass (A') of the solution composition at the start of growth is the desired (target) composition of the solution to be used in the subsequent production of SiC single crystals, and therefore the numerical value is known for each component. On the other hand, the component mass (A R ) is unknown because it varies depending on the solution used, manufacturing conditions, etc. in the production of the SiC single crystal. Therefore, the content ratio of each component is calculated by performing the composition analysis described above. Specifically, the mole fraction of each component in the solidified product of the solution to be reused is calculated from the spectrum obtained by the composition analysis, and the mass content of each component in the solution to be reused can be calculated. Note that when calculating this mole fraction, the carbon (C) component can be ignored.

[0075] For example, assume that a new solution of Si-40 mol% Cr-2 mol% Al with a total mass of 500 g is used to produce a SiC single crystal, and after production of the single crystal is completed, 200 g of the solidified used solution is cut out and reused.

[0076] At this time, since the composition of the solidified used solution is unknown, a sample for composition analysis is prepared and the composition is analyzed using an X-ray fluorescence analyzer (XRF), an analytical scanning electron microscope (SEM-EDS), etc. Here, we will use as an example a case where the molar fractions of each component calculated by composition analysis are 50 mol% Cr and 2 mol% Al (Si-50 mol%Cr-2 mol%Al) (Si is contained as the remainder, and C is not taken into account).

[0077] The content of each component in 200 g of solidified solution can be calculated from its atomic weight and the above molar fraction. Furthermore, using these results, the amount of each component to be added can be calculated as follows: Here, the solution prepared for reuse is shown as having the same composition as the new solution and the same amount (500 g), but this can be changed each time to have any desired composition and mass.

[0078] <Si addition amount> Mass of solution components at the start of growth (A'): 216.46g The mass of the components of the solution to be reused (A R ): 67.37g · Added mass (A): 149.09g (=A'-A R ) <Amount of Cr added> Mass of solution components at the start of growth (A'): 276.37g The mass of the components of the solution to be reused (A R ): 129.93g · Added mass (A): 146.44g (=A'-A R ) <Amount of Al added> Mass of solution components at the start of growth (A'): 7.17g The mass of the components of the solution to be reused (A R ): 2.70g ·Additional mass (A): 4.47g (=A'-A R ) In this way, the amount of each component required to be added is calculated, the solidified solution to be reused and the raw material of each component in an amount equivalent to the mass of the component to be added are placed in a new crucible, and the SiC single crystal manufacturing operation is carried out as described above, thereby enabling the SiC single crystal manufacturing method of this embodiment to be carried out. This makes it possible to effectively utilize the elements and reduce manufacturing costs.

[0079] (b) growing silicon carbide single crystals Next, the crystal holding shaft 14 with the seed crystal 30 attached thereto, which has been introduced into the furnace 11, is lowered to contact the solution, and a SiC single crystal is grown on the crystal growth surface 30a. At this time, first, as shown in FIG. 1, the crystal growth surface 30a of the seed crystal 30 attached to the crystal holding shaft 14 is positioned so as to be in contact with the solution 20.

[0080] In this step (b), the seed crystal 30 only needs to be brought into contact with the solution 20, and in particular, the seed crystal 30 can be brought into contact with the surface of the solution 20 (without being submerged in the solution 20) to initiate the growth of a single crystal.

[0081] The single crystal is grown by moving the crystal holding shaft 14 upward or downward or by maintaining it in a contacting position, thereby growing a silicon carbide single crystal (SiC single crystal) on the underside of the seed crystal 30. On the other hand, when crystal growth is to be terminated, the crystal holding shaft 14 is pulled up to separate the SiC single crystal from the solution 20. This terminates the growth of the SiC single crystal.

[0082] As described above, a SiC single crystal can be produced by operating single crystal production apparatus 101. Although it has been described that crystal growth is terminated by pulling up crystal holding shaft 14 to separate the SiC single crystal from solution 20, this is not limiting, and for example, instead of pulling up crystal holding shaft 14, crucible holding shaft 13 can be pulled down to separate the SiC single crystal from solution 20, thereby terminating crystal growth.

[0083] The above-mentioned operation can be achieved by providing a control unit for operating in that manner, and in single crystal manufacturing apparatus 101, SiC single crystals are manufactured by the "solution growth method."

[0084] Specifically, the control unit controls induction heating by induction coil 12, vertical movement and rotation of crystal holding shaft 14, vertical movement and rotation of crucible holding shaft 13, etc., and operates to ensure that the above-mentioned single crystal manufacturing method proceeds smoothly.

[0085] The seed crystal 30 used here can be any of the known polytypes known as silicon carbide single crystals, such as 2H, 3C, 4H, and 6H. The differences in polytypes are due to differences in atomic arrangement, and the effects of this crystal growth method may vary depending on the type of polytype. In this embodiment, for example, a 4H silicon carbide single crystal is preferably used as the seed crystal, and the resulting silicon carbide single crystal is suitable for use in power semiconductor devices.

[0086] FIG. 2 is an example of a diagram schematically illustrating a cross section of a silicon carbide seed crystal 30. In FIG. 2, the seed crystal 30 is a 4H silicon carbide single crystal, and is an off-substrate having an off-angle θ. Specifically, the crystal growth surface 30a (or the normal 30n of the crystal growth surface 30a) is aligned with the c-plane ( <0001> It is preferable that the off-angle θ is inclined from the <11-20> direction (off-angle θ) to the <11-20> direction. This off-angle θ is preferably 0.5° or more and 8° or less, and more preferably 0.5° or more and 5° or less. The seed crystal 30 may be an on-substrate whose crystal growth surface is the c-plane. Although the on-substrate has an off-angle of 0°, an off-angle of less than 0.5° may inevitably be formed when producing a seed crystal as an on-substrate. Therefore, the on-substrate substantially has an off-angle of 0 to less than 0.5°.

[0087] To improve the crystal quality of silicon carbide single crystals, it is necessary to reduce the dislocation density. However, threading dislocations, such as threading edge dislocations parallel to the c-axis, present in silicon carbide seed crystals propagate and remain in the grown silicon carbide single crystal. For this reason, it is generally not easy to improve the crystal quality beyond that of the silicon carbide seed crystal.

[0088] However, a known method for solving this problem is to use step-flow growth to bend the direction of threading dislocations parallel to the c-axis, i.e., convert them to dislocations parallel to the a-axis, thereby excluding them from the crystal growth direction and reducing the dislocation density. To achieve this, it is effective to use a substrate with an off-angle as a seed crystal to increase the microsteps on the crystal growth surface. In other words, using a seed crystal with such an off-angle can reduce the dislocation (defect) density. On the other hand, step-flow growth can also be performed on an on-substrate by performing concave growth, thereby reducing the dislocation density.

[0089] The temperature during growth of the SiC single crystal is preferably 1800°C or higher and 2200°C or lower, and more preferably 1900°C or higher and 2100°C or lower. It is even more preferable to grow the crystal at a temperature of 1950°C or higher and 2050°C or higher. The conditions for the temperature gradient near the crystal growth surface 30a, the atmosphere in the furnace 11 during growth, the pressure, and other conditions can be set to the same conditions as those used in conventional solution growth methods.

[0090] During the growth of the SiC single crystal, the silicon carbide seed crystal 30 may be rotated by a rotating shaft 14. Similarly, the crucible 10 may be rotated by a rotating shaft 13.

[0091] In this way, silicon carbide single crystals can be produced by the above-described steps (a) and (b). After the silicon carbide single crystal has grown sufficiently, the growth is stopped to obtain a SiC single crystal ingot. At this time, heating by the induction coil or the like in the silicon carbide single crystal production apparatus is stopped, and the used solution is cooled to room temperature and solidified. The solidified material thus obtained can be reused as a raw material for the solution when silicon carbide single crystals are next produced, as described above.

[0092] (Regarding solidified used solution) As described above, this embodiment is characterized in that the solidified product of the used solution is reused as a raw material for the solution in the subsequent method for producing a SiC single crystal. As described above, the solidified product of the used solution may be reused in its entirety or in part, but it is preferable to remove miscellaneous crystals and use only part of it. This is because, if miscellaneous crystals are contained in the solution for producing a SiC single crystal, crystal growth may progress not only on the seed crystal 30 but also on the miscellaneous crystals, which may prevent efficient production of a SiC single crystal.

[0093] The miscellaneous crystals can be removed from the solidified product of the used solution by, for example, cutting the area of ​​the solidified product containing the miscellaneous crystals. Fig. 3 shows a cross-sectional view of crucible 10 containing solidified product 21 obtained by cooling the used solution after once producing a SiC single crystal.

[0094] 3, the inner surface of crucible 10 that was in contact with the solution for producing a SiC single crystal has been deformed by the melting during single crystal production, the cooling operation after heating, and other processes. Also, miscellaneous crystals 22 have formed in the solution. As explained above, miscellaneous crystals 22 tend to form in the vicinity of the contact interface between solution 20 and the bottom surface of crucible 10 and in the region of the liquid surface of solution 20 close to crucible 10.

[0095] As shown in FIG. 3, if miscellaneous crystals 22 are formed in solidified material 21 of the used solution, the miscellaneous crystals 22 are removed by, for example, cutting horizontally along cutting lines X1 and X2 and vertically along cutting lines Y1 and Y2 to cut out a region R (the region surrounded by cutting lines X1, X2 and cutting lines Y1, Y2) that does not contain miscellaneous crystals 22, and the region R is reused as a raw material for a solution in the next method for producing a SiC single crystal.

[0096] The solidified material 21 cut out here can be reused as it is, but it is preferable to cut it into small pieces so that it can be easily dissolved by heating when preparing the next solution. In this case, for example, it is preferable to cut the solidified material 21 so that its maximum dimension is 2 / 3 or less of the inner diameter of the crucible.

[0097] [Modification] In the above embodiment, the following method for producing a SiC single crystal is described as being carried out by performing a composition analysis of the deficient elements, calculating the amount of the deficient element from the composition analysis results, and adding the raw material to make up the deficient amount.

[0098] Incidentally, if a SiC single crystal has previously been produced using a solidified used solution under the same conditions, the above composition analysis can be omitted. For example, if a SiC single crystal production method is being performed using a solidified used solution as a raw material under the same production conditions (solution composition and SiC single crystal growth conditions) as those planned for the next production, the results of the composition analysis obtained in the previous method can be used to calculate the amount of added missing elements without performing composition analysis.

[0099] For example, if SiC production is repeated under the same conditions as described in the composition analysis above, 500 g of a raw material with a composition of Si-40 mol% Cr-2 mol% Al is prepared as the solution before SiC production for reuse, and the solidified material from the used solution after SiC single crystal production has a composition of Si-50 mol% Cr-2 mol% Al. In this case, the amount of solidified material to be reused is not constant and is likely to increase or decrease each time. However, since the molar fraction of each component contained in the resulting solidified material is known, the content of each component can be calculated from the mass to be reused. Therefore, the added mass of each component can be calculated using the above formula (1) without performing composition analysis.

[0100] This is particularly useful when the reuse conditions (solution composition and SiC single crystal manufacturing conditions) are kept the same and the SiC single crystal is manufactured repeatedly. Note that, in this case, the number of times the reuse is repeated is not particularly limited, and the used solution can be reused any number of times.

[0101] <Example> A specific example will be described below.

[0102] As used herein, the term "feed solution" refers to a solution used to grow a silicon carbide single crystal by contacting it with a silicon carbide seed crystal. In other words, it is a solution used to grow a single crystal by a solution growth method. In particular, a "feed solution" is a solution that is used for the first time. In contrast, a "recycled solution" is a solution that has already been used to grow a single crystal by a solution growth method, and is recovered and reused for growing a single crystal.

[0103] First, a "feed solution" was prepared. Specifically, a "feed solution" containing, for example, 53.5 atomic % silicon (Si), 40 atomic % chromium (Cr), and 6.5 atomic % other elements (Al, Y, Mo, La, Ce, Pr, Nd, Tb, Gd, W, etc.) was prepared.

[0104] The prepared "feed solution" was then placed in a graphite crucible to grow a single crystal. The solidified "feed solution" used for growing the single crystal was then cut together with the graphite crucible. Specifically, as shown in FIG. 3, the solidified material in the region R, which did not contain miscellaneous crystals, was cut along the cutting lines X1 and X2 and the cutting lines Y1 and Y2.

[0105] The cut solidified material was then placed in an alumina crucible and remelted by heating at 1500°C for 5 minutes, after which it was solidified. The remelted solution was designated the "recycled solution," and the composition of Si, Cr, and other elements contained in this "recycled solution" was measured by inductively coupled plasma atomic emission spectroscopy (ICP-AES). The composition of Si, Cr, and other elements in the "feed solution" was also measured in a similar manner.

[0106] Table 1 shows the results of measuring the composition of Si, Cr, and other elements in the "feed solution" and the "recycled solution."

[0107] [Table 1]

[0108] As shown in Table 1, the composition of the "recycled solution" differs from that of the "feed solution." To reuse the "recycled solution" for single crystal growth, the composition of the "recycled solution" must be made to match that of the "feed solution." For this reason, additional samples must be added to the "recycled solution." Specifically, additional samples are added to the "recycled solution" so that it contains 53.5 atomic % Si, 40 atomic % Cr, and 6.5 atomic % of other elements.

[0109] Table 2 shows the composition weights of the additional samples and the recycled samples that make up the "recycled solution."

[0110] [Table 2]

[0111] Here, for example, the weight of the additional Si sample can be calculated using the following formula:

[0112] Weight of additional Si sample = (weight of Si in the starting composition (g / mol)) - (weight of recycled Si sample per mol) x (weight of recycled sample) / (weight of alloy per mol) The term "alloy" used here refers to an alloy containing Si, Cr, and other elements, and refers to an alloy with the same composition as the recycled sample. While the formula for calculating the weight of the additional Si sample has been explained here, the weight of the additional Cr sample and the weight of the additional samples of other elements can be calculated in the same way. In this way, additional samples can be added to the recycled sample.

[0113] For example, in the example shown in Table 2, the weight of the recycled sample is 327.4 g and the weight of the additional sample is 301.0 g. Therefore, the recycle rate in the example shown in Table 2 is 52%.

[0114] In this specification, the solution obtained by dissolving the recycled sample as an additional sample is called a "recycled solution." The composition of this "recycled solution" is the same as the composition of the "feed solution."

[0115] Below we describe an example of growing a single crystal using a "recycled solution."

[0116] For example, the "recycled solution" was poured into a graphite crucible with an inner diameter of 90 mm and an outer diameter of 110 mm to a solution height of approximately 25 mm. A seed crystal consisting of a 4H-SiC single crystal with a diameter of 40 mm, a thickness of 800 μm, and a 1° off-angle was prepared. The (0001) plane was attached to a graphite holder with a diameter of approximately 40 mm using graphite adhesive, so that the growth surface would be the (000-1) plane. This graphite holder was attached to a graphite rod to serve as the crystal axis.

[0117] In a high-frequency heating furnace (single crystal growth apparatus), the furnace was evacuated, and then argon (Ar) was introduced up to atmospheric pressure, and single crystals were grown in the Ar atmosphere. The internal configuration of the high-frequency heating furnace is, for example, as shown in Figure 1.

[0118] The growth of the single crystal was carried out by raising the temperature of the graphite crucible and the "recycled raw materials" contained in the graphite crucible to melt the "recycled raw materials." Here, the "recycled raw materials" referred to in this specification refer to the solidified "recycled solution." In other words, the "recycled solution" is a solution formed by melting the "recycled raw materials."

[0119] The single crystal was then grown by immersing the seed crystal in the "recycle solution" and holding it at approximately 1900°C for 16 hours. During this time, the crystal axis and the graphite crucible were rotated. The temperature distribution of the "recycle solution" was adjusted so that the side of the graphite crucible was hot and the seed crystal was cold. After holding this state for 16 hours, the grown single crystal was lifted off the surface of the "recycle solution" and cooled by slowly lowering the temperature over the course of one hour.

[0120] In this way, a "recycled solution" can be used to grow single crystals.

[0121] Table 3 shows the growth rate and growth surface condition of single crystals grown in the "recycled solution" compared with the growth rate and growth surface condition of single crystals grown in the "feed solution."

[0122] [Table 3]

[0123] As shown in Table 3, the growth rate was almost the same for the "recycled solution" and the "starting solution," and the growth rate when the "recycled solution" was used was 83 (μm / h), confirming that the growth rate could be secured even when the "recycled solution" was used.

[0124] Furthermore, when comparing the quality of the crystal faces, the average step height of the crystal faces measured with a laser microscope was almost the same for the "recycled solution" and the "prepared solution," and the step height when the "recycled solution" was used was 5 μm or less, confirming that a low step height could be ensured even when the "recycled solution" was used. The average step height was the average of the step heights measured from 36 laser microscope images (all over the crystal surface).

[0125] These findings demonstrate that it is possible to grow single crystals while maintaining the growth rate and quality even when using a "recycled solution."

[0126] <Further innovations> As described in the above examples, in order to prepare a "recycled solution" having the same composition as the "feed solution," it is necessary to add an additional sample to the "recycled solution" (recycled sample). At this time, it is necessary to measure the composition of the "recycled solution," for example, by ICP analysis. This is because, unless the composition of the "recycled solution" is measured, it is impossible to prepare a "recycled solution" having the same composition as the "feed solution." In other words, unless the composition of the "recycled solution" is measured, it is impossible to determine the amount of missing elements to be added.

[0127] In this regard, if the composition of the "recycled solution" can be grasped without performing measurements such as ICP analysis, the production efficiency of the "recycled solution" can be improved. Therefore, in this embodiment, a device is devised to grasp the composition of the "recycled solution" without performing measurements such as ICP analysis. The technical concept of this device in this embodiment will be described below.

[0128] Specifically, a further innovation in this embodiment is that a system is constructed to estimate the composition of a used solution by using a regression model that outputs an estimated value for the composition of a used solution ("recycled solution") when a pre-use composition value indicating the composition of the solution before use ("feed solution") and manufacturing conditions for growing a single crystal are input.

[0129] In the following, we will mainly explain an example in which the above-mentioned composition estimation system is configured from a single computer (composition estimation device), but the composition estimation system in this embodiment can also be realized as a distributed system consisting of multiple computers.

[0130] <<Configuration of the composition estimation device>> <<<Hardware Configuration>>> First, the hardware configuration of the composition estimation device in this embodiment will be described.

[0131] Fig. 4 is a diagram showing an example of the hardware configuration of composition estimation apparatus 100 according to the present embodiment. Note that the configuration shown in Fig. 4 merely shows an example of the hardware configuration of composition estimation apparatus 100, and the hardware configuration of composition estimation apparatus 100 is not limited to the configuration shown in Fig. 4 and may be other configurations.

[0132] 4, composition estimation apparatus 100 includes a central processing unit (CPU) 101A that executes a program. CPU 101A is electrically connected to, for example, read-only memory (ROM) 102, random access memory (RAM) 103, and hard disk drive 112 via bus 113, and is configured to control these hardware devices.

[0133] The CPU 101A is also connected to input devices and output devices via a bus 113. Examples of input devices include a keyboard 105, a mouse 106, a communication board 107, and a scanner 111. Examples of output devices include a display 104, a communication board 107, and a printer 110. The CPU 101 may also be connected to, for example, a removable disk device 108 and a CD / DVD-ROM device 109.

[0134] Composition estimation device 100 may be connected to, for example, a network. For example, when composition estimation device 100 is connected to other external devices via a network, communication board 107, which constitutes a part of composition estimation device 100, is connected to a LAN (local area network), a WAN (wide area network), or the Internet.

[0135] RAM 103 is an example of a volatile memory, and recording media such as ROM 102, removable disk device 108, CD / DVD-ROM device 109, and hard disk device 112 are examples of non-volatile memory. These volatile memories and non-volatile memories constitute a storage device of composition estimation device 100.

[0136] Hard disk drive 112 stores, for example, an operating system (OS) 201, a program group 202, and a file group 203. The programs included in program group 202 are executed by CPU 101A using operating system 201. RAM 103 also temporarily stores at least some of the programs of operating system 201 and application programs executed by CPU 101A, as well as various data required for processing by CPU 101A.

[0137] A BIOS (Basic Input Output System) program is stored in ROM 102, and a boot program is stored in hard disk drive 112. When composition estimation device 100 is started up, the BIOS program stored in ROM 102 and the boot program stored in hard disk drive 112 are executed, and operating system 201 is started up by the BIOS program and the boot program.

[0138] Program group 202 stores programs that realize the functions of composition estimation apparatus 100, and these programs are read and executed by CPU 101A. Furthermore, file group 203 stores information, data, signal values, variable values, and parameters indicating the results of processing by CPU 101A as each item of a file.

[0139] The files are recorded on a recording medium such as the hard disk drive 112 or a memory. The information, data, signal values, variable values, and parameters recorded on a recording medium such as the hard disk drive 112 or a memory are read by the CPU 101A into the main memory or cache memory and used for the operations of the CPU 101A, such as extraction, search, reference, comparison, calculation, processing, editing, output, printing, and display. For example, during the operations of the CPU 101A described above, the information, data, signal values, variable values, and parameters are temporarily stored in the main memory, registers, cache memory, buffer memory, etc.

[0140] The functions of composition estimation apparatus 100 may be realized by firmware stored in ROM 102, or may be realized by software alone, hardware alone such as elements, devices, substrates, and wiring, a combination of software and hardware, or even a combination of firmware and firmware. Firmware and software are recorded as programs on a recording medium such as hard disk drive 112, a removable disk, a CD-ROM, or a DVD-ROM. The programs are read and executed by CPU 101A. In other words, the programs cause a computer to function as composition estimation apparatus 100.

[0141] Thus, composition estimation device 100 is a computer equipped with CPU 101A as a processing device, hard disk drive 112 and memory as storage devices, keyboard 105, mouse 106, and communication board 107 as input devices, and display 104, printer 110, and communication board 107 as output devices. The functions of composition estimation device 100 are realized using the processing device, storage device, input device, and output device.

[0142] <<<Function block configuration>>> Next, the functional block configuration of the composition estimation device 100 will be described.

[0143] FIG. 5 is a functional block diagram showing the functions of composition estimation device 100.

[0144] The composition estimation device 100 includes an input unit 301 , a regression model generation unit 302 , an estimation unit 303 , an output unit 304 , and a data storage unit 305 .

[0145] The input unit 301 is configured to input relevant data.

[0146] Here, "related data" refers to data consisting of pre-use composition data, manufacturing condition data, and used composition data, and is data that associates pre-use composition data, manufacturing condition data, and used composition data. Pre-use composition data is data consisting of pre-use composition values ​​that indicate the composition of the solution before use ("feed solution"). Manufacturing condition data is data consisting of manufacturing conditions when producing single crystals by solution growth. Used composition data is data consisting of used composition values ​​that indicate the composition of the solution after use ("recycled solution").

[0147] For example, FIG. 6 is a table showing an example of related data.

[0148] In FIG. 6, it can be seen that the related data is composed of pre-use composition data, manufacturing condition data, and used composition data.

[0149] The pre-use composition data is composed of the composition values ​​of elements contained in the solution before use (pre-use composition values). For example, in the example shown in FIG. 6, the solution before use contains at least Si, Cr, and Al, and the composition values ​​of each element are specified. Focusing on the related data "No. 1" shown in FIG. 6, the pre-use composition data includes a composition value of Si of "53.5," a composition value of Cr of "40," and a composition value of Al of "6.5." Similarly, focusing on the related data "No. 2" shown in FIG. 6, the pre-use composition data includes a composition value of Si of "60," a composition value of Cr of "40," and a composition value of Al of "0."

[0150] Next, the manufacturing condition data includes, as items, for example, "growth temperature," "growth time," "crucible diameter," "solution height," and "rotation speed (crucible and crystal axis)."

[0151] Next, the used composition data is composed of the composition values ​​(used composition values) of elements contained in the used solution. For example, in the example shown in FIG. 6, the used solution contains at least Si, Cr, and Al, and the composition values ​​of each element are specified. Focusing on the related data "No. 1" shown in FIG. 6, the used composition data includes a composition value of Si of "43.2," a composition value of Cr of "45.5," and a composition value of Al of "11.3."

[0152] The related data configured in this manner is input from the input unit 301 and then stored in the data storage unit 305. This data storage unit 305 functions as a database that stores multiple pieces of related data. The input unit 301 is also configured to input various types of data other than the related data. For example, data input to the input unit 301 can include single pre-use composition data, single manufacturing condition data, etc. These various types of data are also input from the input unit 301 and then stored in the data storage unit 305.

[0153] The regression model generating unit 302 has a function of generating a regression model based on the related data stored in the data storage unit 305. That is, the regression model generating unit 302 is configured to generate a regression model that correlates pre-use composition values, manufacturing conditions, and used composition values. Specifically, as shown in Fig. 7, the regression model generating unit 302 is configured to generate a regression model that uses pre-use composition values ​​and manufacturing conditions as inputs and uses used composition values ​​(estimated values) as output, by machine learning using the related data as training data.

[0154] Here, a "regression model" is defined as a function that, when a pre-use composition value and manufacturing conditions are input, outputs a used composition value (estimated value) corresponding to these pre-use composition values ​​and manufacturing conditions. In other words, a "regression model" is defined as a function that, when a pre-use composition value and manufacturing conditions whose correspondence with the used composition value is unknown are input, outputs an estimate of the used composition value that is presumed to be realized under these pre-use composition values ​​and manufacturing conditions. In this way, a regression model can be said to be a function used to estimate the composition value of a used solution when growing a single crystal using a pre-use solution whose correspondence with the composition of the used solution is unknown.

[0155] The estimation unit 303 is configured to estimate the composition of the used solution based on a pre-use composition value indicating the composition of the solution before use, manufacturing conditions for growing a single crystal, and a regression model. That is, the estimation unit 303 is configured to obtain an estimate of a used composition value indicating the composition of the used solution using the regression model generated by the regression model generation unit 302. For example, when a pre-use solution whose correspondence with the composition of the used solution is unknown is used and a single crystal is grown under predetermined manufacturing conditions, the estimation unit 303 has a function of estimating the composition value of the used solution based on a regression model whose accuracy has been improved by machine learning.

[0156] The output unit 304 outputs the estimated value of the used composition value estimated by the estimation unit 303.

[0157] In this manner, composition estimation device 100 is configured.

[0158] In this embodiment, composition estimation apparatus 100 is provided with regression model generation unit 302 that generates a regression model and estimation unit 303 that estimates the composition value of a used solution. In other words, in this embodiment, a composition estimation system that estimates the composition of a used solution that has been used to grow a single crystal by contacting a seed crystal with the solution is realized by a single composition estimation apparatus 100.

[0159] However, the technical idea of ​​this embodiment is not limited to this configuration, and it is also possible to configure a composition estimation system using a distributed system in which a regression model generation device including the regression model generation unit 302 and an estimation device including the estimation unit 303 are configured on separate computers, and the regression model generation device and the estimation device are connected using a network, for example.

[0160] <<Operation of the composition estimation device>> Composition estimation apparatus 100 is configured as described above, and its operation will be described below. The operation of composition estimation apparatus 100 includes an "operation for generating a regression model" and an "operation for estimating the composition of a used solution." Therefore, only these operations will be described below.

[0161] <<<Regression model generation operation>>> FIG. 8 is a flowchart illustrating the operation of generating a regression model.

[0162] 8, first, the input unit 301 inputs a plurality of pieces of related data that associate pre-use composition data, manufacturing condition data, and used composition data (S101). Then, the plurality of pieces of related data input to the input unit 301 are stored in the data storage unit 305 (S102).

[0163] Next, the regression model generating unit 302 generates a regression model based on the related data stored in the data storage unit 305 (S103). Specifically, the regression model generating unit 302 generates a regression model by machine learning using the related data as training data, in which the input is the pre-use composition value and the manufacturing conditions, and the output is an estimated value of the used composition value (see FIG. 7).

[0164] Then, the regression model generated by the regression model generating unit 302 is stored in the data storage unit 305 (S104). In this manner, the operation of generating a regression model is performed.

[0165] <<<Estimating the composition of used solutions>>> Next, the operation of estimating the composition of the used solution will be described.

[0166] 9 is a flowchart illustrating the operation of estimating the composition of a used solution. The regression model is already stored in the data storage unit 305.

[0167] In FIG. 9, first, the input unit 301 inputs the composition of the solution before use (pre-use composition value), whose correspondence with the composition of the used solution is unknown, and also inputs the manufacturing conditions for the solution growth method using this pre-use solution (S201).

[0168] The estimation unit 303 then substitutes the above-described pre-use composition values ​​and manufacturing conditions into the regression model (S202). As a result, the regression model outputs an estimated value of the used composition value, which indicates the composition of the used solution, based on the input pre-use composition values ​​and manufacturing conditions. This allows the estimation unit 303 to obtain the estimated value (S203). The output unit 304 then outputs the estimated value obtained by the estimation unit 303 (S204).

[0169] In this way, the composition estimation device 100 can obtain an estimate of the composition of the used solution (used composition value) that is likely to be realized when a single crystal is grown by a solution growth method under specified manufacturing conditions using a pre-used solution having an unknown composition (pre-used composition value) whose correspondence with the composition of the used solution is unknown. As a result, according to this embodiment, the composition of the used solution ("reused solution") that is likely to be realized can be determined without performing measurements such as ICP analysis, thereby improving the efficiency of producing a "recycled solution." In this respect, the technical concept of this embodiment is useful from the perspective of reducing the manufacturing costs of single crystals.

[0170] <Composition estimation program> The composition estimation method performed by the composition estimation apparatus 100 described above can be realized by a composition estimation program that causes a computer to execute a composition estimation process.

[0171] For example, in composition estimation apparatus 100 configured as a computer shown in Fig. 4, the composition estimation program of this embodiment can be installed as one of programs 202 stored in hard disk drive 112. Then, by having the computer that is composition estimation apparatus 100 execute this composition estimation program, the composition estimation method of this embodiment can be realized.

[0172] The composition estimation program that causes a computer to execute each process for generating data related to the composition estimation process can be recorded on a computer-readable recording medium and distributed. Examples of the recording medium include magnetic storage media such as hard disks and flexible disks, optical storage media such as CD-ROMs and DVD-ROMs, and hardware devices such as non-volatile memories such as ROMs and EEPROMs.

[0173] The invention made by the inventor has been specifically described above based on the embodiments thereof, but it goes without saying that the present invention is not limited to the above-described embodiments and can be modified in various ways without departing from the spirit of the invention. [Explanation of symbols]

[0174] 10 Crucible 11 Furnace 12 induction coil 13 Crucible holding shaft 14 Crystal holding shaft 20 solution 21 Solidified materials 22 Miscellaneous crystals 30 seed crystals 30a Crystal growth surface 100 Composition estimation device 101 Single crystal manufacturing equipment 101A CPU 102 ROM 103 RAM 104 Display 105 keyboard 106 Mouse 107 Communication Board 108 Removable disk device 109 CD / DVD-ROM device 110 Printer 111 Scanner 112 Hard disk drive 201 Operating Systems 202 Programs 203 files 301 Input section 302 Regression model generation unit 303 Estimation Department 304 Output section 305 Data storage unit R Region without miscellaneous crystals

Claims

1. (a) providing a solution containing silicon and carbon; (b) contacting a silicon carbide seed crystal with the solution and growing a silicon carbide single crystal on a crystal growth surface of the seed crystal; A method for producing a silicon carbide single crystal, comprising: As a raw material for the solution in the step (a), a solidified material obtained by cooling a solution used in a previous method for producing a silicon carbide single crystal to about room temperature is used. A method for producing silicon carbide single crystals.

2. 2. The method for producing a silicon carbide single crystal according to claim 1, In the step (a), a composition is adjusted by adding a deficient element to the solidified material.

3. The method for producing a silicon carbide single crystal according to claim 2, A method for producing a silicon carbide single crystal, wherein silicon is added as the deficient element.

4. The method for producing a silicon carbide single crystal according to claim 2, The method for producing a silicon carbide single crystal, wherein the amount of the deficient element to be added is calculated based on the results of a composition analysis of the solidified product.

5. The method for producing a silicon carbide single crystal according to claim 4, The method for producing a silicon carbide single crystal, wherein the composition analysis is performed using an analytical device utilizing an X-ray fluorescence analyzer (XRF), an analytical scanning electron microscope (SEM-EDS), or an ICP optical emission spectroscopy.

6. 2. The method for producing a silicon carbide single crystal according to claim 1, The method for producing a silicon carbide single crystal, wherein the solidified product of the used solution is used after removing miscellaneous crystals contained therein.

7. 7. The method for producing a silicon carbide single crystal according to claim 6, The method for producing a silicon carbide single crystal, wherein the removal of the miscellaneous crystals is carried out by cutting out the inside of the solidified product of the used solution.

8. The method for producing a silicon carbide single crystal according to claim 2, A method for producing a silicon carbide single crystal, wherein, when a silicon carbide single crystal has previously been produced under the same conditions using a solidified product of the used solution, the amount of the missing element to be added is calculated based on the composition analysis results obtained in the production of the silicon carbide single crystal carried out under the same conditions.

Images