Carbonized metal-coated carbon material
By forming a tantalum carbide coating with controlled chlorine concentration and thickness on the surface of a carbon substrate, the problem of low yield of single-crystal semiconductors was solved, and polymorphism suppression and product life extension were achieved.
Patent Information
- Application Number
- CN202480022488.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-03-30
- Filing Date
- 2024-03-05
- Publication Date
- 2025-10-31
AI Technical Summary
In existing technologies, when using metal carbide-coated carbon materials as crucibles or guiding components, the yield of semiconductor single crystals is affected by problems such as polymorphism and expansion defects, resulting in a low yield.
A metal carbide coating is formed on the surface of a carbon substrate, with the chlorine concentration in the metal carbide coating controlled to be below 25000 ppm·μm, the film thickness above 10 μm and below 100 μm, the surface arithmetic mean roughness Ra above 0.1 μm and below 9.5 μm, and the surface roughness of the carbon substrate controlled to be above 0.1 μm and below 10.0 μm, and the linear thermal expansion coefficient above 3.5×10-6/℃ and below 8.2×10-6/℃, using tantalum carbide as the main metal carbide.
It effectively suppresses polymorphism in semiconductor single crystals, improves the yield of semiconductor single crystals, and extends the product life of components.
Smart Images

Figure CN120882897A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to metal carbide coated carbon materials having a metal carbide coating film on the surface of a carbon substrate. Background Technology
[0002] Carbides such as tantalum carbide, niobium carbide, zirconium carbide, hafnium carbide, and tungsten carbide have high melting points and excellent chemical stability, strength, toughness, and corrosion resistance. Therefore, coating carbon substrates with carbides can improve their heat resistance, chemical stability, strength, toughness, and corrosion resistance. Carbide-coated carbon materials, especially tantalum carbide-coated carbon materials, with a carbide film on the surface of the carbon substrate are used as components in semiconductor single crystal manufacturing equipment such as Si (silicon), SiC (silicon carbide), and GaN (gallium nitride).
[0003] The sublimation recrystallization method (modified Rayleigh method) is widely known as a method for manufacturing bulk SiC single crystals. In the sublimation recrystallization method, a crucible is filled with SiC raw material, and a SiC seed crystal is placed on top of it. Additionally, a cylindrical guiding member is placed around the SiC seed crystal. Sublimation gas generated by heating the SiC raw material rises along the inner wall of the guiding member, allowing the SiC single crystal to grow using the SiC seed crystal.
[0004] Furthermore, SiC single-crystal substrates used in semiconductor devices are manufactured by epitaxially growing SiC single crystals on a SiC substrate formed from bulk single crystals. Known methods for epitaxial growth of SiC single crystals include liquid phase epitaxy (LPE), vapor phase epitaxy (VPE), and chemical vapor deposition (CVD). Typically, CVD is used for epitaxial growth of SiC single crystals. The CVD-based epitaxial growth method involves placing a SiC substrate on a substrate within a device and supplying a raw material gas at a high temperature above 1500°C to grow SiC single crystals.
[0005] In this method of manufacturing SiC single crystals, in order to obtain higher quality crystals, Patent Document 1 discloses a method using a crucible with the inner surface of a graphite substrate coated with tantalum carbide. Furthermore, Patent Document 2 discloses a method using a guide member with the inner wall coated with tantalum carbide.
[0006] Existing technical documents Patent documents Patent Document 1: Japanese Patent Application Publication No. 2019-99453; Patent Document 2: Japanese Patent Application Publication No. 2019-108611. Summary of the Invention
[0007] The problem that the invention aims to solve It is known that the yield of SiC single crystals grown using metal carbide-coated carbon materials as crucibles or guiding components is higher than that grown using uncoated carbon materials. However, even when using metal carbide-coated carbon materials, further improvements in yield are sought. The main reasons for reduced yield include point defects, expansion defects (dislocations, stacking defects), and polymorphism. This invention focuses on suppressing polymorphism to attempt to improve the yield of semiconductor single crystals.
[0008] Therefore, the purpose of this invention is to suppress polymorphism of semiconductor single crystals and improve yield by using metal carbide-coated carbon materials.
[0009] Methods for solving problems Through in-depth research, the inventors discovered that by setting the chlorine concentration in the metal carbide coating of the metal carbide-coated carbon material to below 25000 ppm·μm, polymorphism in semiconductor single crystals can be suppressed, thereby improving the yield of semiconductor single crystals. This led to the completion of this invention. The key points of this invention are as follows.
[0010] [1] A metal carbide coated carbon material, which comprises a carbon substrate mainly composed of carbon and a metal carbide coated film covering at least a portion of the carbon substrate, wherein the metal carbide constituting the metal carbide coated film is at least one metal carbide selected from tantalum carbide, niobium carbide, zirconium carbide, hafnium carbide and tungsten carbide, and the chlorine concentration in the metal carbide coated film is less than 25,000 ppm·μm.
[0011] [2] The metal carbide coated carbon material described in [1] above, wherein the thickness of the metal carbide coated film is 10 μm or more and 100 μm or less.
[0012] [3] The metal carbide coated carbon material described in [1] or [2] above, wherein the surface arithmetic mean roughness Ra of the metal carbide coated film is 0.1 μm or more and 9.5 μm or less.
[0013] [4] The metal carbide coated carbon material described in any one of [1] to [3] above, wherein the surface arithmetic mean roughness Ra of the carbon substrate is 0.1 μm or more and 10.0 μm or less.
[0014] [5] The metal carbide-coated carbon material described in any one of [1] to [4] above, wherein the linear coefficient of thermal expansion of the carbon substrate is 3.5 × 10⁻⁶. -6 / ℃ or above and 8.2×10 -6 / ℃ below.
[0015] [6] The metal carbide coated carbon material described in any one of [1] to [5] above, wherein the metal carbide constituting at least the surface of the metal carbide coated film is tantalum carbide.
[0016] [7] The metal carbide coated carbon material described in any one of [1] to [6] above, wherein the metal carbide constituting the metal carbide coating is tantalum carbide.
[0017] Invention Effects According to the present invention, the generation of polymorphism during crystal growth and epitaxial growth of semiconductor single crystals can be suppressed, thereby improving the yield of semiconductor single crystals. Attached Figure Description
[0018] [ Figure 1 [Illustration] is a schematic diagram of the externally heated depressurized CVD apparatus according to this embodiment.
[0019] [ Figure 2 [Illustration] is a schematic diagram of a reduced-pressure heating furnace for semiconductor single crystal growth according to this embodiment.
[0020] [ Figure 3 [Illustration] is a schematic diagram of the semiconductor single crystal epitaxial growth apparatus according to this embodiment.
[0021] [ Figure 4 [The result is the GDMS analysis result of the tantalum carbide-coated carbon material in Example 1.]
[0022] [ Figure 5 [This is the result of cross-sectional SEM observation of the tantalum carbide-coated carbon material in Example 1.]
[0023] [ Figure 6 [This is the result of SEM observation of the surface of the tantalum carbide-coated carbon material in Example 1.]
[0024] [ Figure 7 [Figure 1] is a diagram used to illustrate the method for determining the adhesion strength of tantalum carbide coatings. Detailed Implementation
[0025] Taking tantalum carbide coated carbon material as an example, the carbon carbide metal coated carbon material of the present invention will be described.
[0026] [Tantalum carbide-coated carbon materials] One embodiment of the present invention provides a tantalum carbide-coated carbon material comprising a carbon substrate with carbon as the main component and a tantalum carbide coating film comprising at least a portion of the coated carbon substrate, wherein the chlorine concentration in the tantalum carbide coating film is below 25,000 ppm·μm.
[0027] (Carbon substrate) In one embodiment of the present invention, the carbon substrate in the tantalum carbide-coated carbon material is a carbon-based substrate. The carbon substrate may further contain chlorine. Examples of materials for the carbon substrate include isotropic graphite, extruded graphite, pyrolytic graphite, and carbon fiber reinforced carbon composites (C / C composites). There are no particular limitations on the shape or properties of the carbon substrate; carbon substrates processed into any shape can be used depending on the application, etc.
[0028] Arithmetic Surface Roughness Ra The arithmetic mean roughness Ra of the carbon substrate surface affects semiconductor single crystal growth and epitaxial growth. A higher arithmetic mean roughness Ra of the carbon substrate tends to result in greater peel strength between the carbon substrate and the tantalum carbide coating; therefore, the arithmetic mean roughness Ra of the carbon substrate is preferably 0.1 μm or higher. On the other hand, if the arithmetic mean roughness Ra of the carbon substrate is too large, the specific surface area of the carbon substrate increases, making it easier for cracks or peeling to occur in the tantalum carbide coating. As a result, when using tantalum carbide-coated carbon material as a component for semiconductor single crystal growth or epitaxial growth, the product life of the component may sometimes be shortened. Therefore, from the viewpoint of the product life of tantalum carbide-coated carbon material, the arithmetic mean roughness Ra of the carbon substrate is preferably 10.0 μm or lower.
[0029] Taking into account the incidence of cracking and peeling of the tantalum carbide coating, the arithmetic mean surface roughness Ra of the carbon substrate is preferably 0.1 μm or more and 10.0 μm or less, more preferably 2.0 μm or more and 6.0 μm or less. By setting the arithmetic mean surface roughness Ra of the carbon substrate within the above range, the peel strength between the carbon substrate and the tantalum carbide coating can be 10 MPa or more, which can extend the product life when using tantalum carbide coated carbon material as a component for semiconductor single crystal growth and epitaxial growth. It should be noted that the arithmetic mean surface roughness Ra of the carbon substrate is a value measured based on JIS B 0633:2001 (ISO4288:1996).
[0030] <Linear thermal expansion coefficient> The linear thermal expansion coefficient of the carbon substrate is preferably 3.5 × 10⁻⁶. -6 / ℃ or above and 8.2×10 -6 Below / ℃. If the linear thermal expansion coefficient of the carbon substrate is 3.5×10 -6 / ℃ or above and 8.2×10 -6 Below a certain temperature, microcracks in the tantalum carbide coating can be further suppressed. From this perspective, a linear thermal expansion coefficient of 5.0 × 10⁻⁶ is more preferable for the carbon substrate. -6 ~7.5×10⁻⁶ / ℃. It should be noted that the linear thermal expansion coefficient of the tantalum carbide coating is approximately 6.3×10⁻⁶. -6 / ℃. The linear thermal expansion coefficient of the carbon substrate can be determined according to JIS R 1618.
[0031] (Tantalum carbide coating) The tantalum carbide coating is primarily composed of tantalum carbide, and the chlorine concentration in the tantalum carbide coating is below 25,000 ppm·μm. It should be noted that the tantalum carbide coating may coat a portion or all of the carbon substrate. Furthermore, in this specification, the chlorine concentration refers to a quality standard.
[0032] <Chlorine concentration> In one embodiment of the present invention, the chlorine concentration in the tantalum carbide coating of the tantalum carbide-coated carbon material is 25,000 ppm·μm or less. If the chlorine concentration in the tantalum carbide coating is greater than 25,000 ppm·μm, polymorphism will occur in the semiconductor single crystal during crystal growth and epitaxial growth, resulting in a lower yield of the semiconductor single crystal. From this viewpoint, the chlorine concentration in the tantalum carbide coating is preferably 10,000 ppm·μm or less, more preferably 5,000 ppm·μm or less, further preferably 4,000 ppm·μm or less, even more preferably 3,000 ppm·μm or less, even more preferably 1,000 ppm·μm or less, even more preferably 300 ppm·μm or less, even more preferably 200 ppm·μm or less, even more preferably 150 ppm·μm or less, even more preferably 100 ppm·μm or less, even more preferably 50 ppm·μm or less, and particularly preferably 30 ppm·μm or less. Furthermore, there is no particular limitation on the lower limit of the chlorine concentration range in the tantalum carbide coating, which is typically above 0.1 ppm·μm or 1 ppm·μm. It should be noted that the chlorine concentration in the tantalum carbide coating can be determined using the method described in the examples below.
[0033] <Mechanism Explanation of the Decreased Polymorph Generation Rate Due to Reduced Chlorine Concentration in Tantalum Carbide Coating> Next, the mechanism by which the polymorph generation rate decreases due to the reduction in chlorine concentration in the tantalum carbide coating will be explained, but this explanation does not limit the present invention in any way.
[0034] Semiconductor single crystal growth, or epitaxial growth, is carried out in a high-temperature environment. The tantalum carbide-coated carbon material used as its component is also exposed to this high-temperature environment. Therefore, chlorine components contained in the tantalum carbide coating are either thermally removed from the tantalum carbide coating alone, or react with impurity atoms contained in the tantalum carbide coating and the carbon substrate to form halides and thus detach from the tantalum carbide coating. In the case of high supersaturation of the feed gas in the single crystal growth environment, chlorine components or halides may adhere to or adsorb onto the facets of the growing semiconductor single crystal. When a crystal nucleus forms from this adhesion or adsorption point, the orientation of the nucleus is randomly selected. Therefore, it is believed that the probability of generating a polymorph with crystal facets different from those grown from the previous step is increased.
[0035] <Method for determining the chlorine concentration in tantalum carbide coatings> The chlorine concentration in the tantalum carbide coating can be determined, for example, by glow discharge mass spectrometry (GDMS) using a glow discharge mass spectrometer (trade name "VG9000") manufactured by VG Elemental under the following measurement conditions.
[0036] (Measurement conditions) • Discharge gas: Ar (7N); • Insulator: ceramic; • Secondary electrode: In Orifice; • Battery: Flat Cell Assembly; • Standardized: 1kV, 1.6mA; • Ion current: Ta ~ 1.2 × E -11 A; • Detector: Faraday cup: 160 milliseconds; • Multiplier tube: 500 milliseconds.
[0037] The measured data are values corrected for the ion strength ratio (IBR) using the relative sensitivity factor (RSF). The RSF used here is the standard value incorporated into the software. RSF varies depending on the measurement conditions or the shape of the discharge battery; the value corrected using the software's built-in RSF (RSF-corrected value) is considered to be closer to the mass concentration than the IBR. Results from Example 1, described later as one example of the measured data, are shown below. Figure 4 The vertical axis represents concentration (ppm by weight), and the horizontal axis represents depth (μm). The depth is calculated based on the measured value of the crater depth after analysis, assuming a constant sputtering velocity during the analysis. It should be noted that because there are irregularities of about several μm at the bottom of the crater after analysis, the depth-direction decomposition is considered to have an error of about 10%.
[0038] The definition of "in the metal carbide coating" refers to the portion of the film thickness that varies in carbon concentration as determined by GDMS analysis, within a range relevant to the measured film thickness obtained through cross-sectional observation using a scanning electron microscope (SEM). For example, in Example 1 described later, cross-sectional observation by SEM showed a film thickness of 30 μm. Figure 4 The measured data show an increase in carbon concentration around 30 μm. Therefore, for example, in Figure 4 In the measurement data shown, the depth from 0 μm to 30 μm, where the carbon concentration increases sharply, is defined as "in the metal carbide coating".
[0039] The calculation method for chlorine concentration is defined as the integral value of the chlorine concentration over the range of "metal carbide coating". For example, in Figure 4 In the measured data shown, the integral value of the chlorine concentration from a depth of 0 μm to a depth of 30 μm is defined as the "chlorine concentration". Figure 4 The measured data shown shows a chlorine concentration of 27.5 ppm·μm.
[0040] <Other Atoms> The tantalum carbide coating is mainly composed of tantalum carbide and contains chlorine at a concentration of less than 25,000 ppm·μm. However, within the range that does not hinder the effect of the present invention, it may contain trace amounts of atoms other than carbon, tantalum, and chlorine. For example, the tantalum carbide coating may contain impurity elements or dopants other than carbon, tantalum, and chlorine at a concentration of less than 10,000 ppm·μm.
[0041] <Manufacturing Method of Tantalum Carbide Coated Carbon Materials> In one embodiment of the present invention, a tantalum carbide-coated carbon material can be fabricated by forming a tantalum carbide layer on the surface of a carbon substrate. The tantalum carbide coating can be formed on the surface of the carbon substrate by methods such as chemical vapor deposition (CVD), sintering, and carbonization. CVD is preferred as it can form a uniform and dense tantalum carbide coating.
[0042] Furthermore, CVD methods include thermal CVD, photochemical CVD, and plasma CVD, among others. Thermal CVD can be used, for example, in the formation of tantalum carbide layers. Thermal CVD offers advantages such as relatively simple equipment configuration and no damage to the carbon substrate caused by plasma. When forming tantalum carbide coatings using thermal CVD, for example, methods such as... Figure 1 The externally heated vacuum CVD apparatus 10 is shown. In the externally heated vacuum CVD apparatus 10, the carbon substrate 14 is supported by a support device 15 in the reaction chamber 12, which includes a heater 13, a raw material supply section 16, an exhaust section 17, etc.
[0043] Reference Figure 1The present invention describes a method for manufacturing a tantalum carbide-coated carbon material according to one embodiment of the present invention.
[0044] First, a carbon substrate 14 is placed in the reaction chamber 12 of the externally heated vacuum CVD apparatus 10. The carbon substrate 14 is supported by a support means 15 having a support portion with three pointed tips.
[0045] Next, the reaction chamber 12 is heated. For example, the reaction chamber 12 is heated under conditions of atmospheric pressure of 10 to 1000 Pa and temperature of 800 to 2200 °C.
[0046] Next, a tantalum carbide coating is formed on the surface of the carbon substrate 14. Gases containing carbon atoms, such as methane (CH4), hydrogen (H2), and tantalum halide gases, such as tantalum pentachloride (TaCl5), are supplied as raw material gases to the reaction chamber 12 from the raw material supply section 16. The tantalum halide gases can be generated, for example, by heating and vaporizing tantalum halide or by reacting metallic tantalum with halogen gases. Then, the raw material gases supplied from the raw material supply section 16 are subjected to a thermal CVD reaction at a temperature of 800–2200°C and a pressure of 1–1000 Pa to form a tantalum carbide coating on the carbon substrate 14.
[0047] <Methods for controlling chlorine concentration in tantalum carbide coatings> When supplying the raw material gas, hydrogen (H2) is supplied to promote the reaction of chlorine being removed from the tantalum halide gas, thereby making the chlorine concentration in the tantalum carbide layer below 25000 ppm·μm.
[0048] In addition, the molar ratio of tantalum atoms in the preferred tantalum halide gas to hydrogen atoms in the preferred hydrogen gas satisfies the following formula.
[0049] Tantalum atom (Ta): Hydrogen atom (H) = 1:x (4 ≤ x < 16) It should be noted that when x is 16 or higher, although the removal of chlorine from the tantalum halide gas is promoted, a tantalum carbide coating containing excess Ta is formed because the molar ratio of Ta₂C or Ta metal is not 1:1. Here, the molar ratio of tantalum atoms in the tantalum halide gas to the hydrogen atoms in the hydrogen gas in the feed gas can be calculated from the volumetric flow rates of the tantalum halide gas and hydrogen gas at their respective standard states. For example, when using tantalum pentachloride as the tantalum halide gas, 1 mole of tantalum halide gas contains 1 mole of tantalum atoms. On the other hand, 1 mole of hydrogen gas contains 2 moles of hydrogen atoms.
[0050] The tantalum carbide-coated carbon material of one embodiment of the present invention described above is an example of the metal carbide-coated carbon material of the present invention. The metal carbide-coated carbon material of the present invention is not limited to the tantalum carbide-coated carbon material of one embodiment. In the metal carbide-coated carbon material of the present invention, the metal carbide constituting the metal carbide coating of the coated carbon substrate is not limited to tantalum carbide. For example, niobium carbide, zirconium carbide, hafnium carbide, and tungsten carbide can be used as the metal carbide constituting the metal carbide coating of the coated carbon substrate. Furthermore, a metal carbide composed of two or more metal carbides selected from tantalum carbide, niobium carbide, zirconium carbide, hafnium carbide, and tungsten carbide can be used as the metal carbide constituting the metal carbide coating of the coated carbon substrate. It should be noted that, among tantalum carbide, niobium carbide, zirconium carbide, hafnium carbide, and tungsten carbide, tantalum carbide is preferred from the perspective of having the highest melting point and excellent chemical stability, strength, and corrosion resistance.
[0051] When tantalum carbide is used in a metal carbide coating, it is not necessary for the entire metal carbide coating to be composed of tantalum carbide, as long as at least the surface of the metal carbide coating is composed of tantalum carbide. Example
[0052] The following examples are given to illustrate the present invention in more detail, but the present invention is not limited thereto.
[0053] The following procedures were followed to prepare the metal carbide-coated carbon materials of Examples 1-21 and Comparative Example 1.
[0054] (Example 1) First, such as Figure 2 and Figure 3 As shown, isotropic graphite is processed into bottomed cylindrical (crucible 21), frustum-shaped (guide member 22), disk-shaped (base 31), and cylindrical (inner wall member 38) shapes, which are used as carbon substrates (see reference). Figure 1 (Symbol 14). The surface arithmetic mean roughness Ra of carbon substrate 14 is 6.0 μm, and the linear thermal expansion coefficient of carbon substrate 14 is 7.0 × 10⁻⁶. -6 / ℃. It should be noted that the linear thermal expansion coefficient of the carbon substrate was measured using a thermomechanical analysis apparatus (TMA7300) from Yamato Scientific Corporation, with values for the thermal expansion coefficient over a temperature range of 200℃ to 1200℃.
[0055] Next, the carbon substrate 14 is placed on Figure 1Inside the reaction chamber 12 of the externally heated reduced-pressure CVD apparatus 10 shown, the carbon substrate 14 is supported by a support device 15 having a support portion with three pointed tips. At this time, the tips of the support portions are in contact with the outer surface of the truncated cone-shaped carbon substrate 14, the outer surface of the bottomed cylindrical carbon substrate 14, the lower surface of the disc-shaped carbon substrate 14, and the outer surface of the cylindrical carbon substrate 14.
[0056] Then, 0.25SLM of methane (CH4) gas, 1.0SLM of argon (Ar) gas as carrier gas, 0.125SLM of hydrogen (H2) gas, and 0.25SLM of tantalum pentachloride (TaCl5) heated to 220°C and vaporized are supplied to the reaction chamber 12 from the raw material supply section 16. Under the conditions of 100Pa and 1250°C, the raw material gas reacts in the reaction chamber 12 to form a tantalum carbide coating on the entire surface of the carbon substrate 14.
[0057] The carbon substrate 14 coated with tantalum carbide film is removed from the reaction chamber 12, completing the crucible 21 made of tantalum carbide-coated carbon material (see reference). Figure 2 ), guide component 22 (refer to) Figure 2 ), base 31 (refer to) Figure 3 ) and inner wall components 38 (refer to) Figure 3 ).
[0058] Based on cross-sectional observations of the tantalum carbide coating using a scanning electron microscope (SEM), the film thickness was calculated to be 30 μm. A cross-sectional SEM image of the tantalum carbide coating in the tantalum carbide-coated carbon material of Example 1 is shown below. Figure 5 and Figure 6 .
[0059] Furthermore, the arithmetic mean roughness Ra of the tantalum carbide coated surface was measured using a surface roughness meter (manufactured by Mitutoyo Co., Ltd., trade name "SurfTest SJ-210"). The result showed that the arithmetic mean roughness Ra of the tantalum carbide coated surface was 5.5 μm.
[0060] Furthermore, the chlorine concentration in the tantalum carbide coating was determined using the glow discharge mass spectrometry (GDMS) method described in this specification. The measurement data are shown below. Figure 4 The results showed that the chlorine concentration in the tantalum carbide coating was 27.5 ppm·μm.
[0061] The adhesion strength of the tantalum carbide coating was measured to be 19.6 MPa. It should be noted that the adhesion strength of the tantalum carbide coating was determined using the following method.
[0062] Regarding the adhesion strength of the tantalum carbide coating, for example, using a film adhesion strength tester (manufactured by Quad Group, trade name "Romulus"), such as... Figure 7 As shown, tantalum carbide coating 41 and pin 46 are bonded together with adhesive 45. Pin 46 is pulled by pressing with a pressing clamp 44, and the stress during peeling of tantalum carbide coating 41 is measured. Five measurements are performed, and the average value of the five measurements is taken as the adhesion strength of tantalum carbide coating.
[0063] In such Figure 2 The vacuum furnace 20 shown contains a fabricated crucible 21 and a guiding component 22, and SiC single crystals are grown using sublimation recrystallization. SiC raw material 25 is placed in the crucible 21, and a 2-inch diameter SiC seed crystal 24 is placed on top of it. Argon gas flows into the vacuum furnace 20 at a flow rate of 10–30 SLM, and under conditions of 500–1000 Pa pressure and 2000–2500 °C temperature, the SiC raw material 25 sublimates, growing a 5 mm thick SiC single crystal on the SiC seed crystal 24.
[0064] SiC single crystals were repeatedly fabricated to determine the number of times crucible 21 and guide component 22 could be reused. As a result, after 24 uses, peeling of the tantalum carbide coating was observed in crucible 21, necessitating replacement with a new component.
[0065] In such Figure 3 The CVD apparatus 30 shown includes a fabricated base 31 and an inner wall component 38, and SiC single crystals are epitaxially grown using the CVD method. A SiC single crystal substrate 34, formed from a bulk single crystal, is placed on the base 31. Silane (SiH4) gas is introduced into the CVD apparatus at a rate of 30 sccm, and propane (C3H8) gas at a rate of 70 sccm. The pressure is set to 45 Torr, and the temperature to 1550°C, allowing the SiC single crystal to grow epitaxially on the SiC single crystal substrate.
[0066] By repeatedly manufacturing SiC single crystals, the reusability of the base 31 and inner wall component 38 was confirmed. As a result, after 96 uses, the peeling of the tantalum carbide coating was confirmed, necessitating the replacement of the component with a new one.
[0067] Furthermore, when studying the polymorphism rate of the obtained single crystals, the polymorphism rate of SiC single crystals prepared by sublimation recrystallization and SiC single crystals prepared by epitaxial growth were both 0%. It should be noted that the polymorphism rate was determined as follows.
[0068] The appearance of the grown SiC single crystals was studied by irradiating them with ultraviolet (UV) light. Even when a few crystal disturbances were observed, it was considered that polymorphism had occurred.
[0069] (Example 2) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0.2 SLM, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0070] (Example 3) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0.15 SLM and the tantalum pentachloride (TaCl5) flow rate from 0.25 SLM to 0.3 SLM, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0071] (Example 4) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0.2 SLM and the tantalum pentachloride (TaCl5) flow rate from 0.25 SLM to 0.4 SLM, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0072] (Example 5) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0.25 SLM and the tantalum pentachloride (TaCl5) flow rate from 0.25 SLM to 0.5 SLM, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0073] (Example 6) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0.15 SLM, the tantalum pentachloride (TaCl5) flow rate from 0.25 SLM to 0.3 SLM, and shortening the film deposition time to reduce the film thickness from 30 μm to 10 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated using the same method as in Example 1, and their evaluation was conducted. The results are shown in Table 2.
[0074] (Example 7) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0.15 SLM and extending the film formation time to increase the film thickness from 30 μm to 50 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0075] (Example 8) Except for changing the surface arithmetic mean roughness Ra of the carbon substrate 14 from 6.0 μm to 0.1 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0076] (Example 9) Except for changing the surface arithmetic mean roughness Ra of the carbon substrate 14 from 6.0 μm to 10 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0077] (Example 10) In addition to reducing the thermal expansion coefficient of carbon substrate 14 from 7.0×10 -6 / ℃ changed to 3.5×10 -6 Except for / ℃, the crucible 12, guide member 19, base 31 and inner wall member 38 were manufactured in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0078] (Example 11) In addition to reducing the thermal expansion coefficient of carbon substrate 14 from 7.0×10 -6 / ℃ changed to 8.2×10 -6 Except for / ℃, the crucible 12, guide member 19, base 31 and inner wall member 38 were manufactured in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0079] (Example 12) Except for keeping the flow rate of the metal chloride gas (0.25 SLM) unchanged, and changing the metal chloride gas from tantalum pentachloride (TaCl5) to niobium pentachloride (NbCl5), the crucible 12, guide member 19, base 31, and inner wall member 38 were manufactured and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0080] (Example 13) Except for keeping the flow rate of the metal chloride gas (0.25 SLM) unchanged, and changing the metal chloride gas from tantalum pentachloride (TaCl5) to hafnium tetrachloride (HfCl4), the crucible 12, guide member 19, base 31, and inner wall member 38 were manufactured and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0081] (Example 14) Except for keeping the flow rate of the metal chloride gas (0.25 SLM) unchanged, and changing the metal chloride gas from tantalum pentachloride (TaCl5) to zirconium tetrachloride (ZrCl4), the crucible 12, guide member 19, base 31, and inner wall member 38 were manufactured and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0082] (Example 15) Except for keeping the flow rate of the metal chloride gas (0.25 SLM) unchanged, and changing the metal chloride gas from tantalum pentachloride (TaCl5) to tungsten pentachloride (WCl5), the crucible 12, guide member 19, base 31, and inner wall member 38 were manufactured and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0083] (Example 16) Except for keeping the flow rate of the metal chloride gas (0.25 SLM) unchanged, and changing the metal chloride gas from tantalum pentachloride (TaCl5) to a mixture of tantalum pentachloride (TaCl5) and niobium pentachloride (NbCl5) (TaCl5:NbCl5=100:1), the crucible 12, guide member 19, base 31, and inner wall member 38 were manufactured in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0084] (Example 17) Except for keeping the flow rate of the metal chloride gas (0.25 SLM) unchanged, and changing the metal chloride gas from tantalum pentachloride (TaCl5) to a mixture of tantalum pentachloride (TaCl5) and tungsten pentachloride (WCl5) (TaCl5:WCl5=100:1), the crucible 12, guide member 19, base 31, and inner wall member 38 were manufactured and evaluated in the same manner as in Example 1. The results are shown in Table 2.
[0085] (Example 18) Except for shortening the film formation time to reduce the film thickness from 30 μm to 2 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated using the same method as in Example 1, and their evaluation was conducted. The results are shown in Table 2.
[0086] (Example 19) Except for changing the surface arithmetic mean roughness Ra of the carbon substrate 14 from 6.0 μm to 32 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0087] (Example 20) Except for changing the surface arithmetic mean roughness Ra of the carbon substrate 14 from 6.0 μm to 0.05 μm, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0088] (Example 21) In addition to reducing the thermal expansion coefficient of carbon substrate 14 from 7.0×10 -6 / ℃ changed to 1.9×10 -6 Except for / ℃, the crucible 12, guide member 19, base 31 and inner wall member 38 were manufactured in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 1.
[0089] (Comparative Example 1) Except for changing the hydrogen (H2) flow rate from 0.125 SLM to 0 SLM and the tantalum pentachloride (TaCl5) flow rate from 0.25 SLM to 0.5 SLM, the crucible 12, guide member 19, base 31, and inner wall member 38 were fabricated in the same manner as in Example 1, and their evaluation was carried out. The results are shown in Table 2.
[0090] The manufacturing conditions for Examples 1 to 21 and Comparative Example 1 are shown in Table 1.
[0091] [Table 1] The evaluation results of Examples 1-21 and Comparative Example 1 are shown in Table 2.
[0092] [Table 2] Comparing the results of Examples 1 to 21 with the results of Comparative Example 1, it can be seen that by setting the chlorine concentration in the metal carbide coating to below 25000 ppm·μm, the crystal growth of semiconductor single crystals and the occurrence of polymorphism during epitaxial growth can be suppressed.
[0093] Comparing the results of Examples 1 to 5 with those of Comparative Example 1, it was found that when the chlorine concentration in the tantalum carbide coating is below 25000 ppm·μm, the product life of the metal carbide-coated carbon material is long, and polymorphism is suppressed during crystal growth and epitaxial growth of semiconductor single crystals. It should be noted that regarding the reuse of the tantalum carbide-coated carbon material in Comparative Example 1, due to polymorphism, its use was discontinued before it became unusable due to peeling of the tantalum carbide coating.
[0094] Comparing the results of Examples 1, 6, and 7 with those of Example 18, it was found that when the thickness of the tantalum carbide coating is 10 μm or more and 100 μm or less, the product life of the metal carbide-coated carbon material is long, and polymorphism is suppressed during crystal growth and epitaxial growth of semiconductor single crystals. When the thickness of the tantalum carbide coating is 10 μm or less, polymorphism was confirmed during crystal growth and epitaxial growth of single-crystal semiconductors. It should be noted that regarding the reuse of the tantalum carbide-coated carbon material in Example 18, due to polymorphism, its use was discontinued before it became unusable due to peeling of the tantalum carbide coating. Furthermore, when the thickness of the tantalum carbide coating exceeds 100 μm, the film formation time and cost increase, which is not preferred. Therefore, the thickness of the tantalum carbide coating is preferably 10 μm or more and 100 μm or less.
[0095] Comparing the results of Examples 1, 8, and 9 with those of Examples 19 and 20, it was found that when the surface arithmetic mean roughness Ra of the metal carbide coating is 0.1 μm or more and 9.5 μm or less, or the surface arithmetic mean roughness Ra of the carbon substrate is 0.1 μm or more and 10.0 μm or less, the product life of the metal carbide-coated carbon material is long, and polymorphism is suppressed during the crystal growth and epitaxial growth of semiconductor single crystals. When the surface arithmetic mean roughness Ra of the carbon substrate is greater than 10 μm, the product life of the metal carbide-coated carbon material decreases, and polymorphism is confirmed during the crystal growth and epitaxial growth of semiconductor single crystals. When the surface arithmetic mean roughness Ra of the carbon substrate is less than 0.1 μm, the product life of the metal carbide-coated carbon material decreases.
[0096] Comparing the results of Examples 1, 10, and 11 with the results of Example 21, it was found that the coefficient of thermal expansion of the carbon substrate is 3.5 × 10⁻⁶. -6 / ℃ or above and 8.2×10 -6 At temperatures below a certain temperature, metal carbide-coated carbon materials exhibit long product lifespans and suppress polymorphism during the crystal growth and epitaxial growth of semiconductor single crystals. The coefficient of thermal expansion of the carbon substrate is less than 3.5 × 10⁻⁶. -6 At a temperature of / ℃, the product life of metal carbide-coated carbon materials decreases, and polymorphism has been confirmed during the crystal growth and epitaxial growth of semiconductor single crystals.
[0097] Comparing the results of Examples 1 and 12-16, it was found that when at least a portion of the surface of a carbon-based substrate is coated with a metal carbide coating film composed of any one or more of tantalum carbide, niobium carbide, zirconium carbide, hafnium carbide, and tungsten carbide, and the chlorine concentration in the metal carbide coating film is below 25,000 ppm·μm, the metal carbide coated carbon material has a long product life and suppresses the occurrence of polymorphism during the crystal growth and epitaxial growth of semiconductor single crystals.
[0098] Symbol Explanation 10: Externally heated reduced-pressure CVD unit; 11: Crown chamber; 12: Reaction chamber; 13: Heater; 14: Carbon substrate; 15: Support device; 16: Raw Material Supply Department; 17: Exhaust section; 20: Pressure-reducing heating furnace; 21: Crucible; 22: Guiding components; 23: Top cover; 24: SiC seed crystal; 25: SiC raw materials; 30: Epitaxial growth apparatus; 31: Base; 32: Reaction chamber; 33: Heater; 34: SiC single crystal substrate; 35: Exhaust section; 36: Raw Material Supply Department; 37: Top chamber; 38: Inner wall components; 41: Tantalum carbide coating; 42: Carbon substrate; 44: Pressing clamp; 45: Adhesive; 46: Sales.
Claims
1. A metal carbide-coated carbon material, comprising a carbon substrate mainly composed of carbon and a metal carbide coating film covering at least a portion of the carbon substrate. The metal carbide constituting the aforementioned metal carbide coating is at least one metal carbide selected from tantalum carbide, niobium carbide, zirconium carbide, hafnium carbide, and tungsten carbide. The chlorine concentration in the aforementioned metal carbide coating is below 25,000 ppm·μm.
2. The metal carbide-coated carbon material according to claim 1, wherein, The thickness of the aforementioned metal carbide coating is 10 μm or more and 100 μm or less.
3. The metal carbide-coated carbon material according to claim 1, wherein, The surface arithmetic mean roughness Ra of the aforementioned metal carbide coating is greater than 0.1 μm and less than 9.5 μm.
4. The metal carbide-coated carbon material according to claim 1, wherein, The surface arithmetic mean roughness Ra of the aforementioned carbon substrate is greater than 0.1 μm and less than 10.0 μm.
5. The metal carbide-coated carbon material according to claim 1, wherein, The linear thermal expansion coefficient of the aforementioned carbon substrate is 3.5 × 10⁻⁶. -6 / ℃ or above and 8.2×10 -6 / ℃ below.
6. The metal carbide-coated carbon material according to claim 1, wherein, The metal carbide constituting at least the surface of the aforementioned metal carbide coating is tantalum carbide.
7. The metal carbide-coated carbon material according to claim 1, wherein, The metal carbide constituting the aforementioned metal carbide coating is tantalum carbide.
Citation Information
Patent Citations
Tantalum carbide coating carbon material and method of manufacturing the same, and member for semiconductor single-crystal manufacturing apparatus
JP2019099453A
METHOD OF MANUFACTURING TaC COATING LAYER USING CVD AND PROPERTY OF TaC MANUFACTURED USING THE SAME
JP2019108611A