# US 2022/0034829 A1 · Thermal conductivity estimation method, thermal conductivity estimation apparatus, production method for semiconductor crystal product, thermal conductivity calculator, thermal conductivity calculation program, and, thermal conductivity calculation method > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2022/0034829 A1 - Document ID: US20220034829A1 - Status: Published application - Kind: A1 - Application number: 17/297,080 - Filed: 2019-11-18 - Published: 2022-02-03 - Earliest claimed: 2018-11-28 - Assignees: Sumco Corporation - Applicants: Sumco Corporation - Inventors: Ryusuke YOKOYAMA; Toshiyuki FUJIWARA; Yusuke HIGUCHI; Toru UJIHARA - Web page: https://patentgrep.com/patent/US20220034829A1 ## CPC - G01N 25/18: by investigating thermal conductivity (by calorimetry G01N25/20; by measuring change of resistance of an electrically-heated body G01N27/18) - C30B 15/20: Controlling or regulating (controlling or regulating in general G05) - C30B 19/04: the solvent being a component of the crystal composition - C30B 35/00: Apparatus not otherwise provided for, specially adapted for the growth, production or after-treatment of single crystals or of a homogeneous polycrystalline material with defined structure - G01N 27/18: caused by changes in the thermal conductivity of a surrounding material to be tested (G01N27/20 takes precedence) - G06N 20/00: Machine learning - C30B 29/36: Carbides - G01N: INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES (measuring or testing apparatus or processes other than immunoassay, involving enzymes or microorganisms C12M, C12Q) - C30B: SINGLE-CRYSTAL GROWTH (by using ultra-high pressure, e.g. for the formation of diamonds, B01J3/06); UNIDIRECTIONAL SOLIDIFICATION OF EUTECTIC MATERIAL OR UNIDIRECTIONAL DEMIXING OF EUTECTOID MATERIAL; REFINING BY ZONE-MELTING OF MATERIAL (zone-refining of metals or alloys C22B); PRODUCTION OF A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE (casting of metals, casting of other substances by the same processes or devices B22D; working of plastics B29; modifying the physical structure of metals or alloys C21D, C22F); SINGLE CRYSTALS OR HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE; AFTER-TREATMENT OF SINGLE CRYSTALS OR A HOMOGENEOUS POLYCRYSTALLINE MATERIAL WITH DEFINED STRUCTURE (for producing semiconductor devices or parts thereof H10); APPARATUS THEREFOR - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A thermal conductivity estimation method includes: measuring temperature distribution of a measurement sample surface in a steady state by partially heating the measurement sample under predetermined heating conditions; calculating temperature distribution of a sample model surface by performing a heat-transfer simulation on the sample model of the same shape as the measurement sample for a plurality of combinations of provisional thermal conductivities and heating conditions; making a regression model, whose input is temperature distribution of the measurement sample surface and whose output is a thermal conductivity of the measurement sample, by a machine learning technique using training data in a form of a calculation result of the plurality of combinations and the temperature distribution obtained from the plurality of combinations; and estimating the thermal conductivity of the measurement sample by inputting a measurement result of the temperature distribution of the measurement sample surface into the regression model. ## Claims ### Claim 1 A thermal conductivity estimation method comprising: preparing, as a measurement sample, a component of a production apparatus for a semiconductor crystal product; heating a part of the measurement sample under a predetermined heating condition and measuring a temperature distribution of a surface of the measurement sample under a steady state; performing a heat-transfer simulation for a plurality of combinations of provisional thermal conductivities and heating conditions of a sample model of the same shape as the measurement sample to calculate a temperature distribution of the surface of the sample model for each of the plurality of combinations; making a regression model using a machine learning technique on a basis of training data, the training data being defined by the plurality of combinations used in the heat-transfer simulation and a calculation result of the temperature distribution obtained from the plurality of combinations, an input of the regression model comprising the temperature distribution of the surface of the measurement sample, an output of the regression model consist of the thermal conductivity of the measurement sample; and estimating the thermal conductivity of the measurement sample by inputting a result of the measurement of the temperature distribution of the surface of the measurement sample into the regression model. ### Claim 2 (depends on 1) The thermal conductivity estimation method according to claim 1, wherein in making the regression model through the machine learning technique, the input of the regression model is defined by the temperature distribution of the surface of the measurement sample and a heating condition in measuring the temperature distribution, and in estimating the thermal conductivity, the result of the measurement of the temperature distribution and the heating condition in measuring the temperature distribution are input into the regression model to estimate the thermal conductivity of the measurement sample. ### Claim 3 (depends on 1) The thermal conductivity estimation method according to claim 1, wherein the heat-transfer simulation performed in calculating the temperature distribution of the surface of the sample model assumes the same measurement system as a measurement system used in measuring the temperature distribution. ### Claim 4 (depends on 1) The thermal conductivity estimation method according to claim 1, wherein the heat-transfer simulation performed in calculating the temperature distribution of the surface of the sample model assumes the same atmosphere as an atmosphere in measuring the temperature distribution of the surface of the measurement sample. ### Claim 5 (depends on 1) The thermal conductivity estimation method according to claim 1, wherein the measurement sample is a substitution material of the component. ### Claim 6 A thermal conductivity estimation apparatus comprising: a measurement unit configured to heat a part of a component, which is prepared as a measurement sample, of a production apparatus of a semiconductor crystal product under a predetermined heating condition and measure a temperature distribution of a surface of the measurement sample under a steady state; a calculation unit configured to perform a heat-transfer simulation for a plurality of combinations of provisional thermal conductivities and heating conditions of a sample model of the same shape as the measurement sample to calculate a temperature distribution of the surface of the sample model for each of the plurality of combinations; a machine learning unit configured to make a regression model using a machine learning technique on a basis of training data, the training data being defined by the plurality of combinations used in the heat-transfer simulation and a calculation result of the temperature distribution obtained from the plurality of combinations, an input of the regression model comprising the temperature distribution of the surface of the measurement sample, an output of the regression model comprising the thermal conductivity of the measurement sample; and an estimation unit configured to input a result of the measurement of the temperature distribution of the surface of the measurement sample into the regression model to estimate the thermal conductivity of the measurement sample. ### Claim 7 (depends on 6) The thermal conductivity estimation apparatus according to claim 6, wherein the input of the regression model made by the machine learning unit comprises the temperature distribution of the surface of the measurement sample and a heating condition in measuring the temperature distribution, and the estimation unit is configured to input the result of the measurement of the temperature distribution and the heating condition in measuring the temperature distribution into the regression model to estimate the thermal conductivity of the measurement sample. ### Claim 8 (depends on 6) The thermal conductivity estimation apparatus according to claim 6, wherein the heat-transfer simulation performed by the calculation unit assumes the same measurement system as a measurement system used in measuring the temperature distribution of the surface of the measurement sample. ### Claim 9 (depends on 6) The thermal conductivity estimation apparatus according to claim 6, wherein the heat-transfer simulation performed by the calculation unit assumes the same atmosphere as an atmosphere in measuring the temperature distribution of the surface of the measurement sample. ### Claim 10 (depends on 9) The thermal conductivity estimation apparatus according to claim 9, wherein the measurement unit comprises a measurement case that houses the measurement sample. ### Claim 11 (depends on 10) The thermal conductivity estimation apparatus according to claim 10, wherein the measurement unit comprises a temperature holder configured to keep a temperature of the measurement case at a constant temperature. ### Claim 12 (depends on 10) The thermal conductivity estimation apparatus according to claim 10, wherein the measurement unit comprises an inert gas introduction unit configured to introduce an inert gas into the measurement case. ### Claim 13 (depends on 6) The thermal conductivity estimation apparatus according to claim 6, wherein the measurement unit comprises a heater configured to heat the measurement sample, and a heat-transfer reduction unit configured to restrain a heat of the heater from being transferred to the surface of the measurement sample through an atmosphere. ### Claim 14 (depends on 1) A method of producing a semiconductor crystal product, the method comprising: preparing, as a measurement sample, a component of a production apparatus of the semiconductor crystal product; estimating the thermal conductivity of the component using the thermal conductivity estimation method according to claim 1; performing the heat-transfer simulation in a production step of the semiconductor crystal product using a result of the estimation of the thermal conductivity; and controlling a production apparatus of the semiconductor crystal product on a basis of a result of the heat-transfer simulation in the production step to produce the semiconductor crystal product. ### Claim at position 15 15-32. (canceled) ### Claim 33 (depends on 6) A method of producing a semiconductor crystal product, the method comprising: preparing, as a measurement sample, a component of a production apparatus of the semiconductor crystal product; estimating the thermal conductivity of the component using the thermal conductivity estimation apparatus according to claim 6; performing the heat-transfer simulation in a production step of the semiconductor crystal product using a result of the estimation of the thermal conductivity; and controlling a production apparatus of the semiconductor crystal product on a basis of a result of the heat-transfer simulation in the production step to produce the semiconductor crystal product.