Graphite crucible and metal sample oxygen concentration measuring method using the same

The graphite crucible design with a specified height difference and replacement protocol addresses the issue of increased measurement errors in silicon samples by preventing abnormal heat generation and maintaining accuracy in oxygen concentration analysis.

JP2024165123AActive Publication Date: 2024-11-28SUMCO CORP
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
JP2023081006
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-16
Publication Date
2024-11-28
Estimated Expiration
2043-05-16

AI Technical Summary

Technical Problem

The use of a double-structured graphite crucible for measuring oxygen concentration in silicon samples by the GFA method leads to increased measurement errors due to the deterioration of the outer crucible's resistivity with repeated use, causing abnormal heat generation and unintended sample dissolution.

Method used

A graphite crucible design with a height difference of 1.1 mm or less between the upper ends of the inner and outer crucibles, along with a protocol of replacing the inner crucible for each sample and reusing the outer crucible up to a predetermined number of measurements, to prevent abnormal heat generation and maintain measurement accuracy.

Benefits of technology

The solution effectively prevents melt leakage and reduces measurement errors by suppressing resistivity increases in the outer crucible, ensuring accurate oxygen concentration measurements in silicon samples.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024165123000001_ABST
    Figure 2024165123000001_ABST
Patent Text Reader

Abstract

To provide a graphite crucible capable of suppressing increases in measurement error in oxygen concentration even when an outer crucible of a dual structure crucible is used repeatedly in measurement of oxygen concentration in a metal sample using a GFA method.SOLUTION: A graphite crucible 10 includes an inner crucible 11 for housing a specimen 2 and an outer crucible 12 for housing the inner crucible 11. A height difference Δh=h2-h1 between an upper end part 12e of the outer crucible 12 and an upper end part 11e of the inner crucible 11 when the inner crucible 11 is housed within the outer crucible 12 is 1.1 mm or less.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a graphite crucible and a method for measuring the oxygen concentration of a metal sample using the same, and more particularly to a method for measuring the oxygen concentration of a silicon sample using a GFA (Gas Fusion Analyzer). [Background technology]

[0002] FT-IR (Fourier Transform Infrared spectroscopy analysis) is widely used as a method for evaluating the oxygen concentration in silicon crystals, and is standardized by ASTM (American Society for Testing and Materials). This method measures the amount of absorption by interstitial oxygen in silicon crystals when infrared light is transmitted through a wafer, and is based on the principle of quantifying the oxygen in the crystal from the amount of absorption. Since the amount of light absorption is very sensitive to the interstitial oxygen concentration in silicon crystals, it allows for highly sensitive and reliable evaluation.

[0003] However, evaluation by FT-IR is impossible unless the sample transmits infrared light to a certain extent.For example, in the case of silicon single crystals with low resistivity (for example, 20 mΩ·cm or less, especially 10 mΩ·cm or less) doped with a high concentration of boron, which is preferably used as a substrate material for epitaxial wafers, or silicon single crystals with low resistivity doped with a high concentration of phosphorus, arsenic, or antimony, which is preferably used as a substrate material for discrete devices, the light is absorbed by the free electrons contained in large quantities in the sample, and they do not transmit any infrared light at all, so FT-IR cannot be used.

[0004] GFA has been attracting attention as a method for evaluating the oxygen concentration in silicon crystals, which is difficult to evaluate using FT-IR. GFA is based on the principle of melting the sample at high temperature to gasify the oxygen contained in the sample as CO, and then quantifying the oxygen contained in the original sample by chemically analyzing the gas. The GFA method is a completely destructive method that requires melting the sample, but has the advantages of relatively inexpensive measuring equipment, not requiring much skill to perform the measurement, and being able to perform the evaluation in a short time.

[0005] Regarding a method for analyzing the oxygen concentration of a sample using a GFA, Patent Documents 1 to 3 describe a method in which a sample in a graphite crucible is preheated at a temperature lower than the melting point to remove oxygen and other substances adhering to the surface, and then the sample is heated at a temperature equal to or higher than the melting point to perform the main analysis. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2003-188107 A [Patent Document 2] JP 2004-053577 A [Patent Document 3] JP 2004-219364 A Summary of the Invention [Problem to be solved by the invention]

[0007] When electrodes are connected to a graphite crucible used for melting a sample at high temperatures and the graphite crucible is directly heated by a resistance heating method, stress is applied to the electrode connection position, which may cause the graphite crucible to crack and the molten sample to spill out. A method of using a double-structure graphite crucible is available as a method to prevent such accidents. When using a double-structure crucible, even if the outer crucible to which the electrodes are connected cracks, the molten material remains in the inner crucible, preventing the molten material from spilling out. In addition, when measuring a new sample, only the inner crucible can be replaced and the outer crucible can be reused, reducing the cost of replacing the graphite crucible.

[0008] However, when the oxygen concentration of a silicon sample is measured by the GFA method using a double-structure crucible, there is a problem in that the measurement error of the oxygen concentration of the sample increases as the number of measurements of the sample increases.

[0009] Therefore, an object of the present invention is to provide a graphite crucible that can prevent the molten material in the crucible from leaking out and suppress an increase in oxygen concentration measurement error that occurs due to an increase in the number of sample measurements, and a method for measuring the oxygen concentration of a metal sample using the same. [Means for solving the problem]

[0010] As a result of extensive research, the inventors of the present application have found that the cause of the increase in the measurement error of the oxygen concentration is that the upper end of the outer crucible of a double-structure graphite crucible gradually changes in quality and increases in resistivity due to repeated use of the outer crucible, and that the above problem can be solved by suppressing the increase in resistivity. If the resistivity of the graphite crucible becomes high, the graphite crucible will abnormally heat up during the sample pretreatment process, causing unintended melting of the sample, which will reduce the amount of oxygen supplied from the sample in this analysis. If the abnormal heating of the graphite crucible can be suppressed, oxygen in the sample will not evaporate during the pretreatment process, making it possible to accurately measure the oxygen concentration in the sample in this analysis.

[0011] The present invention is based on such technical knowledge, and the graphite crucible of the present invention is a graphite crucible used for measuring the oxygen concentration of a metal sample by the GFA method, and is characterized in that it comprises an inner crucible for accommodating the metal sample and an outer crucible for accommodating the inner crucible, and the difference in height between the upper end of the outer crucible and the upper end of the inner crucible when the inner crucible is accommodated within the outer crucible (the amount by which the upper end of the crucible protrudes from the upper end of the inner crucible) is 1.1 mm or less.

[0012] The present invention also provides a method for measuring the oxygen concentration of a metal sample by the GFA method, comprising a pretreatment step of heating a metal sample contained in a graphite crucible at a temperature lower than the melting point to remove a surface oxide film, and a main analysis step of heating the metal sample from which the surface oxide film has been removed at a temperature equal to or higher than the melting point to generate CO gas derived from oxygen in the metal sample, and quantitatively analyzing the CO gas using an IR (Infrared) detector, wherein the graphite crucible comprises an inner crucible for containing the metal sample and an outer crucible for containing the inner crucible, and the difference in height between an upper end of the outer crucible and an upper end of the inner crucible when the inner crucible is contained within the outer crucible is 1.1 mm or less.

[0013] According to the present invention, even when the outer crucible of a double-structure crucible is repeatedly used, the occurrence of abnormal heating of the outer crucible during pretreatment can be suppressed, and therefore an increase in measurement error of the oxygen concentration of the sample in the main analysis can be suppressed.

[0014] In the present invention, the height difference is preferably 0 to 0.6 mm, which can further reduce the probability of abnormal heating of the outer crucible during pretreatment.

[0015] In the present invention, it is preferable that the inner crucible is replaced every time a new sample is measured, and the outer crucible is not replaced every time a new sample is measured, but is used repeatedly up to a predetermined number of measurements. In this case, it is preferable that the predetermined number of measurements is 50 to 100. This makes it possible to prevent the melt from leaking due to the cracking of the crucible, and to suppress an increase in measurement error of the oxygen concentration caused by an increase in the number of measurements of the sample.

[0016] In the present invention, the metal sample is preferably a silicon piece cut from a silicon wafer and preferably comprises a silicon single crystal doped with a p-type dopant and having a resistivity of 0.2 Ω cm or less, or a silicon single crystal doped with an n-type dopant and having a resistivity of 0.1 Ω cm or less. According to the present invention, it is possible to measure the oxygen concentration in a silicon wafer doped with a high concentration of dopant and having a low resistivity, which is difficult to evaluate by FT-IR.

[0017] In the present invention, it is preferable that an upper electrode and a lower electrode are connected to the upper end and the lower end of the outer crucible, and the graphite crucible is heated by applying high frequency power to the graphite crucible via the upper electrode and the lower electrode, and the upper electrode is not in contact with the inner crucible, whereby the graphite crucible can be directly heated to heat the contents, and the outflow of the molten material due to cracking of the inner crucible can be prevented. Effect of the Invention

[0018] According to the present invention, it is possible to provide a graphite crucible that can prevent the molten material in the crucible from leaking out and suppress an increase in oxygen concentration measurement error that occurs due to an increase in the number of sample measurements, and a method for measuring the oxygen concentration of a metal sample using the same. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a schematic cross-sectional view showing the structure of a graphite crucible according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic exploded view of the graphite crucible shown in FIG. 1, showing a state in which the inner crucible is removed from the outer crucible. [Diagram 3] FIG. 3 is an explanatory diagram of a method for measuring the oxygen concentration of a silicon sample by the GFA method using the above-mentioned graphite crucible. [Figure 4] FIG. 4 is a graph showing the results of using the graphite crucibles according to the comparative example and the example, illustrating a correlation between the number of times the outer crucible was used and the power consumption of the graphite crucible in the pretreatment step. [Diagram 5] FIG. 5 is a graph showing the correlation between the number of times the outer crucible is used and the power consumption of the graphite crucible in the pretreatment process when graphite crucibles with height differences between the outer crucible and the inner crucible of 0.6 mm, 1.1 mm, and 1.2 mm are used. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0021] Fig. 1 is a schematic cross-sectional view showing the configuration of a graphite crucible according to an embodiment of the present invention, and Fig. 2 is a schematic exploded view of the graphite crucible shown in Fig. 1, showing a state in which an inner crucible is removed from an outer crucible.

[0022] As shown in Fig. 1, the graphite crucible 10 is a so-called double crucible, and is composed of an inner crucible 11 that contains a sample, and an outer crucible 12 that contains the inner crucible 11. As shown in Fig. 2, the inner crucible 11 can be removed from the outer crucible 12, and is replaced with a new one every time a new sample is measured. On the other hand, the outer crucible 12 is not replaced every time a new sample is measured, and is repeatedly used up to a predetermined number of times (for example, 50 to 100 times).

[0023] The reason for using a double-structure crucible in the GFA method is to prevent the melt from leaking out if the crucible cracks. When the graphite crucible is heated directly by resistance heating, cracks are likely to occur at the electrode connection positions of the graphite crucible due to differences in thermal expansion coefficients, and the melt inside the crucible may leak out. However, in the case of a double crucible, even if the outer crucible 12 cracks, the melt is contained in the inner crucible 11, so leakage of the melt can be prevented.

[0024] The inner crucible 11 has a cylindrical body portion 11a and a bottom portion 11b provided below the body portion 11a. In this embodiment, the outer bottom surface of the bottom portion 11b is flat, but the inner bottom surface of the inner crucible 11 is rounded.

[0025] The outer crucible 12 has a cylindrical body portion 12a and a bottom portion 12b provided below the body portion 12a. In order to accommodate the inner crucible 11, the inner diameter R2 of the body portion 12a of the outer crucible 12 is slightly larger than the outer diameter R1 of the body portion 11a of the inner crucible 11. The inner bottom surface of the outer crucible 12 is rounded, but a step 12c is provided at the inner corner portion of the outer crucible 12, and the inner crucible 11 in the outer crucible 12 is supported by the step 12c at the inner corner portion of the outer crucible 12.

[0026] In a state where the inner crucible 11 is accommodated in the outer crucible 12 (see FIG. 1), the upper end 12e of the outer crucible 12 is located above the upper end 11e of the inner crucible 11, but the difference Δh=h2-h1 between the height h2 of the upper end 12e of the outer crucible 12 and the height h1 of the upper end 11e of the inner crucible 11 is 1.1 mm or less (Δh≦1.1 mm), preferably 0 to 0.6 mm (0 mm≦Δh≦0.6 mm). If the height h2 of the upper end 12e of the outer crucible 12 is too high compared to the upper end 11e of the inner crucible 11, the exposed area of ​​the inner circumferential surface of the outer crucible 12 becomes large, and as the number of times the outer crucible 12 is reused increases, the impregnation of the upper end 12e of the outer crucible 12 with silicon (Si) progresses, and the resistivity of the upper end 12e becomes high. This makes it easier for abnormal heat to occur at the upper end of the graphite crucible 10, and unintended melting of the silicon sample occurs even in the pretreatment process. However, if the difference Δh between the height h2 of the upper end 12e of the outer crucible 12 and the height h1 of the upper end 11e of the inner crucible 11 is 1.1 mm or less (Δh≦1.1 mm), an increase in the resistivity of the upper end of the outer crucible 12 can be suppressed. Therefore, abnormal heat generation at the upper end 12e of the outer crucible 12 can be prevented, and deterioration of the measurement accuracy of the oxygen concentration of the silicon sample can be prevented.

[0027] FIG. 3 is an explanatory diagram of a method for measuring the oxygen concentration of a silicon sample by the GFA method using the graphite crucible 10.

[0028] As shown in FIG. 3(a), an upper electrode 21 and a lower electrode 22 are connected to the upper and lower ends of the outer crucible 12 of the graphite crucible 10, respectively, and a silicon sample 1 is set in the inner crucible 11. In the first sample measurement, new inner crucibles 11 and outer crucibles 12 are used. It is preferable that the upper electrode 21 is connected only to the upper end of the outer crucible 12 and is not in contact with the upper end of the inner crucible 11. If the upper electrode 21 is in contact with the inner crucible 11, stress is applied to the inner crucible 11, which may cause the inner crucible 11 to crack. However, if the upper electrode 21 is connected only to the outer crucible 12, such a problem can be avoided.

[0029] The silicon sample 1 is a silicon piece cut from a silicon wafer, and is preferably made of silicon single crystal doped with a p-type dopant such as boron, gallium, or indium and having a resistivity of 0.2 Ω cm or less, or silicon single crystal doped with an n-type dopant such as phosphorus, arsenic, or antimony and having a resistivity of 0.1 Ω cm or less. It is difficult to measure the oxygen concentration of such silicon samples by the FT-IR method, and measurement by the GFA method is suitable.

[0030] 3(b), a high-frequency power source 23 is connected between the upper electrode 21 and the lower electrode 22, and a high-frequency power of 1850 to 2200 W is applied to directly heat the graphite crucible 10. At this time, a pretreatment step is performed in which power is supplied to achieve a temperature lower than the melting point of silicon, thereby removing the surface oxide film 2 of the silicon sample 1.

[0031] Next, as shown in Fig. 3(c), high-frequency power of 3500 to 3800 W is applied to melt the silicon sample 1 to a temperature above the melting point of silicon. At this time, carbon (C) supplied from the graphite crucible 10 reacts with oxygen (O) contained in the silicon sample 1 to generate CO gas (or CO2 gas). As shown in Fig. 3(d), in the GFA method, the CO gas generated in the pretreatment in Fig. 3(b) and the main analysis in Fig. 3(c) is converted to absorbance by an IR detector, and the oxygen concentration is calculated from the absorbance area of ​​the main analysis.

[0032] As shown in Fig. 3(e), in the second sample measurement, the inner crucible 11 of the graphite crucible 10 is replaced with a new one, but the outer crucible 12 is reused. As described above, the inner crucible 11 is replaced every time a new sample is measured, but the outer crucible 12 is used repeatedly until it has been used a predetermined number of times. The silicon sample 1 is set in the inner crucible 11, and an upper electrode 21 and a lower electrode 22 are connected to the upper and lower ends of the outer crucible 12, respectively.

[0033] Next, as shown in Fig. 3(f), a pretreatment process is performed to remove the surface oxide film 2 of the silicon sample 1, and then the main analysis process of the silicon sample 1 is performed as shown in Fig. 3(g). As shown in Fig. 3(h), in the GFA method, the CO gas generated in the pretreatment process in Fig. 3(f) and the main analysis in Fig. 3(g) is converted to absorbance by an IR detector, and the oxygen concentration is calculated from the absorbance area in the main analysis.

[0034] When the oxygen concentration is measured by the GFA method using a conventional graphite crucible 10 in which the upper end of the outer crucible protrudes more than the inner crucible, the chemical reaction (conversion to SiC) of the upper end 12e of the outer crucible 12 progresses in the pretreatment of Fig. 3(f) and the main analysis of Fig. 3(g), so that the resistivity of the upper end 12e increases significantly. Therefore, by repeatedly using the outer crucible 12, abnormal heat generation occurs in the graphite crucible 10 in the pretreatment of Fig. 3(f), and the measurement error of the oxygen concentration in the main analysis increases due to unintended melting of the sample in the pretreatment.

[0035] However, as described above, when the oxygen concentration is measured by the GFA method using the graphite crucible 10 according to this embodiment, the chemical reaction (conversion to SiC) of the upper end 12e of the outer crucible 12 is suppressed in the pretreatment of FIG. 3(f) and the main analysis of FIG. 3(g), so that the increase in resistivity of the upper end 12e is suppressed. Therefore, even if the outer crucible 12 is used repeatedly, abnormal heat generation of the graphite crucible 10 does not occur in the pretreatment of FIG. 3(f), and unintended melting of the sample can be prevented. Therefore, the sample can be reliably melted in the main analysis, and the measurement accuracy of the oxygen concentration in the silicon sample can be improved.

[0036] The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention, and it goes without saying that these modifications are also included in the scope of the present invention.

[0037] For example, in the above embodiment, the double-structure graphite crucible is used to quantitatively analyze oxygen in a silicon sample, but the graphite crucible may be used to quantitatively analyze oxygen in other metal samples such as iron, nickel, titanium, etc. Also, the graphite crucible may be used to quantitatively analyze impurity elements other than oxygen, such as nitrogen. EXAMPLES

[0038] (Comparative Examples 1 to 3) The oxygen concentration of silicon samples was repeatedly measured using the graphite crucibles according to Comparative Examples 1 to 3, in which the difference in height between the top end of the inner crucible and the top end of the outer crucible when the inner crucible was set in the outer crucible was 1.4 mm. As described above, the outer crucible was used repeatedly, and the inner crucible was replaced each time a new sample was measured. In the oxygen concentration measurement of silicon samples, the samples were pretreated to remove the surface oxide film before the main analysis. During the sample pretreatment, the power consumption of the graphite crucible and the presence or absence of abnormal elution of the sample were evaluated.

[0039] (Examples 1 and 2) The oxygen concentration of silicon samples was repeatedly measured using the graphite crucibles according to Examples 1 and 2, in which the difference in height between the top end of the inner crucible and the top end of the outer crucible when the inner crucible was set in the outer crucible was 0.6 mm. As described above, the outer crucible was used repeatedly, and the inner crucible was replaced each time a new sample was measured. In the measurement of the oxygen concentration of silicon samples, the samples were pretreated to remove the surface oxide film before the main analysis. During the sample pretreatment, the power consumption of the graphite crucible and the presence or absence of abnormal elution of the samples were evaluated.

[0040] FIG. 4 is a graph showing the results of using the graphite crucibles according to Comparative Examples 1 to 3 and Examples 1 and 2, illustrating a correlation equation between the number of times the outer crucible was used and the power consumption of the graphite crucible in the pretreatment step.

[0041] As shown in Fig. 4, when the graphite crucibles according to Comparative Examples 1 to 3 were used, the power consumption of the graphite crucible increased as the number of times the sample was analyzed increased. In particular, in Comparative Example 1, the regression line showing the relationship between the number of times the outer crucible was used x and the power consumption y of the graphite crucible in the pretreatment step was y = 0.27x + 1798.1, and the coefficient of determination was R 2 In Comparative Example 2, the regression line showing the relationship between the number of times the outer crucible was used x and the power consumption y of the graphite crucible in the pretreatment step was y = 0.156x + 1798.2, and the coefficient of determination was R 2 In Comparative Example 3, the regression line showing the relationship between the number of times the outer crucible was used x and the power consumption y of the graphite crucible in the pretreatment step was y = 0.4079x + 1790.9, and the coefficient of determination was R 2 = 0.8878. The reason why the power consumption of the graphite crucible in the pretreatment process increased with the number of sample analyses is thought to be because silicon impregnated the upper end of the outer crucible, causing the resistance of the outer crucible to increase. In addition, when the number of sample analyses exceeded 30, an abnormality in elution occurred in the pretreatment process before the main analysis.

[0042] On the other hand, when the graphite crucibles of Examples 1 and 2 were used, the power consumption of the graphite crucible in the pretreatment step increased with an increase in the number of times the sample was analyzed, but it was found that the increase was more gradual compared to Comparative Examples 1 to 3. In Example 1, the regression line showing the relationship between the number of times the outer crucible was used x and the power consumption (maximum value) y of the graphite crucible in the pretreatment step was y = 0.2132x + 1778.7, and the coefficient of determination was R 2 In Example 2, the regression line showing the relationship between the number of times the outer crucible was used x and the power consumption y of the graphite crucible in the pretreatment step was y = 0.1361x + 1777.6, and the coefficient of determination was R 2 The reason why the increase in power consumption was gentler than in the comparative example is believed to be because the increase in the resistance value of the outer crucible due to the impregnation of the upper end of the outer crucible with silicon was suppressed.

[0043] Based on the results of Examples 1 and 2 and Comparative Examples 1 to 3, a correlation equation was derived between the number of times the outer crucible was used and the power consumption of the graphite crucible in the pretreatment process when the height difference between the outer crucible and the inner crucible was arbitrarily changed, and based on the correlation equation, the height difference between the inner crucible and the outer crucible at which abnormal heat generation may occur when the number of times the outer crucible was used reached 50 times, specifically, at which the power consumption reached 1810 W, was estimated. This is because it can be predicted from the results of Examples 1 and 2 and Comparative Examples 1 to 3 that when the power consumption reaches 1810 W, there is a very high probability that abnormal elution of the sample will occur.

[0044] FIG. 5 is a graph showing the correlation between the number of times the outer crucible is used and the power consumption of the graphite crucible in the pretreatment process when graphite crucibles with height differences between the outer crucible and the inner crucible of 0.6 mm, 1.1 mm, and 1.2 mm are used.

[0045] As shown in Figure 5, when a graphite crucible with a height difference between the outer crucible and the inner crucible of 1.2 mm was used, the power consumption (MAX) was 1810 W or more when the outer crucible had been used 48 times. However, when a graphite crucible with a height difference between the outer crucible and the inner crucible of 1.1 mm was used, the power consumption (MAX) was 1810 W or less even when the outer crucible had been used more than 50 times. From the above results, it was found that when the height difference between the outer crucible and the inner crucible is greater than 1.1 mm, the probability of sample elution abnormality occurring when the outer crucible has been used 50 times or less is very high, but when the height difference is 1.1 mm or less, the probability of sample elution abnormality occurring when the outer crucible has been used 50 times or less is very low. [Explanation of symbols]

[0046] 1 Silicon sample 2. Surface oxide film 10 Graphite crucible 11 Inner crucible 11a Straight body of inner crucible 11b Bottom of inner crucible 11e Upper end of inner crucible 12 Outer crucible 12a Straight body of outer crucible 12b Bottom of outer crucible 12c Step of outer crucible 12e Upper end of outer crucible 21 Upper electrode 22 Lower electrode 23 High frequency power supply R1 Outer diameter of inner crucible R2 Inner diameter of outer crucible h1 Height of crucible h2 Height of outer crucible

Claims

1. A graphite crucible used in measuring the oxygen concentration of a metal sample by the GFA method, an inner crucible for accommodating the metal sample; and an outer crucible that accommodates the inner crucible, 1. A graphite crucible, characterized in that when the inner crucible is housed within the outer crucible, the difference in height between an upper end of the outer crucible and an upper end of the inner crucible is 1.1 mm or less.

2. The graphite crucible according to claim 1, wherein the height difference is 0 to 0.6 mm.

3. A method for measuring an oxygen concentration of a metal sample by a GFA method, comprising the steps of: A pretreatment step of heating a metal sample contained in a graphite crucible at a temperature below the melting point to remove a surface oxide film; a main analysis step of heating the metal sample from which the surface oxide film has been removed at a temperature equal to or higher than the melting point to generate CO gas derived from oxygen in the metal sample, and quantitatively analyzing the CO gas using an IR detector; The graphite crucible is an inner crucible for accommodating the metal sample; and an outer crucible that accommodates the inner crucible, A method for measuring the oxygen concentration of a metal sample, characterized in that when the inner crucible is contained within the outer crucible, the difference in height between the upper end of the outer crucible and the upper end of the inner crucible is 1.1 mm or less.

4. The inner crucible is replaced every time a new sample is measured.

4. The method for measuring an oxygen concentration in a metal sample according to claim 3, wherein the outer crucible is not replaced each time a new sample is measured, but is repeatedly used up to a predetermined number of times.

5. 5. The method for measuring an oxygen concentration in a metal sample according to claim 4, wherein the predetermined number of uses is 50 to 100 times.

6. 5. The method for measuring an oxygen concentration in a metal sample according to claim 4, wherein the metal sample is a silicon piece cut out from a silicon wafer.

7. 7. The method for measuring an oxygen concentration in a metal sample according to claim 6, wherein the silicon piece is made of a silicon single crystal doped with a p-type dopant and having a resistivity of 0.2 Ω·cm or less, or a silicon single crystal doped with an n-type dopant and having a resistivity of 0.1 Ω·cm or less.

8. An upper electrode and a lower electrode are connected to the upper end and the lower end of the outer crucible, respectively; High frequency power is supplied to the graphite crucible via the upper electrode and the lower electrode to heat the graphite crucible; The method for measuring an oxygen concentration in a metal sample according to claim 3 , wherein the upper electrode is not in contact with the inner crucible.

Citation Information

Patent Citations

  • Gas analyzer

    JP1984000951U

  • Method for analyzing oxygen in si sample

    JP2004219364A

  • Method for evaluating oxygen density in silicon crystal

    JP2018133351A

  • Apparatus for melting a silicon

    KR1020160084184A

  • Oxygen analyzer

    US4601882A