JP6717221B2 - Method for evaluating oxygen concentration in silicon crystal - Google Patents

Description

The present invention relates to a method for evaluating oxygen concentration in a silicon crystal, and more particularly to a method for evaluating oxygen concentration in a silicon wafer by the GFA (the inert Gas Fusion infrared Absorption method).

FT-IR (Fourier Transform Infrared spectroscopy analysis) is widely used as an evaluation method of oxygen concentration in silicon crystals, and ASTM (American Society for Testing and Materials: American Society for Testing and Materials) ) Is standardized. This method measures the amount of absorption by interstitial oxygen in a silicon crystal when infrared light is transmitted through the wafer, and is based on the principle of quantifying oxygen in the crystal from the amount of absorption. .. Since the light absorption amount is very sensitive to the interstitial oxygen concentration in the silicon crystal, it is possible to evaluate with high sensitivity and high reliability.

However, evaluation itself by FT-IR is impossible unless the sample transmits infrared light to some extent. For example, it is preferably used as a substrate material for an epitaxial wafer, a high-concentration boron-doped silicon single crystal having a low resistivity (for example, 20 mΩ·cm or less, particularly 10 mΩ·cm or less), and a substrate material for discrete devices. In the case of a low-resistivity silicon single crystal in which phosphorus, arsenic, and antimony are highly doped, light is absorbed by a large amount of free electrons contained in the sample, and infrared light is not transmitted at all. -There is a problem that IR cannot be applied.

As a method for evaluating the oxygen concentration in a silicon crystal, which is difficult to evaluate by FT-IR, GFA has attracted attention (see Patent Documents 1 to 3, for example). GFA is a principle that oxygen contained in a sample is gasified as CO and CO ₂ by melting the sample at a high temperature, and the gas is chemically analyzed to quantify the oxygen contained in the original sample. It is based on. The GFA method is a complete destructive method that requires melting of the sample, but has the advantage that the measuring device is relatively inexpensive and requires very little skill for measurement and can be evaluated in a short time.

Since the unit of oxygen concentration value measured by GFA is ppma, a calibration curve is required to convert this to the standardized oxygen concentration unit (atoms/cm ³ ) of ASTM. This calibration curve can be prepared by preparing a sample that can be measured by FT-IR and determining the FT-IR measured value and the GFA measured value of the oxygen concentration of the sample.

Japanese Patent Laid-Open No. 11-14543 Japanese Patent Laid-Open No. 11-201963 JP, 2007-121319, A

As described above, in a specific type of silicon wafer used as a substrate material for an epitaxial wafer, the boron concentration in the silicon crystal is increased in order to secure the desired gettering ability and improve the cleanliness of the epitaxial film, and (Bulk Micro Defect) In order to increase the density, a higher oxygen concentration than ever has been required. As described above, the GFA capable of only measuring the oxygen concentration in the silicon crystal doped with boron at a high concentration is required to further improve the measurement accuracy and reliability.

However, for such a silicon crystal having a very high oxygen concentration, there is a problem that the accuracy of analysis of the oxygen concentration by GFA is low.

Further, in GFA, due to the problem of sensitivity in chemical analysis of gas, the absolute amount of oxygen contained in the sample is increased, and therefore a large sample of 2 to 10 mm square is generally required. However, when the sample size is large, there is a problem that the measurement accuracy is deteriorated due to the oxygen concentration distribution in the sample.

Therefore, an object of the present invention is to improve the accuracy of oxygen concentration analysis in the method for evaluating oxygen concentration in silicon crystals by GFA.

In order to solve the above-mentioned problems, the method for evaluating the oxygen concentration in a silicon crystal according to the present invention provides a plurality of standard samples for preparing a calibration curve composed of a silicon crystal measurable by FT-IR, each of the plurality of standard samples being prepared. After measuring the oxygen concentration of FT-IR and GFA respectively, a calibration curve showing a one-to-one relationship between the FT-IR measured value (atoms/cm ³ ) of the oxygen concentration and the GFA measured value (ppma) is prepared. Preparation of a calibration curve and an evaluation sample made of a silicon crystal that is difficult to measure by FT-IR, the oxygen concentration of the evaluation sample was measured by GFA, and the GFA measurement value was FT- by using the calibration curve. An oxygen concentration evaluation step of converting to an IR measurement value, and the highest level of the FT-IR measurement value of the oxygen concentration of the plurality of standard samples is the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step. Is more than the upper limit of the target range of.

According to the present invention, the highest level (high-level side) of the oxygen concentration of the calibration curve preparation sample covers the target range of the oxygen concentration in the silicon crystal to be evaluated. It is possible to improve the quality of the calibration curve for converting the GFA measurement value created by the above into the FT-IR measurement value. Therefore, even when the oxygen concentration in the silicon crystal to be evaluated is very high, the GFA measurement value of the oxygen concentration can be correctly converted into the FT-IR measurement value, and the silicon crystal which is difficult to measure by FT-IR. It is possible to accurately obtain the FT-IR measurement value of the oxygen concentration in the inside.

In the present invention, the highest level of FT-IR measurement value of oxygen concentration of the plurality of standard samples is preferably 15.7×10 ¹⁷ (atoms/cm ³ ) or more, and 16×10 ¹⁷ (atoms/atoms/ It is particularly preferable that the size is not less than cm ³ ). According to this, the quality of the calibration curve used for the oxygen concentration analysis of the evaluation sample in which the oxygen concentration in the silicon crystal is very high can be improved, and the accuracy of the oxygen concentration analysis by GFA can be improved.

In the present invention, the lowest level of the FT-IR measurement value of the oxygen concentration of the plurality of standard samples may be equal to or lower than the lower limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step. It is particularly preferably 8×10 ¹⁷ (atoms/cm ³ ) or less. The lowest level (low level side) oxygen concentration of the calibration curve preparation sample covers the target range of oxygen concentration in the silicon crystal to be evaluated, so GFA measurement made based on the calibration curve preparation sample It is possible to improve the quality of the calibration curve for converting the value into the FT-IR measurement value.

In the present invention, the plurality of standard samples include a first standard sample, a second standard sample having a higher oxygen concentration than the first standard sample, and a third standard sample having a higher oxygen concentration than the second standard sample. Including a standard sample, the FT-IR measurement value of the oxygen concentration of the first standard sample is less than or equal to the lower limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step, The FT-IR measurement value of the oxygen concentration of the standard sample is preferably equal to or higher than the upper limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step. In this case, the FT-IR measurement value of the oxygen concentration of the first standard sample is preferably 8×10 ¹⁷ (atoms/cm ³ ) or less. Further, the FT-IR measurement value of the oxygen concentration of the third standard sample is preferably 15.7×10 ¹⁷ (atoms/cm ³ ) or more, and 16×10 ¹⁷ (atoms/cm ³ ) or more. Is particularly preferable. Furthermore, the FT-IR measurement value of the oxygen concentration of the second standard sample is the FT-IR measurement value of the oxygen concentration of the first standard sample and the FT-IR measurement value of the oxygen concentration of the third standard sample. The average value is preferably ±0.5×10 ¹⁷ (atoms/cm ³ ).

In the present invention, the evaluation sample is preferably a silicon crystal doped with a p-type dopant and having a specific resistance of 0.2 (Ω·cm) or less. This type of silicon crystal is preferably used as a substrate material for an epitaxial wafer and requires a very high oxygen concentration, but it is difficult to measure the oxygen concentration by FT-IR and the effect of the present invention is remarkable. ..

In the present invention, it is also preferable that the evaluation sample is a silicon crystal doped with an n-type dopant and having a specific resistance of 0.1 (Ω·cm) or less. This type of silicon crystal is preferably used as a substrate material for discrete devices, but it is difficult to measure the oxygen concentration by FT-IR and the effect of the present invention is remarkable.

In the present invention, it is preferable that the standard sample is a silicon chip cut out from a region where the in-plane variation in oxygen concentration of the silicon wafer is within ±0.05×10 ¹⁷ (atoms/cm ³ ). The in-plane variation of the oxygen concentration in the silicon wafer becomes more remarkable as the oxygen concentration increases, and the quality of the calibration curve may deteriorate depending on the cutting position from the wafer. However, the quality of the calibration curve can be improved by cutting out a region where the in-plane variation of the oxygen concentration is small and using it as a sample for preparing the calibration curve. Therefore, the accuracy of oxygen concentration analysis by GFA can be improved.

According to the present invention, the accuracy of oxygen concentration analysis can be increased in the method of evaluating oxygen concentration in silicon crystals by GFA.

FIG. 1 is a flowchart illustrating a method for evaluating oxygen concentration in silicon crystals. FIG. 2 is a flowchart for explaining the calibration curve creating step. FIG. 3 is a graph showing an example of in-plane distribution of oxygen concentration of a silicon wafer for creating a calibration curve, where the horizontal axis represents the wafer radial position and the vertical axis represents the oxygen concentration FT-IR measurement value (atoms/ cm ³ ) respectively. FIG. 4 is a graph of oxygen concentration distribution with a calibration curve drawn. FIG. 5 is a flowchart for explaining the oxygen concentration evaluation step of the silicon wafer using the calibration curve. FIG. 6 is a graph showing the one-way analysis result of the slope of the calibration curve with respect to the high level oxygen concentration, where the horizontal axis is the FT-IR measurement value (atoms/cm ³ ) of the high level oxygen concentration and the vertical axis is the calibration amount. The slope of each line is shown. FIG. 7 is a graph showing the one-way analysis result of the intercept of the calibration curve with respect to the high level oxygen concentration, the abscissa is the FT-IR measurement value (atoms/cm ³ ) of the high level oxygen concentration, and the ordinate is the calibration. Each line segment is shown.

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

FIG. 1 is a flowchart illustrating a method for evaluating oxygen concentration in silicon crystals.

As shown in FIG. 1, the method for evaluating the oxygen concentration in a silicon crystal is a calibration curve creating step in which a calibration curve is created in advance using a standard sample for creating a calibration curve of a silicon crystal capable of measuring the oxygen concentration by FT-IR. Prepare a sample for evaluation of S11 and a silicon crystal whose oxygen concentration is difficult to measure by FT-IR, measure the oxygen concentration of this sample for evaluation by GFA, and then use the calibration curve prepared in advance. And an oxygen concentration evaluation step S12 for converting the GFA measurement value of the oxygen concentration into an FT-IR value.

FIG. 2 is a flowchart illustrating the calibration curve creation step S11.

As shown in FIG. 2, in the calibration curve creating step S11, a plurality of silicon wafers are prepared, and the in-plane distribution of oxygen concentration of these silicon wafers is measured by FT-IR (step S21). As the silicon wafer, a wafer whose oxygen concentration can be measured by FT-IR is used. Therefore, low resistance varieties in which boron is highly doped are not used. As the silicon wafer, a silicon wafer that is manufactured so as to substantially satisfy the oxygen concentration conditions required to create the calibration curve is prepared. The oxygen concentration is measured by FT-IR according to ASTM F-121 (1979).

Next, from the FT-IR measurement results, after selecting a silicon wafer that satisfies the oxygen concentration condition required to create a calibration curve, a silicon chip is cut out from the silicon wafer to create a standard sample for creating a calibration curve of a silicon crystal. To do. Specifically, a first level (low level) standard sample (first standard sample) having a relatively low oxygen concentration, and a second level (intermediate level) standard sample (oxygen concentration higher than the first level) A second standard sample) and a third level (high level) standard sample (third standard sample) having an oxygen concentration higher than the second level are prepared (step S22).

Here, the three levels of the oxygen concentration are set so as to cover the target range of the oxygen concentration of one or a plurality of types of silicon crystals to be evaluated in the oxygen concentration evaluation step S12, and particularly, the third standard sample. The level oxygen concentration is set to be equal to or higher than the upper limit value of the target range. Specifically, the third level oxygen concentration is preferably 15.7×10 ¹⁷ (atoms/cm ³ ) or more, and particularly preferably 16×10 ¹⁷ (atoms/cm ³ ) or more. The first level oxygen concentration of the first standard sample is set to be equal to or lower than the lower limit value of the target range. Specifically, the first level oxygen concentration is preferably 8×10 ¹⁷ (atoms/cm ³ ) or less. Furthermore, the second level oxygen concentration of the second standard sample is preferably an average value of the first level and the third level of ±0.5×10 ¹⁷ (atoms/cm ³ ).

Conventionally, of the three levels of the oxygen concentration for preparing the calibration curve, the oxygen concentration on the high level side is a relatively low value of about 14±0.5×10 ¹⁷ (atoms/cm ³ ). When a silicon wafer with a very high oxygen concentration in recent years is evaluated by GFA using a calibration curve created based on such a high level side oxygen concentration, the measurement error becomes large, and there are those that should originally meet the standard. In some cases, it was judged to be out of standard, and the correct evaluation could not be performed. However, by setting the oxygen concentration on the high-level side of the standard sample for preparing the calibration curve to a higher value as in the present embodiment, it is possible to increase the reliability of the calibration curve and reduce the measurement error of the oxygen concentration. Become.

FIG. 3 is a graph showing an example of in-plane distribution of oxygen concentration of a silicon wafer for creating a calibration curve, where the horizontal axis represents the wafer radial position and the vertical axis represents the oxygen concentration FT-IR measurement value (atoms/ cm ³ ) respectively.

As shown in FIG. 3, the oxygen concentration of a silicon wafer having a diameter of about 200 mm changes in the radial direction. In this example, the fluctuation of the oxygen concentration is small in the range from the wafer center to 50 mm (see range a), and the fluctuation of the oxygen concentration is particularly small in the range from the wafer center to 20 mm (see range b), but 50 to 100 mm. The fluctuation of oxygen concentration is large in the range. This tendency is more remarkable as the oxygen concentration in the silicon crystal is higher. If an area with a large variation in oxygen concentration is used as a standard sample for preparing a calibration curve, the measurement error of GFA becomes large, and as a result, the quality of the calibration curve prepared based on the GFA measurement value deteriorates. Therefore, as the standard sample cut out from the silicon wafer, it is desirable to use a region in which the fluctuation of oxygen concentration is as small as possible, particularly the fluctuation range of the oxygen concentration, that is, the oxygen concentration of a specific region of the silicon wafer used as the standard sample. The in-plane variation is preferably within ±0.05×10 ¹⁷ (atoms/cm ³ ).

Next, the oxygen concentration of the silicon chip (standard sample) is measured by GFA (step S23). The GFA measuring device used for measuring the oxygen concentration by GFA is a device for measuring the oxygen content in a sample by charging a silicon chip to be measured into a graphite crucible in an analysis furnace and heating and melting the sample. CO and CO ₂ are generated by reacting the oxygen in the graphite crucible at a high temperature, and these gases are chemically analyzed to quantify the oxygen contained in the original sample. In the present invention, the detailed measuring method of oxygen concentration by GFA is not particularly limited, and various measuring methods can be adopted.

Next, a calibration curve is created from the one-to-one relationship between the FT-IR measurement value and the GFA measurement value of each standard sample (step S24). The calibration curve can be obtained, for example, by performing regression analysis by calculation of the least square method.

FIG. 4 is a graph of oxygen concentration distribution with a calibration curve drawn.

As shown in FIG. 4, the calibration curve can be obtained from the plotted values of oxygen concentration at three levels (low level, intermediate level and high level). In the present embodiment, the FT-IR measured value of the low level oxygen concentration is about 8×10 ¹⁷ (atoms/cm ³ ), and the FT-IR measured value of the intermediate level oxygen concentration is about 11.5×10 ^17. (Atoms/cm ³ ), and the FT-IR measurement value of high-level oxygen concentration is about 16×10 ¹⁷ (atoms/cm ³ ) (see the plotted value with a circle in the figure (after examination)). Conventionally, the high-level oxygen concentration was a low value of approximately 14 to 15×10 ¹⁷ (atoms/cm ³ ) (see the diamond-shaped plot value in the figure (before study)), but the high-level oxygen concentration value The higher the value, the better the quality of the calibration curve. Therefore, by using this calibration curve, the FT-IR measured value (atoms/cm ³ ) corresponding to the arbitrary GFA measured value (ppma) of the oxygen concentration in the silicon crystal can be accurately obtained. Conversely, it is also possible to obtain a GFA measurement value (ppma) corresponding to an arbitrary FT-IR measurement value (atoms/cm ³ ).

The calibration curve creating step S11 is periodically performed before measuring the oxygen concentration using the GFA measuring device. That is, once a calibration curve is created for one GFA measuring device at the time of its first use, it is not possible to continue using the same calibration curve for tens of days. It is the expiration date, and it is necessary to create a new calibration curve. This is because the output of the GFA measuring device slightly changes due to a slight change in the measurement environment, and the measurement error (conversion error) due to the calibration curve becomes large unless a new calibration curve is created.

Next, the oxygen concentration evaluation step S12 using this calibration curve will be described.

FIG. 5 is a flow chart for explaining the oxygen concentration evaluation step S12 of the silicon wafer using the calibration curve.

As shown in FIG. 5, in the evaluation of the oxygen concentration of a silicon wafer by GFA, first, a silicon wafer that is difficult to measure by FT-IR, such as a silicon wafer doped with a high concentration of dopant, is prepared (step S31). For example, in a silicon wafer that is heavily doped with a p-type dopant such as boron, it is preferable to evaluate a silicon wafer having a specific resistance of 0.2 (Ω·cm) or less. Further, in a silicon wafer that is heavily doped with an n-type dopant such as phosphorus, arsenic, or antimony, it is preferable to evaluate silicon wafers having a specific resistance of 0.1 (Ω·cm) or less.

Next, a plurality of silicon chips (evaluation samples) cut out from appropriate places on the surface of the silicon wafer are prepared. Specifically, a silicon chip of a predetermined size including measurement points set at predetermined intervals along the radial direction of the wafer is manufactured (step S32).

Next, the oxygen concentration of each silicon chip is measured by GFA (step S33). The method for measuring the oxygen concentration by GFA is as described above, and the detailed measuring method is not particularly limited.

Next, the GFA measurement value (ppma) of the oxygen concentration of each chip is converted into the FT-IR measurement value (atoms/cm ³ ) using the calibration curve (step S34). Then, the in-plane distribution of the oxygen concentration of the silicon wafer can be obtained by mapping the FT-IR measurement value in association with the coordinate position of the silicon chip cut out from the silicon wafer.

As described above, the method for evaluating the oxygen concentration in the silicon crystal according to the present embodiment measures the oxygen concentration of the standard sample of the silicon crystal by FT-IR and GFA, and measures the oxygen concentration of the standard sample by the GFA measurement value (ppma). ) With the FT-IR measurement value (atoms/cm ³ ) in the calibration curve creation step S11 for creating a calibration curve showing a one-to-one relationship, the oxygen concentration evaluation step S12 Since it is a value equal to or higher than the upper limit value of the target range of the oxygen concentration of the silicon crystal type to be evaluated in, the quality of the calibration curve created in the calibration curve creation step S11 can be improved, and the GFA measurement value (ppma) The accuracy of conversion from FT-IR measured values (atoms/cm ³ ) can be improved, and the accuracy of measurement of oxygen concentration in silicon crystals by GFA can be improved. Therefore, it is possible to accurately obtain the FT-IR measurement value of the oxygen concentration in the silicon crystal, which is difficult to measure by FT-IR.

Although the preferred embodiments of the present invention have been described above, 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. It goes without saying that it is included in the range.

For example, in the above-described embodiment, the case where the oxygen concentration of a low-resistance type silicon wafer in which boron is highly doped is evaluated is described as an example, but the present invention is limited to the case of evaluating such a silicon wafer. However, various silicon wafers that are difficult to measure by FT-IR can be targeted. Therefore, a silicon wafer that is highly doped with phosphorus, arsenic, or antimony may be an evaluation target. Further, the shape of the silicon crystal is not required to be a wafer shape, and various shapes can be targeted. Further, it is not limited to silicon single crystal, and may be polycrystalline silicon.

Further, in the above embodiment, three levels were prepared as the FT-IR measurement value of the oxygen concentration for preparing the calibration curve, but it may be two levels or four or more levels. In this case, the highest level of the FT-IR measurement value of the oxygen concentration of the plurality of standard samples may be equal to or higher than the upper limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step S12. The lowest level of the FT-IR measurement value of the oxygen concentration of the standard sample may be equal to or lower than the lower limit value of the target range.

Of the three levels of oxygen concentration used to create the calibration curve, the effect of the high level oxygen concentration value on the slope and intercept of the calibration curve was examined. The low level oxygen concentration FT-IR measurement value used to create the calibration curve is about 8×10 ¹⁷ (atoms/cm ³ ), and the intermediate level oxygen concentration FT-IR measurement value is about 11.5×10 ¹⁷ (Atoms/cm ³ ). Then, the FT-IR measurement value of the high level oxygen concentration is about 13×10 ¹⁷ (atoms/cm ³ ) (Comparative example 1), about 14.5×10 ¹⁷ (atoms/cm ³ ) (Comparative example 2). , About 16×10 ¹⁷ (atoms/cm ³ ) (Example 1).

The number of samples of the calibration curve according to Comparative Example 1 was 42, the number of samples of the calibration curve according to Comparative Example 2 was 194, and the number of samples of the calibration curve according to Example 1 was 318, and the slope and intercept values of each calibration curve. Was calculated and graphed.

FIG. 6 is a graph showing the one-way analysis result of the slope of the calibration curve with respect to the high level oxygen concentration, where the horizontal axis is the FT-IR measurement value (atoms/cm ³ ) of the high level oxygen concentration and the vertical axis is the calibration amount. The slope of each line is shown.

As shown in the plot group on the left side of FIG. 6, the average value of the slope of the calibration curve of Comparative Example 1 in which the high level oxygen concentration value was about 13×10 ¹⁷ (atoms/cm ³ ) was about 0.479, The standard deviation was about 0.036. The lower 95% value was about 0.468, and the upper 95% value was about 0.490.

Further, as shown in the group of plots in the center of FIG. 6, the average value of the slope of the calibration curve of Comparative Example 2 in which the high level oxygen concentration value was about 14.5×10 ¹⁷ (atoms/cm ³ ), was about 0. It was 0.493 and the standard deviation was about 0.031. The lower 95% value was about 0.489, and the upper 95% value was about 0.498.

Further, as shown in the group of plots on the right side of FIG. 6, the average value of the slope of the calibration curve of Example 1 in which the value of the high level oxygen concentration was about 16×10 ¹⁷ (atoms/cm ³ ) was about 0.504. And the standard deviation was about 0.021. The lower 95% value was about 0.501, and the upper 95% value was about 0.506.

Furthermore, when a straight line that passes the average value of each level is created in FIG. 6, the FT-IR measurement value at which this straight line becomes the theoretical value (0.5) or more is about 15.7×10 ¹⁷ (atoms/cm ³ ). Therefore, it was found that a highly reliable calibration curve can be prepared if the high-level oxygen concentration value used for preparing the calibration curve is 15.7×10 ¹⁷ (atoms/cm ³ ) or more.

FIG. 7 is a graph showing the one-way analysis result of the intercept of the calibration curve with respect to the high level oxygen concentration, the abscissa is the FT-IR measurement value (atoms/cm ³ ) of the high level oxygen concentration, and the ordinate is the calibration. Each line segment is shown.

Further, as shown in the group of plots on the left side of FIG. 7, the average value of the intercepts of the calibration curve of Comparative Example 1 in which the high level oxygen concentration value was about 13×10 ¹⁷ (atoms/cm ³ ) was about 1.338. , And the standard deviation was about 0.409. The lower 95% value was about 1.210, and the upper 95% value was about 1.465.

Further, as shown in the plot group in the center of FIG. 7, the average value of the intercept of the calibration curve of Comparative Example 2 in which the value of the high level oxygen concentration was about 14.5×10 ¹⁷ (atoms/cm ³ ) was about 0. .911 and the standard deviation was about 0.419. The lower 95% value was about 0.911, and the upper 95% value was about 1.030.

Further, as shown in the plot group on the right side of FIG. 7, the average value of the intercept of the calibration curve of Example 1 in which the value of the high level oxygen concentration was about 16×10 ¹⁷ (atoms/cm ³ ) was about 0.656. , And the standard deviation was about 0.265. The lower 95% value was about 0.627, and the upper 95% value was about 0.686.

As described above, the intercept of the calibration curve of Example 1 in which the value of the high level oxygen concentration was about 16×10 ¹⁷ (atoms/cm ³ ) showed a blank value (0. It turned out to be the closest to 5). The value when measured in an empty state without putting a sample into the GFA measuring device should ideally be zero, but it does not actually become zero and has a blank value (background value). .. Therefore, it can be said that if the intercept value is around 0.5, it is almost a theoretical value.

S11 Calibration curve preparation step S12 Oxygen concentration evaluation step S21 FT-IR measurement step S22 Calibration curve preparation standard sample preparation step S23 GFA measurement step S24 Calibration curve preparation step S31 Wafer preparation step S32 Evaluation sample preparation step S33 GFA measurement step S34 conversion Step

Claims (11)

Prepare a plurality of calibration curve standard samples made of silicon crystals that can be measured by FT-IR, measure the oxygen concentration of each of the plurality of standard samples by FT-IR, and melt each of the plurality of standard samples. After measuring the oxygen concentration of the gas generated by GFA, the FT-IR measurement value (atoms/cm ³ ) of the oxygen concentration of the standard sample and the GFA measurement value (ppma) of the oxygen concentration of the gas A calibration curve creating step for creating a calibration curve showing a one-to-one relationship,
An evaluation sample made of a silicon crystal that is difficult to measure by FT-IR is prepared, and the oxygen concentration of the gas generated by melting the evaluation sample is measured by GFA, and then the GFA measurement value is obtained by using the calibration curve. And an oxygen concentration evaluation step for converting the FT-IR measurement value to
The highest level of the FT-IR measurement value of the oxygen concentration of the plurality of standard samples is not less than the upper limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step ,
The upper limit value is 15.7×10 ¹⁷ (atoms/cm ³ ), which is an oxygen concentration evaluation method in a silicon crystal. The upper limit, 16 × a ^{10 17 (atoms / cm 3)} , the oxygen concentration evaluating method in the silicon crystal according to claim 1. The lowest levels of FT-IR measurement of the oxygen concentration of the plurality of standard samples is less than the lower limit of the target range of oxygen concentration in the silicon crystal to be evaluated by the oxygen concentration evaluating step, according to claim 1 or 2 The method for evaluating the oxygen concentration in a silicon crystal according to 1. The method for evaluating oxygen concentration in a silicon crystal according to claim 3 , wherein the lower limit value is 8×10 ¹⁷ (atoms/cm ³ ). Prepare a plurality of calibration curve standard samples made of silicon crystals that can be measured by FT-IR, measure the oxygen concentration of each of the plurality of standard samples by FT-IR, and melt each of the plurality of standard samples. After measuring the oxygen concentration of the gas generated by GFA, the FT-IR measurement value (atoms/cm ³ ) of the oxygen concentration of the standard sample and the GFA measurement value (ppma) of the oxygen concentration of the gas A calibration curve creating step for creating a calibration curve showing a one-to-one relationship,
An evaluation sample made of a silicon crystal that is difficult to measure by FT-IR is prepared, and the oxygen concentration of the gas generated by melting the evaluation sample is measured by GFA, and then the GFA measurement value is obtained by using the calibration curve. And an oxygen concentration evaluation step for converting the FT-IR measurement value to
The highest level of the FT-IR measurement value of the oxygen concentration of the plurality of standard samples is not less than the upper limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step ,
The lowest level of the FT-IR measurement value of the oxygen concentration of the plurality of standard samples is equal to or lower than the lower limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step,
The lower limit value is 8×10 ¹⁷ (atoms/cm ³ ), which is an oxygen concentration evaluation method in a silicon crystal. Prepare a plurality of calibration curve standard samples made of silicon crystals that can be measured by FT-IR, measure the oxygen concentration of each of the plurality of standard samples by FT-IR, and melt each of the plurality of standard samples. After measuring the oxygen concentration of the gas generated by GFA, the FT-IR measurement value (atoms/cm ³ ) of the oxygen concentration of the standard sample and the GFA measurement value (ppma) of the oxygen concentration of the gas A calibration curve creating step for creating a calibration curve showing a one-to-one relationship,
An evaluation sample made of a silicon crystal that is difficult to measure by FT-IR is prepared, and the oxygen concentration of the gas generated by melting the evaluation sample is measured by GFA, and then the GFA measurement value is obtained by using the calibration curve. And an oxygen concentration evaluation step for converting the FT-IR measurement value to
The plurality of standard samples include a first standard sample, a second standard sample having an oxygen concentration higher than that of the first standard sample, and a third standard sample having an oxygen concentration higher than that of the second standard sample. ,
The FT-IR measurement value of the oxygen concentration of the first standard sample is equal to or lower than the lower limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step,
The FT-IR measurement value of the oxygen concentration of the third standard sample is not less than the upper limit value of the target range of the oxygen concentration of the silicon crystal to be evaluated in the oxygen concentration evaluation step ,
The lower limit is 8×10 ¹⁷ (atoms/cm ³ ),
The upper limit value is 15.7×10 ¹⁷ (atoms/cm ³ ), the method for evaluating the oxygen concentration in a silicon crystal. The method for evaluating oxygen concentration in a silicon crystal according to claim 6 , wherein the upper limit value is 16×10 ¹⁷ (atoms/cm ³ ). The FT-IR measurement value of the oxygen concentration of the second standard sample is an average value ± of the FT-IR measurement value of the oxygen concentration of the first standard sample and the FT-IR measurement value of the oxygen concentration of the third standard sample. The method for evaluating oxygen concentration in a silicon crystal according to claim 6 or 7 , wherein the oxygen concentration is 0.5×10 ¹⁷ (atoms/cm ³ ). The oxygen concentration in the silicon crystal according to any one of claims 1 to 8 , wherein the evaluation sample is a silicon crystal doped with a p-type dopant and having a specific resistance of 0.2 (Ω·cm) or less. Evaluation method. The oxygen concentration in the silicon crystal according to any one of claims 1 to 8 , wherein the evaluation sample is a silicon crystal doped with an n-type dopant and having a specific resistance of 0.1 (Ω·cm) or less. Evaluation method. The standard sample in-plane variation of the oxygen concentration of the silicon wafer is a silicon chip cut from a region within ^{± 0.05 × 10 17 (atoms /} cm 3), in any one of claims 1 to 10 A method for evaluating oxygen concentration in a silicon crystal as described.