Carbon concentration measuring method
The FT-IR method for measuring carbon concentration in silicon wafers and epitaxial wafers accurately matches SIMS results by subtracting reference spectra and applying a derived relational expression, allowing precise high-concentration measurements with minimal deviation.
Patent Information
- Application Number
- JP2023110244
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-07-04
- Publication Date
- 2025-01-20
- Estimated Expiration
- 2043-07-04
AI Technical Summary
Existing FT-IR methods for measuring carbon concentration in silicon wafers and epitaxial wafers deviate significantly from the accurate SIMS method when measuring high carbon concentrations, making them unsuitable for large-scale product inspections.
A method using FT-IR to measure carbon concentration by subtracting the absorbance spectrum of a reference silicon single crystal without carbon from the measured sample, calculating the differential absorbance, and applying a derived relational expression to determine carbon concentration accurately.
Enables accurate measurement of carbon concentration in silicon single crystals and epitaxial wafers with high concentrations, matching the precision of SIMS methods without destruction, with errors less than 20%.
Smart Images

Figure 2025008250000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a method for measuring carbon concentration. [Background technology]
[0002] Impurities such as carbon atoms and oxygen atoms contained in silicon wafers are important factors that control the electrical quality of the wafer. With the miniaturization and high integration of semiconductor devices in recent years, carbon in silicon wafers is thought to be the cause of leakage current, so it is generally desirable to reduce it as much as possible. On the other hand, advances in film formation technology using low-pressure CVD furnaces have made it possible to form epitaxial layers with locally high carbon concentrations using silicon wafers as substrates, and new applications that have never been seen before are beginning to be considered. Against this background, evaluation technology for carbon concentration in silicon wafers and epitaxial wafers is required for extremely low concentrations (1×10 15 atoms / cm 3 or less) to high concentrations (5 × 10 17 atoms / cm 3 It is necessary to respond to a wide range of issues, including the above.
[0003] Photoluminescence (PL) and secondary ion mass spectrometry (SIMS) are well-established methods for evaluating carbon concentration in silicon wafers and epitaxial wafers. While these methods are highly accurate, they require electron beam irradiation prior to measurement and require long measurement times, making them unsuitable for large-scale product delivery inspections. Therefore, quantitative analysis using Fourier transform infrared spectroscopy (FT-IR) is widely adopted as a rapid method for measuring carbon concentration in silicon wafers and epitaxial wafers. FT-IR has the advantage of being able to rapidly measure carbon concentration in the depth direction of the substrate nondestructively. In practice, the method involves determining the localized vibrational absorption peak of substitutional carbon (Cs) from the difference absorbance spectrum between the absorbance spectrum of the evaluation sample and that of a reference sample containing essentially no carbon, and then converting the difference absorption coefficient into carbon concentration. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] Japanese Patent Application Publication No. 2020-012772 [Non-patent literature]
[0005] [Non-Patent Document 1] JEITA EM-3503 Summary of the Invention [Problem to be solved by the invention]
[0006] As described above, there has been a demand for the evaluation of silicon wafers and epitaxial wafers containing carbon in various concentration ranges by a simple FT-IR method. For example, Patent Document 1 discloses that the FT-IR method can be used to evaluate extremely low carbon concentrations (5 × 10 15 atoms / cm 3In Non-Patent Document 1, a method for quantifying the amount of 605 cm -1 Determine the differential absorption coefficient of carbon at [Carbon concentration of silicon wafer (atoms / cm 3 )]=8.2×10 16 × [difference absorption coefficient] … (1) A method for estimating carbon concentration using the following relational expression is described.
[0007] However, the carbon concentration of the silicon wafer to which this relational expression applies is 2 × 10 as described in the text of Non-Patent Document 1. 15 ~3×10 17 atoms / cm 3 When measuring an epitaxial wafer containing carbon at a concentration that greatly exceeds the solid solubility in the epitaxial layer, the measured carbon concentration will deviate from the actual carbon concentration obtained by SIMS or the like.
[0008] The present invention has been made to solve the above problems, and the FT-IR method has been used to measure 5×10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 The object of the present invention is to provide a carbon concentration measurement method that does not deviate from the carbon concentration determined by SIMS when measuring silicon single crystals containing carbon in the following high concentration range. [Means for solving the problem]
[0009] The present invention has been made to achieve the above object, and provides a method for measuring the substitutional carbon concentration in a silicon single crystal by FT-IR, wherein the effective range of the carbon concentration measurement is 5×10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3a differential absorption spectrum obtained by subtracting an absorbance spectrum obtained by irradiating a reference silicon single crystal containing no carbon with infrared light from an absorbance spectrum obtained by irradiating the silicon single crystal to be measured with infrared light, the differential absorption coefficient being calculated from the differential absorbance obtained from the localized vibrational absorption peak of substitutional carbon and the thickness of the measurement object; [Carbon concentration (atoms / cm 3 )]=1.2×10 17 × [Differential absorption coefficient] … (2) The carbon concentration of the silicon single crystal to be measured is determined using the following relational expression:
[0010] By using this carbon concentration measurement method, when measuring silicon single crystals containing locally high concentrations of carbon, the carbon concentration determined by the FT-IR method can be measured more accurately without significantly differing from the carbon concentration determined by the highly accurate SIMS method.
[0011] In this case, the carbon concentration of the silicon single crystal to be measured can be the carbon concentration in the epitaxial layer of the silicon epitaxial wafer.
[0012] This allows for more accurate measurement of the carbon concentration determined by FT-IR, without significant deviation from that determined by the highly accurate SIMS method, when measuring a sample containing locally high concentrations of carbon in the epitaxial layer of a silicon epitaxial wafer.
[0013] In this case, the differential absorption coefficient is [Differential absorption coefficient] = [Differential absorbance] × 23.03 ÷ [Thickness of measurement object (mm)] … (3) It can be calculated using the following relational expression.
[0014] This allows the carbon concentration in the silicon single crystal or the epitaxial layer determined by the FT-IR method to be measured more accurately without deviating significantly from the carbon concentration determined by the highly accurate SIMS method. [Effects of the Invention]
[0015] As described above, the carbon concentration measurement method of the present invention can measure the carbon concentration of 5 × 10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 For silicon single crystals, particularly epitaxial wafers, containing carbon in the high concentration range below, the FT-IR method makes it possible to measure the carbon concentration easily, quickly, and non-destructively with accuracy comparable to that of the SIMS method. [Brief explanation of the drawings]
[0016] [Figure 1] 1 shows a flow of carbon concentration measurement according to the present invention. [Figure 2] An example of a difference absorbance spectrum showing how to draw a baseline is shown. [Figure 3] 1 shows differential absorbance spectra (after baseline correction) of epitaxial wafers in which the carbon concentration in the epitaxial layer was varied in an experimental example. [Figure 4] 1 shows the relationship between the difference absorbance read from the difference absorbance spectrum and the carbon concentration measured by the SIMS method in an experimental example. [Figure 5] 1 shows the relationship between the differential absorption coefficient of the epitaxial layer obtained by the FT-IR method and the carbon concentration of the epitaxial layer measured by the SIMS method in an experimental example. [Figure 6] The carbon concentrations of the epitaxial wafers shown in Examples 1 to 5 were measured using the carbon concentration measuring method of the present invention, and the results are shown below. [Figure 7] 1 shows the difference in carbon concentration when the carbon concentration measurement method of the present invention and the existing carbon concentration measurement method are applied to the epitaxial wafers shown in Comparative Examples 1 to 5. DETAILED DESCRIPTION OF THE INVENTION
[0017] The present invention will be described in detail below, but the present invention is not limited thereto.
[0018] As mentioned above, 5 x 10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 There has been a demand for the ability to measure carbon concentration by FT-IR with the same accuracy as SIMS, even for silicon single crystals, particularly epitaxial wafers, that contain carbon in the high concentration range below.
[0019] As a result of extensive research into the above-mentioned problems, the present inventors have discovered a method for measuring the substitutional carbon concentration in a silicon single crystal by FT-IR, in which the effective range of the measurement is 5×10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 The inventors have found that the carbon concentration can be measured by the FT-IR method with an accuracy comparable to that of the SIMS method even for silicon single crystals containing carbon in the above concentration range, by deriving a differential absorption coefficient from the differential absorbance spectrum obtained by subtracting the absorbance spectrum of a reference silicon single crystal containing no carbon from the absorbance spectrum of the silicon single crystal to be measured, and then calculating the carbon concentration using relational expression (2), thereby completing the present invention.
[0020] The present inventors have also found that the carbon concentration measurement method of the present invention is preferably applied to measurement of the carbon concentration in the epitaxial layer of a silicon epitaxial wafer.
[0021] The reference sample and the method of baseline correction used in the present invention are not particularly limited. For example, from the viewpoint of measurement accuracy, the reference sample can be an epitaxial wafer in which an epitaxial layer essentially containing no carbon is formed to the same film thickness as the evaluation sample on the same silicon wafer as the evaluation sample. In addition, the baseline correction is performed at 640 cm following Non-Patent Document 1. -1 and 560cm -1 can be drawn so that both ends are This will prevent errors caused by the person making the measurement, resulting in a more accurate carbon concentration evaluation method.
[0022] [Carbon concentration measurement method] The carbon concentration measurement method of the present invention will be described below with reference to Fig. 1. The carbon concentration measurement of the present invention is carried out in accordance with the steps shown in Fig. 1.
[0023] (Process 1) First, prepare an epitaxial wafer to be used as an evaluation sample and an epitaxial wafer to be used as a reference sample. These silicon substrates should have the same thickness and resistivity and be surface-polished. For example, when measuring carbon concentration, a silicon substrate with a thickness of 2.0 mm and a resistivity of 3 Ω·cm or more if the conductivity type is p-type, or 1 Ω·cm or more if the conductivity type is n-type, can be prepared. In addition, the epitaxial layer to be formed on the silicon substrate should be prepared with the same thickness and resistivity, with only the carbon concentration being different. The epitaxial layer of the evaluation sample should have a thickness of 5×10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 A reference sample containing carbon in the following concentration range and substantially no carbon in the epitaxial layer is prepared.
[0024] In this case, at least one of SiH4, SiH2Cl2, and SiHCl3 can be used as a silicon source for the mixed gas atmosphere containing silicon and carbon, and at least one of SiH3(CH3), SiH2(CH3)2, SiH(CH3)3, CH4, C2H6, and C3H8 can be used as a carbon source for the mixed gas atmosphere containing silicon and carbon.
[0025] (Process 2) The evaluation sample and the reference sample are placed in the optical path of the FT-IR instrument, and infrared light is irradiated to obtain an absorbance spectrum. There are no particular restrictions on the gas atmosphere in the sample chamber or the number of integrations in the FT-IR instrument. However, a highly accurate absorbance spectrum can be obtained by using a vacuum pump to remove the influence of the atmosphere, or by setting the number of integrations to, for example, 40,000. The difference absorbance spectrum is then obtained by subtracting the absorbance spectrum of the reference sample from the absorbance spectrum of the evaluation sample. As shown in Figure 2, the absorbance spectrum at 640 cm -1 and 560cm -1 By drawing a line so that both ends of the line are the baseline, 605cm -1 The distance between the localized vibrational absorption peak of Cs observed nearby and the baseline is read, and this is taken as the differential absorbance.
[0026] (Step 3) The differential absorbance obtained in step 2 and the thickness of the epitaxial layer are substituted into the following relational expression (3) in Non-Patent Document 1 to calculate the differential absorption coefficient. [Differential absorption coefficient] = [Differential absorbance] × 23.03 ÷ [Thickness of measurement object (mm)] … (3)
[0027] The carbon concentration contained in the epitaxial layer of the epitaxial wafer is estimated by applying the differential absorption coefficient to the following relational expression (2). [Carbon concentration (atoms / cm 3 )]=1.2×10 17 × [Differential absorption coefficient] … (2)
[0028] By using the carbon concentration measurement method of the present invention, the carbon concentration of an epitaxial wafer containing carbon in the epitaxial layer at a carbon concentration that far exceeds the solid solubility can be estimated by the FT-IR method.
[0029] [Derivation of the carbon concentration calculation formula] Next, the process leading to the relational expression (2) described in step 3 will be explained along with experimental examples, but the present invention is not limited to these.
[0030] As one of the experiments, 5 × 10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 Epitaxial wafers containing carbon in the epitaxial layer in the following concentration ranges were prepared. That is, a silicon epitaxial film (carbon concentration: 5.0 × 10) was grown on a silicon substrate using a low-pressure CVD apparatus. 17 , 2.0×10 18 , 7.0×10 18 , 2.0×10 19 , 1.0×10 20 atoms / cm 3 The carbon concentration and thickness of the silicon epitaxial film were adjusted by changing the supply amount of source gas and the processing time in the low-pressure CVD furnace.
[0031] As a reference sample, a silicon epitaxial film was formed on a silicon substrate of the same quality as the silicon substrate used for forming the carbon-containing epitaxial wafer, without supplying a source gas containing carbon.
[0032] The absorbance spectrum of the evaluation sample was obtained by performing FT-IR measurement at room temperature. The absorbance spectrum of the reference sample was also obtained using the same instrument, and the difference absorbance spectrum was obtained by subtracting the absorbance spectrum of the reference sample from the absorbance spectrum of the evaluation sample.
[0033] In the differential absorbance spectrum calculated from the evaluation sample and the reference sample, as shown in FIG. -1 and 560cm -1 The baseline is a line drawn so that both ends are 605cm. -1The difference absorbance was calculated by reading the distance between the Cs local vibration absorption peak observed near the base line. Figure 3 shows each difference absorbance spectrum, and Figure 4 shows the relationship between the read difference absorbance and the carbon concentration of the evaluation sample measured by SIMS. As shown in Figure 3, the higher the carbon concentration, the greater the distance between the Cs local vibration absorption peak and the base line. Also, as shown in Figure 4, it can be seen that the higher the carbon concentration in the epitaxial layer of the evaluation sample, the greater the difference absorbance. Note that in Figure 4, when the carbon concentration is 1 x 10 20 atoms / cm 3 The reason why the linearity seems to be broken in sample 1 is due to the difference in film thickness (the film thickness of this sample was targeted at 1 μm, while the film thickness of the other four evaluation samples was targeted at 5.5 μm).
[0034] The differential absorption coefficient was calculated by substituting the differential absorbance into the relational expression (3) in Non-Patent Document 1. In this case, the film thickness of the epitaxial layer was determined from the profile of a separately measured SIMS measurement.
[0035] The relationship between the carbon concentration measured by SIMS and the differential absorption coefficient derived from FT-IR for each of the evaluation samples was investigated, and the results are shown in Figure 5. As shown in Figure 5, it was found that the two are proportional to each other, and using y = ax as a fitting function, the proportionality constant a was determined so as to minimize the error from the carbon concentration measured by SIMS. As a result, the relational expression (2) for calculating the carbon concentration in the present invention was derived. Note that "α" in the relational expression in Figure 5 represents the differential absorption coefficient. [Example]
[0036] EXAMPLES The present invention will be explained in more detail below by showing examples and comparative examples of the present invention, but the present invention is not limited to these.
[0037] Example 1 In the step 1, an epitaxial film (carbon atom concentration: 5.0 × 10) was grown on a silicon single crystal using monomethylsilane as a carbon source gas in a low-pressure CVD apparatus. 17 atoms / cm 3A 5.5 μm thick epitaxial film containing substantially no carbon was formed on a silicon single crystal substrate of the same standard as the evaluation sample, and this was used as the reference sample.
[0038] In the step 2, the absorbance spectra of the evaluation sample and the reference sample were measured using an FT-IR device, and the absorbance spectrum of the reference sample was subtracted from the absorbance spectrum of the evaluation sample to obtain a differential absorbance spectrum. -1 and 560cm -1 The baseline is a straight line drawn so that both ends of the line are at 605 cm (hereinafter referred to as "baseline correction"). -1 The difference absorbance was calculated by reading the distance between the localized vibrational absorption peak of Cs observed in the vicinity and the baseline, and it was found to be approximately 0.00081.
[0039] In step 3, the differential absorption coefficient was calculated from the differential absorbance obtained in step 2 and the film thickness of the epitaxial film formed on the silicon single crystal (5.7 μm (determined from the SIMS profile)) (the differential absorption coefficient at this time was approximately 3.3). As a result of substituting the differential absorption coefficient into relational expression (2) of the present invention, the carbon concentration in the epitaxial film formed on the silicon single crystal was approximately 4.0 × 10 17 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 5.0 × 10 17 atoms / cm 3 Therefore, the error was approximately 20%. The results are shown in Figure 6. Figure 6 shows the relationship between the carbon concentration in the epitaxial film formed on the silicon single crystal, which was determined by substituting the differential absorption coefficients derived in Examples 1 to 5 into relational formula (2) of the present invention, and the measurement results of the carbon concentration in the epitaxial film formed on the silicon single crystal by SIMS.
[0040] Example 2 In the step 1, the carbon concentration in the epitaxial film is 2.0×10 18 atoms / cm 3 An epitaxial wafer was manufactured under the same conditions as in Example 1, except that the target was the same as that of Example 1, and this was used as an evaluation sample. As a reference sample, a 5.5 μm thick epitaxial film containing substantially no carbon was formed on a silicon single crystal substrate of the same standard as the evaluation sample, and this was used as the reference sample.
[0041] In the step 2, the absorbance spectra of the evaluation sample and the reference sample were measured using an FT-IR device, and the absorbance spectrum of the reference sample was subtracted from the absorbance spectrum of the evaluation sample to obtain a differential absorbance spectrum. -1 The difference absorbance was calculated by reading the distance between the localized vibrational absorption peak of Cs observed in the vicinity and the baseline, and it was found to be approximately 0.0031.
[0042] In step 3, the differential absorption coefficient was calculated from the differential absorbance obtained in step 2 and the film thickness of the epitaxial film formed on the silicon single crystal (5.1 μm (determined from the SIMS profile)) (the differential absorption coefficient at this time was approximately 13.9). As a result of substituting the differential absorption coefficient into relational expression (2) of the present invention, the carbon concentration in the epitaxial film formed on the silicon single crystal was approximately 1.7×10 18 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 1.4 × 10 18 atoms / cm 3 Therefore, the error was approximately 20% (see Figure 6).
[0043] Example 3 In the step 1, the carbon concentration in the epitaxial film is 7.0×10 18 atoms / cm 3An epitaxial wafer was manufactured under the same conditions as in Example 1, except that the target was the same as that of Example 1, and this was used as an evaluation sample. As a reference sample, a 5.5 μm thick epitaxial film containing substantially no carbon was formed on a silicon single crystal substrate of the same standard as the evaluation sample, and this was used as the reference sample.
[0044] In the step 2, the absorbance spectra of the evaluation sample and the reference sample were measured using an FT-IR device, and the absorbance spectrum of the reference sample was subtracted from the absorbance spectrum of the evaluation sample to obtain a differential absorbance spectrum. -1 The difference in absorbance was calculated by reading the distance between the localized vibrational absorption peak of Cs observed in the vicinity and the baseline, and it was found to be approximately 0.011.
[0045] In step 3, the differential absorption coefficient was calculated from the differential absorbance obtained in step 2 and the film thickness of the epitaxial film formed on the silicon single crystal (5.1 μm (determined from the SIMS profile)) (the differential absorption coefficient at this time was approximately 48.2). As a result of substituting the differential absorption coefficient into relational expression (2) of the present invention, the carbon concentration in the epitaxial film formed on the silicon single crystal was approximately 5.8×10 18 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 5.1 × 10 18 atoms / cm 3 Therefore, the error was approximately 13% (see Figure 6).
[0046] Example 4 In the step 1, the carbon concentration in the epitaxial film is 2.0×10 19 atoms / cm 3 An epitaxial wafer was manufactured under the same conditions as in Example 1, except that the target was the same as that of Example 1, and this was used as an evaluation sample. As a reference sample, a 5.5 μm thick epitaxial film containing substantially no carbon was formed on a silicon single crystal substrate of the same standard as the evaluation sample, and this was used as the reference sample.
[0047] In the step 2, the absorbance spectra of the evaluation sample and the reference sample were measured using an FT-IR device, and the absorbance spectrum of the reference sample was subtracted from the absorbance spectrum of the evaluation sample to obtain a differential absorbance spectrum. -1 The difference absorbance was calculated by reading the distance between the localized vibrational absorption peak of Cs observed in the vicinity and the baseline, and it was found to be approximately 0.028.
[0048] In step 3, the differential absorption coefficient was calculated from the differential absorbance obtained in step 2 and the film thickness of the epitaxial film formed on the silicon single crystal (5.1 μm (determined from the SIMS profile)) (the differential absorption coefficient at this time was approximately 125.9). As a result of substituting the differential absorption coefficient into relational expression (2) of the present invention, the carbon concentration in the epitaxial film formed on the silicon single crystal was approximately 1.5×10 19 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 1.5 × 10 19 atoms / cm 3 Therefore, the error was approximately 0% (see Figure 6).
[0049] Example 5 In the step 1, the carbon concentration in the epitaxial film is 1.0×10 20 atoms / cm 3 An epitaxial wafer was manufactured under the same conditions as in Example 1, except that the target thickness and film thickness were 1 μm, and this was used as an evaluation sample. As a reference sample, a 1 μm thick epitaxial film that was substantially free of carbon was formed on a silicon single crystal substrate of the same standard as the evaluation sample, and this was used as the reference sample.
[0050] In the step 2, the absorbance spectra of the evaluation sample and the reference sample were measured using an FT-IR device, and the absorbance spectrum of the reference sample was subtracted from the absorbance spectrum of the evaluation sample to obtain a differential absorbance spectrum. -1 The difference absorbance was calculated by reading the distance between the localized vibrational absorption peak of Cs observed in the vicinity and the baseline, and it was found to be approximately 0.027.
[0051] In step 3, the differential absorption coefficient was calculated from the differential absorbance obtained in step 2 and the film thickness of the epitaxial film formed on the silicon single crystal (1.1 μm (determined from the SIMS profile)) (the differential absorption coefficient at this time was approximately 558.2). As a result of substituting the differential absorption coefficient into relational expression (2) of the present invention, the carbon concentration in the epitaxial film formed on the silicon single crystal was approximately 6.7 × 10 19 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 8.7 × 10 19 atoms / cm 3 Therefore, the error was approximately 23% (see Figure 6).
[0052] (Comparative Example 1) In the same evaluation sample as in Example 1, the carbon concentration was estimated by substituting the differential absorption coefficient indicated by the FT-IR method into the relational expression (1) in Non-Patent Document 1. The carbon concentration in the epitaxial film formed on the silicon single crystal was found to be approximately 2.7 × 10 17 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 5.0 × 10 17 atoms / cm 3 Therefore, the error was approximately 46% (see Figure 7).
[0053] (Comparative Example 2) In the same evaluation sample as in Example 2, the carbon concentration was estimated by substituting the differential absorption coefficient indicated by the FT-IR method into the relational expression (1) in Non-Patent Document 1. The carbon concentration in the epitaxial film formed on the silicon single crystal was found to be approximately 1.1 × 10 18 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 1.4 × 10 18 atoms / cm 3 Therefore, the error was approximately 21% (see Figure 7).
[0054] (Comparative Example 3) In the same evaluation sample as in Example 3, the carbon concentration was estimated by substituting the differential absorption coefficient indicated by the FT-IR method into the relational expression (1) in Non-Patent Document 1. The carbon concentration in the epitaxial film formed on the silicon single crystal was found to be approximately 4.0 × 10 18 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 5.1 × 10 18 atoms / cm 3 Therefore, the error was approximately 23% (see Figure 7).
[0055] Comparative Example 4 In the same evaluation sample as in Example 4, the carbon concentration was estimated by substituting the differential absorption coefficient indicated by the FT-IR method into the relational expression (1) in Non-Patent Document 1. The carbon concentration in the epitaxial film formed on the silicon single crystal was found to be approximately 1.0 × 10 19 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 1.5 × 10 19 atoms / cm 3 Therefore, the error was approximately 33% (see Figure 7).
[0056] (Comparative Example 5) In the same evaluation sample as in Example 5, the carbon concentration was estimated by substituting the differential absorption coefficient indicated by the FT-IR method into the relational expression (1) in Non-Patent Document 1. The carbon concentration in the epitaxial film formed on the silicon single crystal was found to be approximately 4.6 × 10 19 atoms / cm 3 The carbon concentration in the epitaxial film formed on the silicon single crystal, as determined by SIMS, was approximately 8.7 × 10 19 atoms / cm 3 Therefore, the error was approximately 47% (see Figure 7).
[0057] As described above, according to Examples 1 to 5 of the present invention, it was found that the error between the carbon concentration calculated using the calculation formula (2) of the present invention and the carbon concentration measured by the SIMS method was about 20% or less, as shown in Fig. 6. Furthermore, when the relational formula (1) of Non-Patent Document 1, which is used as an existing FT-IR method, was applied (Comparative Examples 1 to 5), it was found that the carbon concentration measured by the SIMS method deviated significantly, as shown in Fig. 7. From the above, the carbon concentration is 5 × 10 17 atoms / cm 3 More than 1×10 20 atoms / cm 3 When measuring the substitutional carbon concentration in a silicon single crystal by FT-IR within the following range, it has been confirmed that by using the carbon concentration measurement method of the present invention, it is possible to perform the measurement more simply, accurately, and non-destructively without significant deviation from the carbon concentration determined by the highly accurate SIMS method.
[0058] The present invention is not limited to the above-described embodiments. The above-described embodiments are merely examples, and anything that has substantially the same configuration as the technical idea described in the claims of the present invention and that exhibits similar effects is included within the technical scope of the present invention.
Claims
1. A method for measuring a substitutional carbon concentration in a silicon single crystal by an FT-IR method, comprising the steps of: The effective range of the carbon concentration measurement is 5×10 17 atoms / cm 3 Above 1 x 10 20 atoms / cm 3 The concentration is A differential absorption coefficient is calculated from the differential absorbance obtained from the localized vibration absorption peak of substitutional carbon in a differential absorbance spectrum obtained by subtracting an absorbance spectrum obtained by irradiating an infrared light onto a reference silicon single crystal containing no carbon from an absorbance spectrum obtained by irradiating an infrared light onto the silicon single crystal to be measured, and the thickness of the measurement object; [Carbon concentration (atoms / cm 3 ) = 1.2 x 10 17 × [Differential absorption coefficient] The carbon concentration of the silicon single crystal to be measured is calculated using the following relational expression:
2. 2. The method for measuring a carbon concentration according to claim 1, wherein the carbon concentration of the silicon single crystal to be measured is a carbon concentration in an epitaxial layer of a silicon epitaxial wafer.
3. The differential absorption coefficient is [Differential absorption coefficient] = [Differential absorbance] × 23.03 ÷ [Thickness of object to be measured (mm)] 3. The method for measuring carbon concentration according to claim 1, wherein the carbon concentration is calculated using the following relational expression:
Citation Information
Patent Citations
Method for determining content of substituted carbon in monocrystal silicon through low-temperature infrared spectrum
CN103712946A
Method for detecting content of substitutional carbon and interstitial oxygen in monocrystalline silicon
CN105092511A
Method for producing substitutionally carbon-doped crystalline si-containing materials by chemical vapor deposition
JP2008530784A
Epitaxial wafer and method of producing the same
JP2009164590A
Method and device for measuring carbon concentration in silicon single crystal
JP2020012772A