Method for evaluating the substitutional Pt concentration and vacancy concentration of a silicon single crystal wafer

Abstract

This invention provides a simple and accurate method for evaluating the concentration of substitutional Pt in a silicon wafer using lifetime measurement and DLTS measurement. [Solution] A method for evaluating the substitutional Pt concentration of a silicon single crystal wafer, comprising: step 1, intentionally contaminating a silicon single crystal wafer for calibration curve creation by varying the concentration of Pt; step 2, performing a heat treatment to diffuse the Pt; step 3, measuring the lifetime; step 4, measuring the substitutional Pt concentration by the DLTS method; step 5, creating a calibration curve from the lifetime and substitutional Pt concentration; step 6, intentionally contaminating the Pt; step 7, performing a heat treatment to diffuse the Pt; step 8, measuring the lifetime; and step 9, estimating the in-plane distribution of substitutional Pt concentration using the calibration curve, wherein in step 2, the formation mechanism of substitutional Pt formed at least after the diffusion heat treatment is the same as the formation mechanism of substitutional Pt to be evaluated in step 7, and the method is performed in a temperature range under these conditions.

Description

[Technical Field]

[0001] The present invention relates to a method for evaluating the substitution type Pt concentration and a method for evaluating the vacancy concentration of a silicon single crystal wafer. [Background technology]

[0002] Platinum (Pt) diffusion into silicon single-crystal wafers is used for lifetime control and vacancy concentration evaluation. In lifetime control, substitutional Pt forms deep energy levels in the silicon band gap, promoting carrier recombination and reducing lifetime. In vacancy concentration evaluation, the reaction in which substitutional Pt is formed by the diffusion of Pt into atomic vacancies in silicon is used, and the vacancy concentration is evaluated from the substitutional Pt concentration. Therefore, accurately evaluating the substitutional Pt concentration in silicon wafers is particularly important for vacancy concentration evaluation.

[0003] One method for evaluating the concentration of metals such as Pt in silicon wafers is to assess the concentration by measuring the deep energy levels created by impurity metals using DLTS measurement. However, DLTS measurement is time-consuming and does not allow for map-based measurement. Patent document 1 proposes a method for evaluating the concentration of metal impurities from a calibration curve using DLTS measurement or lifetime measurement. However, when the metal impurity is Pt, the diffusion behavior of Pt during heat treatment is complex, and non-patent document 1 reports that the formation mechanism of substitutional Pt differs depending on the heat treatment temperature.

[0004] When diffusing Pt into silicon, if the diffusion temperature is below 700°C, substitutional Pt is formed by Pt entering vacancies, a mechanism called the Frank-Turnbull mechanism. For evaluating vacancy concentration using Pt diffusion, heat treatment of Pt diffusion where the Frank-Turnbull mechanism is dominant is used. On the other hand, when the diffusion temperature is above 700-800°C, the mechanism of substitutional Pt formation, which involves the ejection of silicon at lattice positions and the generation of interstitial silicon, becomes dominant, a mechanism called the kick-out mechanism. For lifetime control using Pt diffusion, heat treatment of Pt diffusion where the kick-out mechanism is dominant is generally used. Basic research on lifetime control using Pt diffusion revealed that the correlation between lifetime and substitutional Pt concentration differs depending on the substitutional Pt formation mechanism. The metal concentration evaluation method in Patent Document 1 does not mention or address the substitutional Pt formation mechanism, resulting in a problem of poor accuracy in evaluation using calibration curves. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2017-199775 [Non-patent literature]

[0006] [Non-Patent Document 1] H. Zimmermann at el.,Phys.Rev.B 44,9064(1991). [Overview of the project] [Problems that the invention aims to solve]

[0007] As described above, conventional methods have been used to evaluate the metal impurity concentration of silicon single-crystal wafers using calibration curves. However, when the metal impurity is Pt, there is a problem in that the accuracy of the evaluation is poor due to the existence of multiple formation mechanisms for substitutional Pt.

[0008] This invention was made to solve the above problems, and aims to provide a simple and accurate method for evaluating the substitutional Pt concentration in a silicon wafer using lifetime measurement and DLTS measurement. [Means for solving the problem]

[0009] To solve the above problems, the present invention provides: A method for evaluating the concentration of substitutional Pt in a silicon single crystal, Step 1 involves intentionally contaminating a silicon single crystal wafer for calibration curve preparation with Pt by varying its concentration, Step 2 involves performing a heat treatment to diffuse Pt into the silicon single crystal wafer used for calibration curve preparation, Step 3 involves measuring the lifetime of the silicon single crystal wafer used for calibration curve preparation in which the aforementioned Pt is diffused. Step 4, after the lifetime measurement, is performed to measure the displacement type Pt concentration by the DLTS method. Step 5 involves creating a calibration curve from the measured lifetime and substitution Pt concentration. Step 6 involves intentionally contaminating the silicon single-crystal wafer to be evaluated with Pt, Step 7 involves performing a heat treatment to diffuse Pt after the aforementioned intentional contamination, Step 8 involves measuring the lifetime of the silicon single crystal wafer to be evaluated, Step 9 involves using the calibration curve created above to estimate the in-plane distribution of substitutional Pt concentration from the lifetime of the silicon single crystal wafer to be evaluated, Includes, The present invention provides a method for evaluating the substitutional Pt concentration of a silicon single-crystal wafer, characterized in that, in step 2, a heat treatment is performed to diffuse Pt into the silicon single-crystal wafer used for calibration curve creation, the treatment is carried out in a temperature range under conditions such that the formation mechanism of substitutional Pt formed after the diffusion heat treatment is the same as the formation mechanism of substitutional Pt formed after the diffusion heat treatment of the silicon single-crystal wafer to be evaluated in step 7.

[0010] With such an evaluation method for the substitutional Pt concentration of a silicon single crystal wafer, the in-plane distribution of the substitutional Pt concentration of the silicon single crystal wafer can be easily and accurately evaluated.

[0011] In addition, in the step 2, diffusion heat treatment is performed in two temperature regions, a temperature region where the substitutional Pt formation mechanism is dominated by the Frank-Turnbull mechanism and a temperature region where the kick-out mechanism is dominant. In the step 5, calibration curves for the Frank-Turnbull mechanism dominance and the kick-out mechanism dominance are respectively created, and in the step 9, it is preferable to use a calibration curve corresponding to the formation mechanism of substitutional Pt formed after diffusion heat treatment of the silicon single crystal wafer to be evaluated.

[0012] Such an evaluation method is preferable because the coefficient of determination of the calibration curve is high and the accuracy is high.

[0013] In the step 9, with the heat treatment time being t [h] and the heat treatment temperature being T [°C], if t·exp[(T - 800) / 10] ≥ 0.25, a calibration curve of the kick-out mechanism dominance type is used, and if t·exp[(T - 800) / 10] < 0.25, a calibration curve of the Frank-Turnbull mechanism dominance type is used to obtain the substitutional Pt concentration from the lifetime.

[0014] With such an evaluation method, the substitutional Pt concentration in the silicon wafer can be easily and accurately evaluated.

[0015] In the step 6 in the evaluation method for the substitutional Pt concentration of the silicon single crystal wafer, when the Pt contamination concentration is in the range of 1×10 13 atoms / cm 2 or more and 1×10 14 atoms / cm 2 or less, and in the step 7, when it is a Pt diffusion heat treatment dominated by the Frank-Turnbull mechanism satisfying t·exp[(T - 800) / 10] < 0.25, the in-plane distribution of the vacancy concentration can be estimated from the substitutional Pt concentration after the step 9.

[0016] The vacancy concentration can be estimated using this method of evaluating vacancy concentration. [Effects of the Invention]

[0017] As described above, the method for evaluating the substitutional Pt concentration in a silicon single-crystal wafer according to the present invention allows for simple and accurate evaluation of the substitutional Pt concentration in a silicon wafer without being affected by the differing correlation between substitutional Pt concentration and lifetime due to the different formation mechanisms of substitutional Pt depending on the heat treatment temperature. Furthermore, if the Pt diffusion heat treatment is predominantly the Frank-Turnbull mechanism, the in-plane distribution of vacancy concentration can be estimated from the substitutional Pt concentration. Also, if the kick-out mechanism is dominant, guidelines for conducting lifetime control tests can be established from the correlation between lifetime and substitutional Pt concentration. [Brief explanation of the drawing]

[0018] [Figure 1] This flowchart shows an example of a method for evaluating the concentration of substitutional Pt according to the present invention. [Figure 2] This is a correlation equation (calibration curve) showing the correlation between lifetime measurements by the μ-PCD method and the substitution type Pt concentration measured by DLTS, which were prepared in the examples and comparative examples. [Figure 3] This is an example of the in-plane distribution of substitutional Pt concentration evaluated from lifetime using a correlation formula. [Figure 4] This is an example of the in-plane distribution of vacancy concentration evaluated from lifetime using a correlation equation. [Modes for carrying out the invention]

[0019] As mentioned above, methods for evaluating the metal impurity concentration of silicon single-crystal wafers from calibration curves have been used. However, when the metal impurity is Pt, the accuracy of the evaluation suffers due to the existence of multiple formation mechanisms for substitutional Pt. Therefore, there has been a need for the development of a simple and accurate method for evaluating substitutional Pt concentration.

[0020] As a result of diligent research into the above-mentioned problems, the inventors have discovered a method for easily and accurately evaluating the substitutional Pt concentration in a silicon wafer without being affected by the differing correlation between substitutional Pt concentration and lifetime, which is caused by different formation mechanisms of substitutional Pt depending on the heat treatment temperature. This led to the completion of the present invention.

[0021] In other words, the present invention is a method for evaluating the substitutional Pt concentration in a silicon single crystal, Step 1 involves intentionally contaminating a silicon single crystal wafer for calibration curve preparation with Pt by varying its concentration, Step 2 involves performing a heat treatment to diffuse Pt into the silicon single crystal wafer used for calibration curve preparation, Step 3 involves measuring the lifetime of the silicon single crystal wafer used for calibration curve preparation in which the aforementioned Pt is diffused. Step 4, after the lifetime measurement, is performed to measure the displacement type Pt concentration by the DLTS method. Step 5 involves creating a calibration curve from the measured lifetime and substitution Pt concentration. Step 6 involves intentionally contaminating the silicon single-crystal wafer to be evaluated with Pt, Step 7 involves performing a heat treatment to diffuse Pt after the aforementioned intentional contamination, Step 8 involves measuring the lifetime of the silicon single crystal wafer to be evaluated, Step 9 involves using the calibration curve created above to estimate the in-plane distribution of substitutional Pt concentration from the lifetime of the silicon single crystal wafer to be evaluated, Includes, The method for evaluating the substitutional Pt concentration of a silicon single-crystal wafer is characterized in that, in step 2, a heat treatment is performed to diffuse Pt into the silicon single-crystal wafer used for creating the calibration curve, and the treatment is performed in a temperature range under conditions such that the formation mechanism of substitutional Pt formed after the diffusion heat treatment is the same as the formation mechanism of substitutional Pt formed after the diffusion heat treatment of the silicon single-crystal wafer to be evaluated in step 7.

[0022] The present invention will be described in detail below, but the present invention is not limited to these descriptions.

[0023] The following will provide a detailed explanation with reference to the diagrams.

[0024] Figure 1 is a flowchart showing an example of the evaluation method for substitution type Pt concentration according to the present invention.

[0025] [Method for evaluating the concentration of substitution-type Pt] The method for evaluating the substitutional Pt concentration of the present invention, as shown in Figure 1, is broadly divided into a preliminary test and a main test. The preliminary test is the step of creating a calibration curve for lifetime and substitutional Pt concentration using a calibration wafer. It mainly consists of the steps of preparing the calibration wafer, intentionally contaminating the calibration wafer with Pt and Pt diffusion, measuring lifetime and DLTS, and creating a calibration curve for each temperature range in which the respective substitutional Pt formation mechanism is dominant. The main test is the step of evaluating the in-plane distribution of substitutional Pt concentration from the lifetime of the wafer to be evaluated using the created calibration curve. It mainly consists of the steps of preparing the wafer to be evaluated, intentionally contaminating the wafer with Pt, and measuring the lifetime of the wafer to be evaluated.

[0026] The following details each step in each stage. The explanation will use the step numbers (S11-S16 and S21-S25) shown in Figure 1. If there are corresponding step numbers (Steps 1-9) in Claim 1, these step numbers will also be included in the explanation.

[0027] <Preliminary Examination> First, let me explain the preliminary examination.

[0028] (S11: Preparation of wafers for calibration curve creation) Step S11 in Figure 1 is the process of preparing wafers for calibration curve creation.

[0029] A calibration wafer is a silicon single-crystal wafer. For example, if the evaluation wafer used in the main test described later is known, it can be a silicon single-crystal wafer manufactured under the same wafer manufacturing conditions as the main test, with similar concentrations of light elements such as hydrogen, carbon, nitrogen, and oxygen, and similar resistivity. Since the lifetime can change due to the influence of light element concentrations and resistivity, even in materials other than substitutional Pt, using a calibration wafer with similar conditions allows for more accurate measurements in the main test.

[0030] Here, the method for manufacturing the single crystal for the calibration curve wafer is not particularly limited. A wafer manufactured by the Czochralski method (hereinafter referred to as the CZ method) may be used, or a wafer manufactured by the Floating Zone Method (hereinafter referred to as the FZ method) may be used.

[0031] Furthermore, there are no particular restrictions on the concentrations of light elements such as hydrogen, carbon, nitrogen, and oxygen.

[0032] The resistivity is preferably 1 to 100 Ω·cm.

[0033] (S12: Pt contamination treatment of wafers for calibration curve creation) Step S12 in Figure 1 is the process of intentionally contaminating the calibration wafer with Pt (Step 1).

[0034] There are no particular restrictions on the method used to intentionally contaminate the wafer for calibration curve preparation. Vacuum deposition, sputtering, or spin coating may be used. There are also no particular restrictions on the vacuum deposition method; resistance heating or electron beam methods may be used. For spin coating, a Pt contamination solution obtained by diluting an atomic absorption standard solution with pure water is used, and its concentration is preferably 100 ppb to 1000 ppm. Furthermore, the Pt surface concentration after Pt contamination can be measured by total reflection X-ray fluorescence (TXRF) analysis.

[0035] (S13: Pt diffusion heat treatment of wafers for calibration curve creation) Step S13 in Figure 1 is the process of performing a heat treatment to diffuse Pt into the wafer for calibration curve creation (Step 2).

[0036] A vertical or horizontal furnace may be used for the diffusion heat treatment furnace. Either a single-wafer or batch furnace is acceptable. The heat treatment temperature may be 500°C to 1000°C, and the heat treatment time may be 3 minutes to 32 hours. In this case, it is preferable to perform the diffusion heat treatment in two different temperature ranges: one where the Frank-Turnbull mechanism is dominant below 700°C, and another where the kick-out mechanism is dominant above 700°C. The atmosphere during heat treatment is preferably an inert gas, such as nitrogen (N2), helium (He), neon (Ne), argon (Ar), krypton (Kr), xenon (Xe), or radon (Rn).

[0037] As for specific heat treatment conditions, for example, for Pt diffusion heat treatment conditions where the Frank-Turnbull mechanism is dominant, it is preferable to have a heat treatment temperature of 650°C, a heat treatment time of 8 hours, and a nitrogen atmosphere, while for Pt diffusion heat treatment conditions where the kick-out mechanism is dominant, it is preferable to have a heat treatment temperature of 800°C, a heat treatment time of 8 hours, and a nitrogen atmosphere.

[0038] (S14: Lifetime measurement of wafers used for calibration curve creation) Step S14 in Figure 1 is the process of measuring the lifetime of the wafer used for calibration curve creation (Step 3).

[0039] The μ-PCD (Microwave Photo Conductivity Decay) method can be used to measure the lifetime of wafers used for calibration curve creation.

[0040] The principle of measuring lifetime using the μPCD method will be explained.

[0041] When a silicon single-crystal wafer is irradiated with light exceeding its band gap, electrons in the valence band are excited into the conductor, generating excess carriers. These excess carriers increase conductivity, thus increasing microwave reflectivity. When light irradiation is stopped, the excess carriers decrease due to recombination. As a result, conductivity decreases, and microwave reflectivity declines. The μ-PCD method utilizes the time evolution of microwave reflectivity and measures the lifetime from the decay curve.

[0042] In this case, if dangling bonds are present on the surface, excess carriers may recombine at the dangling bonds on the surface, making it impossible to measure the bulk lifetime. To measure the bulk lifetime, it is necessary to terminate the dangling bonds on the surface. This can be done by chemical passivation through immersion in an iodine alcohol solution, or by forming an oxide film on the surface through oxidative heat treatment and then performing passivation.

[0043] (S15: DLTS measurement of wafers for calibration curve creation) Step 4 in Figure 1 is the process of measuring the displacement Pt concentration of the calibration wafer using the DLTS (Deep Level Transient Spectroscopy) method.

[0044] The measurement principle of the DLTS method will be explained.

[0045] [Measurement of the time evolution of junction capacitance] DLTS (Deep Layer Transistor Tectonics) utilizes the depletion layer formed in Schottky junctions and pn junctions for measurement. When a forward bias is applied to these junctions, the Fermi level rises, and electrons are trapped in the lower-energy electron trap levels. Subsequently, when a reverse bias is instantaneously applied, the Fermi level falls, and the electrons trapped are released into the conductor. Since the traps that have released electrons become positively charged, the depletion layer width decreases to compensate for this charge, and the junction capacitance increases. Therefore, by measuring the time change of the junction capacitance when a reverse bias is applied, the process of electron emission from the trap levels can be observed.

[0046] [Acquisition of DLTS spectrum by temperature scan] This explains the principle of measuring DLTS spectra using Lang's method.

[0047] Lang's method measures the junction capacitances C1 and C2 at two times t1 and t2, and calculates the difference ΔC = C1 - C2. By measuring ΔC while varying the temperature, a DLTS spectrum with a peak at the temperature at which electron emission from the trap is thermally activated can be obtained.

[0048] The DLTS spectra measured for substitutional Pt in single-crystal silicon differ depending on the conductivity type of silicon. For substitutional Pt in p-type single-crystal silicon, a DLTS spectrum with one peak around 180K is observed in the temperature range of 77K to 300K. This peak around 180K is attributed to substitutional Pt. On the other hand, for substitutional Pt in n-type single-crystal silicon, a DLTS spectrum with five peaks around 100K, 135K, 170K, 250K, and 270K is observed in the temperature range of 77K to 300K. Of these, the peak around 135K is suggested to be attributed to substitutional Pt, and the peak around 270K is suggested to be attributed to a Pt-oxygen complex. The other peaks around 100K, 170K, and 250K are of unknown origin.

[0049] (S16: Creation of calibration curves for each temperature range of the substitutional Pt formation mechanism) Step S16 in Figure 1 is the process of creating a calibration curve for lifetime and substitutional Pt concentration from the results of S14 and S15 (Step 5).

[0050] The diffusion behavior of Pt during heat treatment is complex, and it has been reported that the formation mechanism of substitutional Pt differs depending on the heat treatment temperature. When Pt is diffused into silicon, at diffusion temperatures below 700°C, substitutional Pt is formed by Pt entering vacancies, which is called the Frank-Turnbull mechanism. On the other hand, when the diffusion temperature is 700-800°C or higher, the formation mechanism of substitutional Pt accompanied by the generation of interstitial silicon by ejecting silicon at lattice positions becomes dominant, which is called the kick-out mechanism. Since the correlation between lifetime and substitutional Pt differs between the Frank-Turnbull mechanism and the kick-out mechanism, it is preferable to create calibration curves for each. For example, in S13: Wafer Pt diffusion heat treatment for calibration curve creation (step 2), the diffusion temperature can be set to 650°C and the diffusion time to 8 hours for the Frank-Turnbull mechanism, and the diffusion temperature can be set to 800°C and the diffusion time to 8 hours for the kick-out mechanism. Thus, it is preferable to create calibration curves in temperature ranges where each substitutional Pt formation mechanism is dominant.

[0051] <Main Exam> (S21: Preparation of wafers to be evaluated) Step S21 in Figure 1 is the process of preparing the wafer to be evaluated.

[0052] Regarding the wafers being evaluated, the manufacturing conditions and light element concentrations are the same as those for the wafers used to create the calibration curve.

[0053] (S22: Pt contamination treatment of wafers to be evaluated) Step S22 in Figure 1 is the process of intentionally contaminating the wafer to be evaluated with Pt (Step 6).

[0054] This method of intentionally contaminating Pt is the same as the Pt contamination treatment of the wafer used for calibration curve creation (S12, step 1).

[0055] (S23: Pt diffusion heat treatment of the wafer to be evaluated) Step S23 in Figure 1 is the process of performing a heat treatment to diffuse Pt into the wafer to be evaluated (Step 7).

[0056] The heat treatment temperature is preferably in the range of 500°C to 1000°C, and the heat treatment time is preferably in the range of 3 minutes to 32 hours. The substitutional Pt formation mechanism and heat treatment conditions are determined according to the evaluation objective.

[0057] Furthermore, if t·exp[(T-800) / 10] < 0.25, the Pt diffusion heat treatment will be predominantly Frank-Turnbull mechanism-driven, and the vacancy concentration can be evaluated from the substitutional Pt concentration after S25 (step 9) described later. Also, if t·exp[(T-800) / 10] ≥ 0.25, the Pt diffusion heat treatment will be predominantly kick-out mechanism-driven, and guidelines can be established for lifetime control testing based on the correlation between lifetime and substitutional Pt concentration.

[0058] For example, in the case of Pt diffusion heat treatment conditions where the Frank-Turnbull mechanism is dominant, it is preferable to set the heat treatment temperature to 650°C, the heat treatment time to 8 hours, and the atmosphere to nitrogen, and in the case of Pt diffusion heat treatment conditions where the kick-out mechanism is dominant, it is preferable to set the heat treatment temperature to 800°C, the heat treatment time to 8 hours, and the atmosphere to nitrogen.

[0059] (S24: Lifetime measurement of the wafer under evaluation) Step S24 in Figure 1 is the process of measuring the lifetime of the wafer being evaluated (Step 8).

[0060] This lifetime measurement is the same as the lifetime measurement of the calibration curve wafer (step 3) mentioned above.

[0061] (S25: Evaluation of the substitutional Pt concentration of the wafer under evaluation from the lifetime using a calibration curve) Step S25 in Figure 1 is the process of evaluating the substitutional Pt concentration of the wafer to be evaluated (Step 9).

[0062] The following formula is used to determine which of the two calibration curves created in the preliminary test should be used.

[0063] That is, assuming the heat treatment time of the Pt diffusion heat treatment conditions of the evaluation target wafer used in the above S23 (Step 7) is t [h] and the heat treatment temperature is T [°C], if t·exp[(T - 800) / 10] ≥ 0.25, a calibration curve of the kick-out mechanism dominant type is used; if t·exp[(T - 800) / 10] < 0.25, a calibration curve of the Frank-Turnbull mechanism dominant type is used, and the substitutional Pt concentration can be evaluated from the lifetime.

[0064] [Vacancy Concentration Evaluation Method] If the Pt contamination concentration in the above S22 (Step 6) is in the range of 1×10 13 atoms / cm 2 or more and 1×10 14 atoms / cm 2 or less, and in S23 (Step 7), if the diffusion heat treatment satisfies the condition of the Frank-Turnbull mechanism dominant type where t·exp[(T - 800) / 10] < 0.25, the in-plane distribution of the vacancy concentration can be further estimated from the substitutional Pt concentration.

[0065] The vacancy concentration can be evaluated from the following formula. [Equation]

[0066] Here, C V represents the vacancy concentration before the Pt diffusion heat treatment, C Pt represents the substitutional Pt concentration after the Pt diffusion heat treatment, C * V represents the equilibrium concentration of vacancies at the Pt diffusion temperature, and C * Pt represents the equilibrium concentration of substitutional Pt at the Pt diffusion temperature. At this time, substituting the reported value of Zimmermann at the diffusion temperature of 650°C for the equilibrium concentrations of vacancies and substitutional Pt, C V = 3.1C Pt is obtained (Non-Patent Document 1). Using this formula, the vacancy concentration can be estimated from the substitutional Pt concentration. [Examples]

[0067] The present invention will be specifically described below using examples and comparative examples, but the present invention is not limited to these.

[0068] [Example 1] The substitutional Pt concentration measurement method of the present invention was carried out by performing the following preliminary and main tests, as shown in Figure 1.

[0069] <Preliminary Examination> First, a silicon single-crystal wafer was prepared for creating the calibration curve for the preliminary test (S11). The single-crystal manufacturing method, conductivity type, diameter, crystal plane orientation, and resistivity of the prepared calibration wafer are as follows. Single crystal manufacturing method: FZ method Conductivity type of substrate: n-type Diameter: 200mm Crystal plane orientation: (100) Resistivity: 50Ω cm

[0070] Next, the prepared calibration wafers were immersed in hydrofluoric acid to remove the native oxide film, and then coated using a Pt contamination solution obtained by diluting an atomic absorption standard solution with pure water by spin coating (S12, step 1). Five levels of Pt concentration were used in the contamination solution: 100 ppb, 1 ppm, 10 ppm, 100 ppm, and 1000 ppm. Subsequently, to ensure that the Frank-Turnbull mechanism was dominant for substitutional Pt formation, Pt diffusion heat treatment was performed with a heat treatment temperature / heat treatment time / atmosphere of 650°C / 8 hours / nitrogen atmosphere (S13, step 2). Furthermore, to ensure that the kick-out mechanism was dominant, Pt diffusion heat treatment was performed on the calibration wafers with a heat treatment temperature / heat treatment time / atmosphere of 800°C / 8 hours / nitrogen atmosphere (S13, step 2). The lifetime of the calibration wafer with diffused Pt was measured using the μ-PCD method, and the displacement Pt concentration was measured using the DLTS method (S14 and S15, steps 3 and 4). A calibration curve was created from the measured lifetime and the measured displacement Pt concentration using the DLTS method (S16, step 5).

[0071] The calibration curves created are shown in Figure 2. The horizontal axis represents the lifetime measured by the μ-PCD method, and the vertical axis represents the displacement Pt concentration measured by DLTS. Measurement points when the heat treatment temperature / heat treatment time / atmosphere was 650°C / 8 hours / nitrogen atmosphere are indicated by square markers, and measurement points when the heat treatment temperature / heat treatment time / atmosphere was 800°C / 8 hours / nitrogen atmosphere are indicated by circular markers. Calibration curves created by distinguishing between the Frank-Turnbull mechanism and the kick-out mechanism are shown as solid lines. Calibration curves created without distinguishing between kick-out mechanism dominant and Frank-Turnbull mechanism dominant data are shown as dotted lines. In the main test of Example 1, the diffusion heat treatment conditions for the kick-out mechanism dominant type were determined, so the calibration curve for the kick-out mechanism dominant type was used. In Example 2, which will be described later, the diffusion heat treatment conditions for the Frank-Turnbull mechanism dominant type were determined, so the calibration curve for the Frank-Turnbull mechanism dominant type was used. In Comparative Examples 1 and 2, calibration curves were used that did not distinguish between data where the kick-out mechanism was dominant and data where the Frank-Turnbull mechanism was dominant.

[0072] <Main Exam> The evaluation wafer for this test was the same as the wafer used for calibration curve creation (S21). The evaluation wafer was intentionally contaminated with Pt using a 100 ppb diluted contamination solution applied by spin coating (S22, step 6). Subsequently, Pt diffusion heat treatment was performed with a heat treatment temperature / time / atmosphere of 800°C / 8 hours / nitrogen atmosphere (S23, step 7). The lifetime of the evaluation wafer with diffused Pt was measured by the μ-PCD method (S24, step 8). The in-plane distribution of substitutional Pt concentration was estimated from the lifetime using the calibration curve (S25, step 9). At this time, the heat treatment temperature / time for the Pt diffusion heat treatment conditions performed in S23 (step 7) was 800°C / 8 hours, and since 8·exp[(800-800) / 10]=8>0.25, a kick-out mechanism dominant calibration curve was used. Figure 3 shows the estimated in-plane distribution of substitutional Pt concentration, and Table 1 summarizes the results for Example 1 and the comparative example.

[0073] [Comparative Example 1] In Comparative Example 1, the method for measuring substitutional Pt from lifetime was carried out as follows.

[0074] Using a calibration curve created without distinguishing between data dominant by the kick-out mechanism and data dominant by the Frank-Turnbull mechanism obtained in the preliminary test of Example 1 described above, the substitutional Pt concentration was estimated by adapting it to the lifetime measured in the main test of Example 1. Table 1 shows the data obtained by estimating the substitutional Pt concentration from the lifetime using the method proposed in the present invention (Example 1) and the conventional method (Comparative Example 1) at the four measurement positions (A, B, C, and D) in Figure 3.

[0075] [Table 1]

[0076] [Example 2] In Example 2, the substitutional Pt concentration was estimated from the lifetime using the Frank-Turnbull mechanism-dominant calibration curve created in the preliminary test of Example 1, and then the vacancy concentration was estimated.

[0077] <Main Exam> The same silicon single-crystal wafer as described in the preliminary test of Example 1 was prepared as the wafer to be evaluated (S21). For intentional Pt contamination of the wafer to be evaluated, a 100 ppm diluted contamination solution was applied by spin coating (S22, step 6). Then, Pt diffusion heat treatment was performed with a heat treatment temperature / heat treatment time / atmosphere of 650°C / 8 hours / nitrogen atmosphere (S23, step 7). The lifetime of the Pt-diffused wafer to be evaluated was measured by the μ-PCD method (S24, step 8). The in-plane distribution of substitutional Pt concentration was estimated from the lifetime using a calibration curve (S25, step 9). At this time, the heat treatment temperature / heat treatment time for the Pt diffusion heat treatment conditions performed in S23 (step 7) was 650°C / 8 hours, and 8·exp[(650-800) / 10] ≈ 2.4×10⁻⁶ -6 Since the value is <0.25, a calibration curve predominantly based on the Frank-Turnbull mechanism was used.

[0078] The Pt contamination concentration in step S22 (process 6) of the main test in Example 2 was 1 × 10⁻⁶ 13 atoms / cm 2 Furthermore, the Pt diffusion heat treatment of S23 is 8·exp[(650-800) / 10]≈2.4×10 -6 Since the Frank-Turnbull mechanism is dominant, satisfying <0.25, the vacancy concentration can be evaluated from the substitutional Pt concentration. The in-plane distribution of vacancy concentration is C V =3.1C Pt The results estimated from the substitution type Pt concentration are shown in Figure 4.

[0079] [Comparative Example 2] In Comparative Example 2, the method for measuring substitutional Pt from lifetime was carried out as follows. In the preliminary test of Example 2 described above, a calibration curve was created without distinguishing between data where the kick-out mechanism was dominant and data where the Frank-Turnbull mechanism was dominant. This curve was used in Comparative Example 2 and adapted to the lifetime measured in the main test of Example 2 to estimate the substitutional Pt concentration. Table 2 shows the data obtained by estimating the substitutional Pt concentration from lifetime using the method proposed in the present invention (Example 2) and the conventional method (Comparative Example 2) at the four measurement positions (A, B, C, and D) in Figure 4. [Table 2]

[0080] In Example 1, the substitutional Pt concentration estimated from lifetime using a calibration curve predominantly representing the kick-out mechanism agreed well with the substitutional Pt concentration measured by the DLTS method, demonstrating the effectiveness of the created calibration curve. On the other hand, the calibration curves created in Comparative Examples 1 and 2, which did not distinguish between the kick-out mechanism and the Frank-Turnbull mechanism, deviated significantly from the measured values ​​by the DLTS method. Thus, as can be seen from Tables 1 and 2, estimating the substitutional Pt concentration from calibration curves created for each substitutional Pt formation mechanism, as in Examples 1 and 2, yields better accuracy. The reason for the poor accuracy of the comparative examples' methods is that they do not take into account the differences in the correlation between lifetime and substitutional Pt concentration depending on the substitutional Pt formation mechanism. From this, it can be seen that the method in the examples has a higher coefficient of determination of the calibration curve than the methods in Comparative Examples 1 and 2, indicating that it can evaluate with better accuracy.

[0081] It should be noted that the present invention is not limited to the embodiments described above. The embodiments described above are illustrative, and any configuration that has substantially the same technical idea as described in the claims of the present invention and achieves similar effects is included within the technical scope of the present invention.

Claims

1. A method for evaluating the concentration of substitutional Pt in a silicon single crystal, Step 1 involves intentionally contaminating a silicon single crystal wafer for calibration curve preparation with Pt by varying the concentration, Step 2 involves performing a heat treatment to diffuse Pt into the silicon single crystal wafer used for creating the calibration curve, Step 3 involves measuring the lifetime of the silicon single crystal wafer used for calibration curve preparation in which the aforementioned Pt is diffused. Step 4, after the lifetime measurement, is performed to measure the substitution type Pt concentration by the DLTS method, Step 5 involves creating a calibration curve from the measured lifetime and substitution Pt concentration. Step 6 involves intentionally contaminating the silicon single crystal wafer to be evaluated with Pt, Step 7 involves performing a heat treatment to diffuse Pt after the aforementioned intentional contamination, Step 8 involves measuring the lifetime of the silicon single crystal wafer to be evaluated, Step 9 involves using the calibration curve created above to estimate the in-plane distribution of substitutional Pt concentration from the lifetime of the silicon single crystal wafer to be evaluated, Includes, A method for evaluating the substitutional Pt concentration of a silicon single crystal wafer, characterized in that, in step 2, a heat treatment is performed to diffuse Pt into the silicon single crystal wafer for calibration curve preparation, the treatment is performed in a temperature range under conditions such that the formation mechanism of substitutional Pt formed after the diffusion heat treatment is the same as the formation mechanism of substitutional Pt formed after the diffusion heat treatment of the silicon single crystal wafer to be evaluated in step 7.

2. The method for evaluating the substitutional Pt concentration of a silicon single crystal wafer according to claim 1, characterized in that, in step 2, diffusion heat treatment is performed in two temperature regions: one in which the substitutional Pt formation mechanism is predominantly the Frank-Turnbull mechanism and the other in which the kick-out mechanism is predominant; in step 5, calibration curves for the Frank-Turnbull mechanism dominance and the kick-out mechanism dominance are created, respectively; and in step 9, a calibration curve corresponding to the substitutional Pt formation mechanism formed after the silicon single crystal wafer to be evaluated is used.

3. The method for evaluating the substitutional Pt concentration of a silicon single crystal wafer according to claim 2, characterized in that, in step 9, the heat treatment time is t [h] and the heat treatment temperature is T [°C], and if t・exp[(T-800) / 10] ≥ 0.25, a calibration curve of the kick-out mechanism dominant type is used, and if t・exp[(T-800) / 10] < 0.25, a calibration curve of the Frank-Turnbull mechanism dominant type is used to determine the substitutional Pt concentration from the lifetime.

4. In step 6 of the method for evaluating the substitutional Pt concentration of a silicon single crystal wafer according to any one of claims 1 to 3, the Pt contamination concentration is 1 × 10 １３ atoms / cm ２ The above 1 x 10 １４ atoms / cm ２ A method for evaluating the vacancy concentration of a silicon single-crystal wafer, characterized in that, when the Pt diffusion heat treatment is predominantly Frank-Turnbull mechanism, within the following range and satisfies t・exp[(T-800) / 10] < 0.25 in step 7, the in-plane distribution of vacancy concentration is estimated from the substitutional Pt concentration after step 9.

Images