Method for evaluating silicon single crystals, and method for manufacturing epitaxial silicon wafers.

Abstract

To provide a silicon single crystal evaluation method capable of determining whether a predetermined area of a grown silicon single crystal is an area, in which an epitaxial silicon wafer having high gettering ability and capable of restraining inferior quality of a semiconductor device can be created.SOLUTION: In a silicon single crystal evaluation method, an oxygen precipitation nucleus growth treatment step is performed to an evaluating epitaxial wafer having an evaluating wafer obtained from a straight body of a silicon single crystal grown by a Czochralski method, added with boron, and having resistivity of 10 mΩ cm or less and oxygen concentration of 12×1017 atoms / cm3 or more and 16×1017 atoms / cm3 or less. When a BMD density of the evaluating wafer of the evaluating epitaxial wafer is not less than a threshold, a predetermined area including an acquisition position of the evaluating wafer is determined as a rejection area, and when the density is less than the threshold, the predetermined area is determined as an acceptance area.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] This invention relates to a method for evaluating silicon single crystals and a method for manufacturing epitaxial silicon wafers. [Background technology]

[0002] Epitaxial silicon wafers (hereinafter sometimes referred to as "epiwafers") are widely used as substrate materials for semiconductor devices. Epiwafers are made by forming an epitaxial layer (hereinafter sometimes referred to as "epi layer") on the surface of a silicon wafer (hereinafter sometimes referred to as "wafer") cut from a silicon single crystal grown by the Czochralski method (hereinafter sometimes referred to as the "CZ method"). Because of the high crystal integrity of the epi layer, it is possible to manufacture high-quality and reliable semiconductor devices.

[0003] In recent years, there has been a demand for epitaxial wafers, which are p+ silicon wafers doped with a high concentration of boron and have an epitaxial layer formed on their surface, in order to prevent latch-up phenomena in device manufacturing processes and to prevent depletion layer expansion when voltage is applied around trenches when using trench-structured capacitors.

[0004] On the other hand, if the epitaxial layer of an epiwafer is contaminated with metallic impurities, the semiconductor device characteristics deteriorate. Therefore, there is a need for epiwafers that have a gettering effect in order to eliminate contamination of the epitaxial layer by metallic impurities. As a gettering technique, intrinsic gettering (IG), which captures impurity elements by utilizing oxygen-induced minute oxygen deposits (BMDs: Bulk Micro Defects) induced during the heat treatment of the device process, is generally employed.

[0005] However, the high-temperature heat treatment applied to the wafer during the epitaxial layer formation process causes the tiny oxygen precipitation nuclei inherent in the wafer to shrink and disappear. As a result, in subsequent device processes, sufficient BMDs, which act as gettering sources, cannot be induced within the wafer, leading to the problem that epitaxial wafers have low gettering ability. To solve these problems, Patent Document 1 reports a method for producing silicon single crystals with uniform and high-density BMD density throughout the entire straight body from top to bottom by controlling the manufacturing conditions in the stage of producing silicon single crystals by the CZ method. This is achieved by increasing the residence time of the growing silicon single crystal in the oxygen nucleation temperature range of around 650°C, and by ensuring that the oxygen concentration incorporated into the silicon single crystal is uniform.

[0006] Furthermore, Patent Document 2 reports that since boron increases the stability of oxygen precipitation nuclei, adding a high concentration of boron to silicon single crystals facilitates the formation of oxygen precipitates. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2010-208894 [Patent Document 2] Japanese Patent Publication No. 2017-24965 [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] In the manufacturing process of silicon single crystals, adjusting the oxygen concentration and thermal history of the silicon single crystal to form oxygen precipitation nuclei is an effective method, and the higher the boron concentration, the greater the increase in BMD density in the wafer. For this reason, the inventors conceived of providing a p++ silicon wafer with a higher concentration of boron added, specifically an epitaxial wafer in which an epitaxial layer is formed on a wafer with a resistivity of 10 mΩ·cm or less to which boron is added.

[0009] However, when heat treatment such as FLA (flash lamp annealing) is performed on an epi-wafer in which an epi-layer is formed on the surface of a wafer with a high oxygen concentration and a high concentration of boron added, it has been found that dislocations may occur starting from BMDs in the wafer, and dislocation defects reaching the epitaxial layer may occur. Through intensive research on the cause, the following findings were obtained.

[0010] Normally, since the BMDs in the wafer of an epi-wafer are small in size, the BMD density in the wafer cannot be directly measured. Therefore, to measure the BMD density in the wafer, generally, after performing a low-temperature heat treatment of 800 °C or lower on the epi-wafer to form oxygen precipitation nuclei in the wafer, a high-temperature heat treatment of 1000 °C or higher is performed to grow the oxygen precipitation nuclei and make them manifest as oxygen precipitates. Specifically, a two-step heat treatment is carried out, in which heat treatment is performed at a temperature of 800 °C for 3 hours and then continuously at a temperature of 1000 °C for 16 hours. Thereafter, for the epi-wafer subjected to this two-step heat treatment, the BMD density in the wafer is measured using an optical microscope, infrared scattering tomography, or the like.

[0011] However, according to the experiments of the present inventors, in the BMD density evaluation after performing the above-described two-step heat treatment on an epi-wafer, it has become clear that even though the BMD density observed in the wafer is almost at the same level, there are cases where dislocation defects occur in the epi-layer and cases where they do not occur.

[0012] Upon investigating this cause, it was found that the occurrence situation of dislocation defects changes depending on the position where the wafer is cut out from the silicon single crystal. In the process of growing the silicon single crystal, it was confirmed that dislocation defects may occur in the epitaxial layer in the wafers obtained from the bottom side region in the straight body part of the silicon single crystal where the residence time in the oxygen precipitation nucleation temperature band is the longest. However, it is not the case that dislocation defects always occur in the wafers obtained from the bottom side region of the straight body part, and cases where dislocation defects do not occur in the epitaxial layer were also found.

[0013] An object of the present invention is to provide a method for evaluating a silicon single crystal capable of determining whether a predetermined region of a grown silicon single crystal is a region where an epitaxial silicon wafer with high gettering ability and capable of suppressing the occurrence of poor semiconductor device quality can be produced, and a method for manufacturing an epitaxial silicon wafer using the silicon single crystal evaluated by the method for evaluating the silicon single crystal.

Means for Solving the Problems

[0014] The method for evaluating a silicon single crystal of the present invention is grown by the Czochralski method, doped with boron, having a resistivity of 10 mΩ·cm or less, and an oxygen concentration of 12×10 17 atoms / cm 3 or more and 16×10 17 atoms / cm 3The following is a method for evaluating a silicon single crystal, comprising: an evaluation wafer acquisition step of acquiring an evaluation wafer from the bottom region of the straight body portion of the silicon single crystal; an evaluation epitaxial wafer preparation step of growing an epitaxial layer on the surface of the evaluation wafer to create an evaluation epitaxial wafer; an oxygen deposition nucleation growth treatment step of performing a heat treatment on the evaluation epitaxial wafer under conditions satisfying the following formulas (1) and (2) in an oxidizing atmosphere; a BMD density measurement step of measuring the BMD density of the evaluation wafer constituting the evaluation epitaxial wafer after the oxygen deposition nucleation growth treatment step; and an evaluation step of determining that if the BMD density measured in the BMD density measurement step is above a threshold, a predetermined region including the acquisition position of the evaluation wafer in the straight body portion is a failure region, and if it is below the threshold, the predetermined region is a passing region. H≧-0.06×T+70 … (1) 900 ≤ T ≤ 1100 … (2) H: Heat treatment time (hours) T: Heat treatment temperature (°C)

[0015] In the silicon single crystal evaluation method of the present invention, it is preferable to include a threshold setting step in which, when a flash lamp heat treatment is performed on the evaluation epitaxial wafer after the oxygen deposition nucleation growth treatment step, the BMD density of the evaluation wafer constituting the evaluation epitaxial wafer in which dislocation defects have occurred in the epitaxial layer is measured, and the threshold is set to be less than or equal to the measured value.

[0016] In the silicon single crystal evaluation method of the present invention, the straight body portion of the silicon single crystal is subjected to heat treatment at 800°C for 3 hours in an oxidizing atmosphere, followed by heat treatment at 1000°C for 16 hours, and the BMD density in the silicon wafer is 3 × 10⁻¹⁶. 9 pieces / cm 3 The above 1 x 10 11 pieces / cm 3 It is preferable that the straight body portion is formed to satisfy the following conditions.

[0017] In the silicon single crystal evaluation method of the present invention, it is preferable that the straight body portion of the silicon single crystal does not include the OSF region.

[0018] The present invention provides a method for evaluating silicon single crystals, comprising: first determining the oxygen nucleation temperature range by measuring the temperature inside the pulling furnace or by simulation; calculating the time the silicon single crystal in growth stays in the oxygen nucleation temperature range; and in the evaluation wafer acquisition step, acquiring the evaluation wafer from the bottom region where the residence time calculated in the residence time calculation step exceeds a predetermined residence time. Preferably .

[0019] The present invention provides a method for manufacturing an epitaxial silicon wafer, which involves performing an epitaxial growth process on a wafer obtained from a region determined to be a suitable region using the silicon single crystal evaluation method described above. [Brief explanation of the drawing]

[0020] [Figure 1] The figures show the experimental results obtained to derive the present invention. (A) is a graph showing the relationship between the solidification rate and resistivity of silicon single crystals representing the wafer acquisition positions in Experimental Examples 1 to 3. (B) is a graph showing the relationship between the solidification rate and the formation temperature residence time of silicon single crystals in Experimental Examples 1 to 3. [Figure 2] The figures show the experimental results obtained to derive the present invention. (A) is a graph showing the relationship between the solidification rate of silicon single crystals in Experimental Examples 1 to 3 and the BMD density of wafers subjected to 2-step deposition heat treatment. (B) is a graph showing the relationship between the resistivity of wafers in Experimental Examples 1 to 3 and the BMD density of wafers subjected to 2-step deposition heat treatment. [Figure 3] This figure shows the state of dislocation defect generation in experiments conducted to derive the present invention. [Figure 4]A diagram showing the experimental results for leading to the present invention. (A) is a graph showing the relationship between the solidification rate of the silicon single crystal in Experimental Examples 1 to 3 and the BMD density of the wafer on which the oxygen precipitation nucleation growth treatment was performed, and (B) is a graph showing the relationship between the resistivity of the wafers in Experimental Examples 1 to 3 and the BMD density of the wafer on which the oxygen precipitation nucleation growth treatment was performed. [Figure 5] A flowchart showing the threshold setting process according to an embodiment of the present invention. [Figure 6] An explanatory diagram showing an example of a method for setting a threshold according to an embodiment of the present invention. [Figure 7] A flowchart showing the evaluation process of a silicon single crystal according to an embodiment of the present invention.

Embodiments for Carrying Out the Invention

[0021] [Background Leading to the Present Invention] As a result of intensive research, the inventor has speculated that the cause of the generation of dislocation defects is due to excessive oxygen precipitation during the growth of the silicon single crystal due to the long residence time in the oxygen precipitation nucleation temperature range, resulting in an excessive increase in the BMD density in the wafer. As a result, the inventor has completed an evaluation method for a silicon single crystal that can determine whether the grown silicon single crystal is a crystal region where no dislocation defects occur in the epitaxial layer. Hereinafter, the details of the background leading to the present invention will be described with reference to the drawings.

[0022] Using a silicon single crystal manufacturing apparatus not shown, boron was added as a dopant, and the growth conditions by the CZ method were controlled so that the straight body part satisfied the following conditions, and a silicon single crystal of Experimental Example 1 having a shoulder part, a straight body part, and a tail part was manufactured. The specifications of the straight body part of the silicon single crystal are as follows. In the following description, the oxygen concentration is the oxygen concentration measured using a Fourier transform infrared spectrophotometer (FT-IR) in accordance with the infrared absorption method defined in ASTM F121-1979. Resistivity of the upper end (top part): 15 mΩ·cm or less <​​atoms / cm 3 The above 13.5 × 10 17 atoms / cm 3 below Diameter: 300mm (diameter after outer cylindrical grinding) Crystalline regions: Do not include COP (Crystal Originated Particle), OSF (Oxidation-Induced Stacking Fault) regions, or dislocation clusters.

[0023] Furthermore, except for setting the resistivity of the upper end of the straight section to 10 mΩ·cm, the growth conditions were controlled in the same way as in Experimental Example 1 to obtain the silicon single crystal of Experimental Example 2. Furthermore, except for setting the resistivity of the upper end of the straight section to 5 mΩ·cm, the growth conditions were controlled in the same way as in Experimental Example 1 to obtain the silicon single crystal of Experimental Example 3. The silicon single crystals of Examples 1 to 3 were subjected to outer cylindrical grinding to process them into silicon single crystals with a straight body diameter of 300 mm.

[0024] Next, three wafers for Experimental Example 1 were obtained from a total of 11 locations (hereinafter sometimes referred to as "wafer acquisition locations") in the straight body of the silicon single crystal of Experimental Example 1, where the solidification rates were 5%, 95%, and 10 × N% (where N is an integer between 1 and 9). Similarly, three wafers for Experimental Examples 2 and 3 were obtained from the straight body of the silicon single crystal of Experimental Example 1, each from the wafer acquisition locations. In Experimental Examples 1 to 3, wafers were cut from the straight body of the silicon single crystal using a band saw. Note that the position where the solidification rate is 0% is the upper end in the pulling direction of the straight section, and the position where it is 100% is the lower end. Then, each acquired wafer was subjected to a predetermined processing treatment (grinding, etching, polishing, and cleaning), and the resistivity of each wafer from Experimental Examples 1 to 3, which were acquired from different locations, was measured. Figure 1(A) shows the relationship between the solidification rate of the silicon single crystal representing the wafer acquisition location for Experimental Examples 1 to 3 and the resistivity.

[0025] As shown in Figure 1(A), the resistivity of the straight sections of the silicon single crystals in experimental examples 1, 2, and 3 was confirmed to be 15 mΩ·cm or less, 10 mΩ·cm or less, and 5 mΩ·cm or less, respectively.

[0026] The temperature range between 600°C and 800°C during the cooling process of silicon single crystals is known as the oxygen nucleation temperature range where oxygen nuclei are formed in silicon single crystals. Therefore, we investigated the time required for each position of the silicon crystals in Experimental Examples 1 to 3 to cool from 800°C to 600°C, that is, the time spent in the oxygen nucleation temperature range (hereinafter sometimes referred to as "residence time at the formation temperature"). Figure 1(B) shows the relationship between the solidification rate and the residence time at the formation temperature of the silicon single crystals in Experimental Examples 1 to 3. Note that the residence time at the formation temperature for positions with a solidification rate of less than 80% is less than or equal to the residence time at the formation temperature for positions with an 80% solidification rate, and therefore the residence time at the formation temperature for positions with a solidification rate of less than 80% is not shown in Figure 1(B). Furthermore, the relationship shown in Figure 1(B) can be calculated based on actual values ​​of the pulling speed during the formation of the straight body and tail portion of the silicon single crystal during growth, after determining the oxygen nucleation temperature range in advance by measuring the temperature inside the pulling furnace or by using known simulations.

[0027] As shown in Figure 1(B), in the silicon single crystals of Experimental Examples 1-3, it was confirmed that the residence time at the formation temperature was locally longest at a solidification rate of approximately 90%. Thus, the location with the longest residence time at the formation temperature occurs in the region of the straight body where the solidification rate is 85% or higher. Hereafter, the location with the longest residence time at the formation temperature may be referred to as the "location with the longest residence time at the formation temperature." The longest residence time at the formation temperature was observed in the silicon single crystal of Experimental Example 3, while the shortest residence time was observed in the silicon single crystal of Experimental Example 1. In other words, it was confirmed that the lower the resistivity (the higher the boron concentration), the longer the residence time at the formation temperature. This is presumed to be because the cooling rate of the silicon single crystal itself decreases as the boron concentration increases, thus increasing the residence time at the formation temperature. Furthermore, the longer the residence time at the formation temperature, the more oxygen precipitation nuclei are formed in the silicon single crystal. Based on the results shown in Figure 1(B), it can be estimated that the oxygen precipitation nuclei in the silicon single crystals of Experimental Examples 1 to 3 are most numerous at the position with the longest residence time at the formation temperature, where the solidification rate is approximately 90%. Comparing the number of oxygen precipitation nuclei at the position with the longest residence time at the formation temperature in the silicon single crystals of Experimental Examples 1 to 3, it can be estimated that the silicon single crystal of Experimental Example 3 has the most nuclei, while the silicon single crystal of Experimental Example 1 has the fewest.

[0028] Next, a total of 99 epitaxial layers with a thickness of 3 μm were grown on each of the wafers for Experimental Examples 1 to 3 using a vapor phase growth apparatus (not shown) at a temperature between 1000°C and 1175°C, thereby creating 99 epitaxial wafers for each of Experimental Examples 1 to 3. Then, a two-step deposition heat treatment for depositing BMDs, commonly performed, was carried out on two epiwafers each from experimental examples 1 to 3, which were acquired from different wafer locations. The two-step deposition heat treatment comprises a first heat treatment, in which heat treatment is performed at 800°C, the oxygen deposition nucleation temperature, for 3 hours in a heat treatment furnace (not shown) in an oxidizing atmosphere, and a second heat treatment, in which the temperature is raised to 1000°C without removing the epiwafer from the heat treatment furnace, and heat treatment is performed at 1000°C, the oxygen deposition nucleation growth temperature, for 16 hours. The temperature of the first heat treatment can be any temperature within the oxygen deposition nucleation temperature range, and is set in the range of 600°C to 800°C, while the temperature of the second heat treatment must be set in the temperature range in which oxygen deposition nuclei grow, and is set to at least 900°C or higher.

[0029] Then, from the epiwafers that underwent the 2-step deposition heat treatment, one epiwafer each from Experimental Examples 1 to 3, which were acquired from different wafer locations, were cleaved. After etching the cleaved surface with a light etching solution to a thickness of 2 μm, the BMD density of the wafer was measured by observing the cleaved surface with an optical microscope. The measurement range for BMD density was from a position 5 mm outside the center of the epiwafer to a position 10 mm inside the outer edge. Figure 2(A) shows the relationship between the solidification rate of silicon single crystals in Experimental Examples 1-3 and the BMD density of the epiwafer wafers that underwent the 2-step deposition heat treatment. Figure 2(B) shows the relationship between the resistivity of the wafers in Experimental Examples 1-3 and the BMD density of the wafers that underwent the 2-step deposition heat treatment.

[0030] As shown in Figure 2(A), in the silicon single crystal of Experimental Example 1, the BMD density increases as the solidification rate increases, but the BMD density is 3 × 10⁻⁶. 9 pieces / cm 3 It was confirmed that the straight-shell region satisfying the above conditions could only be obtained in a very small crystalline region beyond 80% solidification. On the other hand, in the silicon single crystals of Experimental Examples 2 and 3, since they are silicon single crystals with a resistivity of 10 mΩ·cm or less due to high concentration of boron addition, the BMD density is higher than in Experimental Example 1, and the BMD density is 3 × 10⁻⁶ throughout the entire straight-shell region. 9 pieces / cm 3 The above 1 x 10 11 pieces / cm 3 The following results confirmed that it is a silicon single crystal with a uniform BMD density throughout almost the entire straight body. Furthermore, as shown in Figure 2(B), in all of the experimental examples 1 to 3, it was confirmed that the BMD density increased as the resistivity of the silicon single crystal decreased.

[0031] Next, flash lamp heat treatment was performed on the remaining epiwafers from Experimental Examples 1-3 that had undergone 2-step deposition heat treatment but had not been cleaved (and therefore BMD density was not measured). This treatment simulated a part of the semiconductor device manufacturing process. In Experimental Examples 1-3, a flash lamp heat treatment device (LA-3000-F) manufactured by SCREEN Semiconductor Solutions was used for the flash lamp heat treatment. After preheating from room temperature to an assist heat temperature of 800°C, the heat treatment was performed at 1200°C in a nitrogen atmosphere for 20 msec. Note that the flash lamp heat treatment conditions are not limited to the above conditions. For example, the heat treatment temperature could be 1300°C, and the heat treatment time could be set to 1.4 msec or 2 msec. Any heat treatment conditions that simulate the device process conditions implemented by the user are acceptable. Next, the epitaxial layer surface of the epiwafers in Experimental Examples 1-3 was observed using X-ray topography to investigate the occurrence of dislocation defects in the epitaxial layer. As a result, it was found that dislocation defects occurred only in the epiwafer containing the wafer at the position with a solidification rate of 90% among the epiwafers of experimental examples 2 and 3. In other words, it was found that dislocation defects occurred only in the epiwafer containing the wafer at approximately the same position as the position with the longest residence time at the formation temperature in a silicon single crystal with a resistivity of 10 mΩ·cm or less in the straight body. Furthermore, when an epiwafer in which dislocation defects had occurred was cleaved and the cleaved cross-section was observed with a TEM (Transmission Electron Microscope), as shown in Figure 3, it was confirmed that BMDs on the wafer surface in regions with particularly high BMD density (high-density BMD regions in Figure 3) were inducing dislocation defects. This indicates that the higher the BMD density of the wafer, the more likely dislocation defects are to occur.

[0032] Based on the relationship between BMD density and the likelihood of dislocation defect occurrence, and the occurrence of dislocation defects after flash lamp heat treatment, it can be estimated that the BMD density at the position with the longest residence time at the formation temperature in the silicon single crystals of Experimental Examples 2 and 3 is locally higher than at other positions. However, the results shown in Figures 2(A) and (B) do not indicate that the BMD density at the position with the longest residence time at the formation temperature in the silicon single crystals of Experimental Examples 2 and 3 is locally higher than at other positions. The inventors considered that the 2-step deposition heat treatment was not suitable as a heat treatment to explain why dislocation defects are likely to occur in epiwafers having a silicon single crystal at a solidification rate of 90% in Experimental Examples 2 and 3. Therefore, in the process of conducting diligent research, the inventors carried out the following experiments.

[0033] First, a new oxygen precipitation nucleation growth treatment was performed on the remaining epiwafers from experimental examples 1-3, which had not undergone the 2-step precipitation heat treatment. In this oxygen precipitation nucleation growth treatment, heat treatment was performed at 900°C, the temperature within the oxygen precipitation nucleation temperature range, for 16 hours in a heat treatment furnace with an oxidizing atmosphere, without performing heat treatment in the oxygen precipitation nucleation temperature range. Then, the BMD density was measured for each epiwafer after the oxygen deposition nucleation growth treatment using the same method as when the 2-step deposition heat treatment was performed. Figure 4(A) shows the relationship between the solidification rate of silicon single crystals in Experimental Examples 1-3 and the BMD density of wafers treated with oxygen deposition nucleation growth. Figure 4(B) shows the relationship between the resistivity of wafers in Experimental Examples 1-3 and the BMD density of wafers treated with oxygen deposition nucleation growth.

[0034] As shown in Figure 4(A), in all of the silicon single crystals in Experimental Examples 1 to 3, the BMD density of the wafer was locally higher at the position where the solidification rate of the silicon single crystal was 90%, and it was confirmed that the difference between the BMD density at the 90% solidification rate position and the BMD density at the 80% solidification rate position increased in the order of Experimental Example 1, Experimental Example 2, and Experimental Example 3. Furthermore, as shown in Figure 4(B), it was confirmed that the BMD density increased locally in Experimental Examples 2 and 3, where the resistivity was low. In other words, it was confirmed that the lower the resistivity, the greater the difference between the BMD density at 90% solidification and the BMD density at 80% solidification. Note that even when the oxygen precipitation nucleation growth treatment conditions were changed to a heat treatment at 1100°C for 4 hours, which is within the oxygen precipitation nucleation growth temperature range, the results were almost the same as those shown in Figures 4(A) and (B).

[0035] The solidification rates in which the BMD density is locally high, as shown in Figures 4(A) and 4(B), are roughly consistent with the solidification rates in which the residence time at the formation temperature in the oxygen nucleation temperature band is locally longer, as shown in Figure 1, and with the solidification rates at the wafer acquisition locations of epiwafers where dislocation defects occur after flash lamp heat treatment. From the above results, the inventors have found that by performing oxygen deposition nucleation growth treatment on an epiwafer and measuring the BMD density, it is possible to determine whether a predetermined region of a silicon single crystal is a region in which an epiwafer with high gettering ability and the occurrence of semiconductor device quality defects can be created. The inventors then hypothesized the following reasons for the difference in BMD generation between the two-step deposition heat treatment and the oxygen deposition nucleation growth treatment applied to the epiwafer.

[0036] In the straight body of a silicon single crystal, there are oxygen precipitation nuclei of a size that can grow into BMDs by heat treatment within the oxygen precipitation nucleus growth temperature range (hereinafter sometimes referred to as "growable precipitation nuclei") and oxygen precipitation nuclei of a size that cannot grow into BMDs by heat treatment within the oxygen precipitation nucleus growth temperature range (hereinafter sometimes referred to as "ungrowable precipitation nuclei"). Furthermore, the longer the residence time at the formation temperature and the greater the number of oxygen precipitation nuclei present at a location, the higher the proportion of growable precipitation nuclei at that location (hereinafter sometimes referred to as the "proportion of growable precipitation nuclei"). For this reason, the number of growable precipitation nuclei at the location with the longest residence time at the formation temperature of a silicon single crystal is locally higher compared to other locations.

[0037] On the other hand, some oxygen nuclei present in the wafer disappear due to heating during epitaxial layer growth. However, the relationship remains that the wafer at the location with the longest residence time at the formation temperature has a locally greater number of growable nuclei compared to wafers at other locations.

[0038] When an epiwafer is subjected to a heat treatment in the oxygen nucleation temperature range, such as a two-step deposition heat treatment, non-growthable nuclei enlarge and become growable nuclei, resulting in an increase in the number of growable nuclei on the wafer. The lower the proportion of growable nuclei before heat treatment in the oxygen nucleation temperature range, that is, the higher the proportion of non-growthable nuclei, the more non-growthable nuclei will become growable nuclei after heat treatment in the oxygen nucleation temperature range. Therefore, by first performing a heat treatment in the oxygen nucleus formation temperature range, such as in a 2-step deposition heat treatment, on epiwafers having wafers obtained from various positions in the straight body, the difference in the number of growable nuclei on each epiwafer after the heat treatment becomes smaller, and the number of growable nuclei on the wafer at the position with the longest residence time at the formation temperature does not become locally higher compared to wafers at other positions. Then, by performing heat treatment in the oxygen precipitation nucleus growth temperature range, only growable precipitation nuclei will grow into BMDs. Therefore, as in the 2-step precipitation heat treatment, when heat treatment is performed in the oxygen precipitation nucleation growth temperature range after heat treatment in the oxygen precipitation nucleation temperature range, it was estimated that the relationship between the solidification rate of the silicon single crystal and the BMD density of the wafer subjected to the 2-step precipitation heat treatment would result in a distribution where the BMD density of the wafer at the position with a solidification rate of 90%, which is approximately the same as the position with the longest residence time at the formation temperature, does not become locally high, as shown in Figure 2(A).

[0039] On the other hand, for each epiwafer that has a relationship in which the wafer at the position with the longest residence time at the formation temperature has a locally higher number of growable nuclei compared to wafers at other positions, if heat treatment is performed in the oxygen precipitation nucleus growth temperature range without performing heat treatment in the oxygen precipitation nucleus formation temperature range, as in oxygen precipitation nucleus growth treatment, non-growable nuclei do not grow larger and become growable nuclei, and only the growable nuclei that existed before heat treatment grow into BMDs. For this reason, the relationship between the solidification rate of the silicon single crystal and the BMD density of wafers that have undergone oxygen precipitation nucleus growth treatment is estimated to be such that the BMD density of wafers at the position with a solidification rate of 90%, which is almost the same as the position with the longest residence time at the formation temperature, is locally higher, as shown in Figure 4(A). The silicon single crystal evaluation method of the present invention was completed based on the above findings.

[0040] [Embodiment] Next, I will describe one embodiment of the present invention.

[0041] [Overview of the Embodiment] This embodiment first performs a threshold setting process to set a plurality of thresholds associated with the growth conditions for silicon single crystals and the growth conditions for epitaxial layers, wherein at least one of the growth conditions and the growth conditions is different from each other.

[0042] Next, using one of several thresholds, an evaluation process is performed on the silicon single crystal using the silicon single crystal evaluation method of the present invention. The evaluation process for silicon single crystals involves creating an evaluation epitaxial wafer by growing an epitaxial layer on an evaluation wafer obtained from the straight body portion of the silicon single crystal to be evaluated, and measuring the BMD density of the evaluation epitaxial wafer that has undergone the oxygen deposition nucleation growth treatment described above. Furthermore, a threshold is selected from multiple thresholds that is associated with the same growth conditions as the silicon single crystal to be evaluated and the same growth conditions as the epitaxial layer of the evaluation epitaxial wafer. If the BMD density of the evaluation wafer is above the threshold, a predetermined region including the acquisition location of the evaluation wafer is determined to be a failure region; if it is below the threshold, the predetermined region is determined to be a success region. The failure region is an epitaxial wafer having an epitaxial layer formed under the same growth conditions as the epitaxial layer of the evaluation epitaxial wafer, and it is impossible to create an epitaxial wafer that has high gettering ability and can suppress the occurrence of quality defects in semiconductor devices. The acceptable region is an epitaxial wafer having an epitaxial layer formed under the same growth conditions as the evaluation epitaxial layer, and is a region in which it is possible to create an epitaxial wafer that has high gettering capability and can suppress the occurrence of quality defects in semiconductor devices. The following describes the details of the threshold setting process and the silicon single crystal evaluation process.

[0043] [Threshold setting process] The threshold setting process will be explained with reference to the diagram. Figure 5 is a flowchart of the threshold setting process. Figure 6 is an explanatory diagram showing an example of how to set a threshold. The threshold setting process constitutes a part of the silicon single crystal evaluation method of the present invention.

[0044] The threshold setting process shown in Figure 5 sets a single threshold associated with predetermined silicon single crystal growth conditions and predetermined epitaxial layer growth conditions. In the threshold setting process, a silicon single crystal for threshold setting is manufactured using a silicon single crystal manufacturing apparatus under predetermined growth conditions (configuration of the hot zone of the silicon single crystal manufacturing apparatus, dopant of the silicon single crystal, resistivity, shape, and pulling speed) (Step S1). Furthermore, the silicon single crystal used for threshold setting manufactured in step S1 has the following properties. Dopant: Boron Resistivity: 1mΩ·cm or more and 10mΩ·cm or less Oxygen concentration: 12 × 10 17 atoms / cm 3 The above 16 x 10 17 atoms / cm 3 below Diameter: 300mm (diameter after outer cylindrical grinding) Crystalline regions: Do not include COP, OSF regions, or dislocation clusters. BMD density: The BMD density obtained from the entire straight body of a wafer after a two-step deposition heat treatment (for example, first heat treatment: 800°C for 3 hours or more, second heat treatment: 1000°C for 16 hours or more) (measured using the same method as the BMD density measurement method for wafers in Experimental Examples 1-3) is 3 × 10⁻⁶ 9 pieces / cm 3 The above 1 x 10 11 pieces / cm 3

[0045] As shown in Figure 1(B), the relationship between the solidification rate and the residence time at the formation temperature of the silicon single crystal manufactured in this manner for threshold setting may show that the residence time at the formation temperature is locally longer at the lower end of the straight body.

[0046] Next, a wafer is obtained from the bottom region of the straight body of the silicon single crystal used for threshold setting (step S2). In step S2 and step S22 described later, the bottom region of the straight body is the region with a solidification rate of 85% or more. Within the region with a solidification rate of 85% or more, a crystal region occurs that has the longest residence time in the oxygen precipitation nucleation temperature band. The wafer obtained in step S2 is also the evaluation wafer of the present invention. In step S2, two wafers are acquired from multiple different positions along the length of the straight body. Then, each acquired wafer is subjected to a predetermined processing treatment (grinding, etching, polishing, cleaning) using various devices.

[0047] Subsequently, using a vapor phase growth apparatus (not shown), an epitaxial layer is grown on the wafer that has undergone predetermined processing under predetermined growth conditions (growth temperature, growth time, growth gas, and epitaxial layer thickness) to create the first and second epitaxial wafers (Step S3: Epitaxial wafer fabrication process). The first epiwafer is one of two epiwafers obtained from the same position on the straight body, and the second epiwafer is the other epiwafer.

[0048] Next, oxygen deposition nucleation growth treatment is performed on each first epiwafer and each second epiwafer using a heat treatment furnace (not shown) (Step S4: BMD density measurement step for threshold setting). In step S4, the oxygen precipitation nucleation growth treatment is performed under an oxidizing atmosphere within the oxygen precipitation nucleation growth temperature range, without performing heat treatment within the oxygen precipitation nucleation formation temperature range. The conditions for the oxygen precipitation nucleus growth treatment are preferably such that the following conditions (3) and (4) are met, where T (°C) is the oxygen precipitation nucleus growth temperature and H (hours) is the heat treatment time. For example, the oxygen precipitation nucleus growth treatment may be performed at a temperature of 900°C for 16 hours or more, or at a temperature of 1100°C for 4 hours or more. If the oxygen precipitation nucleus growth treatment does not satisfy the conditions of formula (3), as in the case of a 2-step precipitation heat treatment, the ungrowable nuclei may become large enough to become growable nuclei, and the BMD density of the first epiwafer obtained from a predetermined region including the position with the longest residence time at the formation temperature may not be locally higher than the BMD of the first wafer obtained from other positions. H≧-0.06×T+70 … (3) 900 ≤ T ≤ 1100 … (4)

[0049] Next, the BMD density of the first wafers constituting each first epiwafer after the oxygen deposition nucleation growth treatment is measured (Step S5: BMD density measurement step for threshold setting). As a method for measuring BMD density, a method similar to the method used to measure the BMD density of the wafers in Experimental Examples 1-3 can be exemplified.

[0050] Next, flash lamp heat treatment is performed on each second epiwafer after oxygen deposition nucleation growth treatment using a flash lamp heat treatment apparatus (not shown) (Step S6: flash lamp heat treatment step). For flash lamp heat treatment, the same conditions as those used for the flash lamp heat treatment on the epiwafers in Experimental Examples 1-3 can be used as examples.

[0051] Next, in order to identify the location of defects in each second epiwafer, the surface of each second epiwafer is observed (Step S7: Defect presence / absence determination step, defect location identification step). As an example of a surface observation method, one can observe the surface of the epitaxial layer side of the second epiwafer using X-ray topography. Next, based on the surface observation results, it is determined whether or not defects have occurred in each second epiwafer (Step S8: Defect presence / absence determination step, defect location identification step).

[0052] If it is determined that a defect has occurred in the second epiwafer (Step S8: YES), a cross-sectional observation of the location of the defect in the second epiwafer is performed (Step S9: Defect presence / absence determination step, determination step). As an example of a cross-sectional observation method, one can exemplified by cleaving the second epiwafer and observing the cleaved cross-section using TEM. Next, based on the cross-sectional observation results, it is determined whether the defect is a dislocation defect caused by BMD in the second epiwafer (Step S10: Defect presence / absence determination step, determination step).

[0053] If it is determined that the defect is a dislocation defect caused by BMD (Step S10: YES), the BMD density of the first wafer, which is acquired at the same location as the second wafer that constitutes the second epiwafer, is set to a rejection value (Step S11).

[0054] On the other hand, if it is determined that no defects have occurred in the second epiwafer based on the cross-sectional observation results (Step S8: NO), or if it is determined that the defects are not dislocation defects caused by BMD based on the surface observation results (Step S10: NO), the BMD density of the first wafer at the acquisition location, which is the same as that of the second wafer constituting the second epiwafer, is set to an acceptable value (Step S12).

[0055] After setting the BMD density of all first wafers to either a failing or passing value, it is determined whether or not a failing BMD density exists (step S13).

[0056] If it is determined that there is a BMD density below the acceptable level (Step S13: YES), a threshold is set based on the acceptable and unacceptable values ​​(Step S14: Setting step). For example, a value less than or equal to the smallest failing value and greater than the largest passing value is set as the threshold. Specifically, if the relationship between the solidification rate, which represents the acquisition position of the first wafer, and the pass / fail status of the BMD density is as shown in Figure 6 (circles indicate passing values, crosses indicate failing values), then a predetermined value within the threshold-setting range between the smallest failing value, which is a BMD density with a solidification rate of 90%, and the largest passing value, which is a BMD density with a solidification rate of 80%, is set as the threshold. Alternatively, instead of considering the passing value, a threshold value that is a predetermined value smaller than the smallest failing value may be set.

[0057] Here, in step S14, the threshold is set based not on the BMD density in the second wafer of the second epiwafer where the dislocation defect occurred, but on the BMD density in the first wafer obtained from a position adjacent to the second wafer. However, since the BMD densities of two adjacent wafers are considered to be approximately the same, the process in step S14 can be considered as a process of setting the threshold to a value less than or equal to the BMD density in the evaluation wafer (first wafer) of the epiwafer (first epiwafer) where the dislocation defect occurred in the epitaxial layer. For example, instead of performing the processes in steps S9 and S10, the BMD density of the wafer whose surface was observed in step S7 may be measured, and based on whether or not it was determined that dislocation defects had occurred in the surface observation, the process in step S12 or step S13 may be performed using the BMD density of the wafer whose surface was observed. In this case, the epiwafer created in step S3 may not be distinguished as a first or second epiwafer, and the process in step S5 may not be performed, and the processes in steps S4, S6 and onward may be performed.

[0058] The threshold set in step S14 is associated with the growth conditions for the silicon single crystal used for threshold setting and the growth conditions for the epitaxial layer of the evaluation epiwafer, i.e., predetermined growth conditions for the silicon single crystal and predetermined growth conditions for the epitaxial layer. The threshold setting process in step S14 may be performed by an operator or by a computer. Furthermore, the threshold may be recorded on a medium such as paper, or it may be stored in the computer's memory so that it can be displayed on a display device.

[0059] On the other hand, if it is determined that there are no BMD densities that do not meet the criteria (step S13: NO), the threshold setting process is terminated without performing the process in step S14.

[0060] By performing the above threshold setting process multiple times, with at least one of the silicon single crystal growth conditions for threshold setting and the epitaxial layer growth conditions of the evaluation epiwafer differing, it is possible to set multiple thresholds associated with the silicon single crystal growth conditions and the epitaxial layer growth conditions, where at least one of the growth conditions and the growth conditions is different from each other.

[0061] [Evaluation process for silicon single crystals] Next, the evaluation process for silicon single crystals, which constitutes a part of the silicon single crystal evaluation method of the present invention, will be explained with reference to the drawings. Figure 7 is a flowchart showing the evaluation process for silicon single crystals.

[0062] As shown in Figure 7, a silicon single crystal to be evaluated is manufactured using a silicon single crystal manufacturing apparatus under predetermined growth conditions (step S21). The silicon single crystal to be evaluated, manufactured in step S21, has the same properties (dopant, resistivity, oxygen concentration, diameter, crystal region, BMD density) as the silicon single crystal used for threshold setting described above.

[0063] Next, an evaluation wafer is obtained from the bottom region (the region with a solidification rate of 85% or more) of the straight body of the silicon single crystal to be evaluated (Step S22: Evaluation wafer acquisition process). In step S22, first, for example, the oxygen nucleation temperature range is determined in advance by measuring the temperature inside the lifting furnace or by simulation, and the time that the silicon single crystal stayed in the oxygen nucleation temperature range is calculated (residence time calculation step). Then, evaluation wafers are obtained from the bottom region where the calculated residence time exceeds a predetermined residence time. After this, various devices are used to perform predetermined processing (grinding, etching, polishing, and cleaning) on ​​each evaluation wafer obtained. Furthermore, evaluation wafers may be obtained as follows. For example, the position with the longest dwell time at the formation temperature in the straight body can be identified. Then, one evaluation wafer may be obtained from the position with the longest dwell time at the formation temperature in the straight body, and from positions that are the same distance away from the position with the longest dwell time at the formation temperature, both on the upper and lower ends in the pulling direction. For example, if the position with the longest dwell time at the formation temperature is at a solidification rate of 85%, evaluation wafers may be obtained from positions with solidification rates of 70%, 85%, and 100%. Furthermore, the acquisition location for evaluation wafers may be one location or two or more locations.

[0064] Next, an evaluation epitaxial wafer is created by growing an epitaxial layer on an evaluation wafer under predetermined growth conditions using a vapor phase growth apparatus (not shown) (Step S23: Evaluation epitaxial wafer creation process).

[0065] Next, the operator performs an oxygen precipitation nucleation growth treatment on each evaluation epiwafer using a heat treatment furnace (not shown) (Step S24: Oxygen Precipitation Nucleation Growth Treatment Process). The oxygen precipitation nucleation growth treatment in step S24 is performed under the same conditions as in step S4 of the threshold setting treatment.

[0066] Next, the BMD density of each evaluation wafer that makes up the evaluation epiwafer after the oxygen deposition nucleation growth treatment is measured (Step S25: BMD density measurement step). BMD density is measured at the center of each evaluation wafer, at a position half a radius outside the center, and at a position 10 mm inside the outer edge, using the same method as in step S5 of the threshold setting process. The measurement locations for BMD density may be within the same range as those for the epiwafers in Experimental Examples 1-3, or they may be one, two, or four or more predetermined locations.

[0067] Next, a threshold is selected from multiple thresholds that is associated with the same growth conditions as the silicon single crystal under evaluation and the same growth conditions as the epitaxial layer of the evaluation epiway (Step S26: Threshold selection step). Note that the threshold selection process in step S26 may be performed by a computer.

[0068] Then, an evaluation is performed based on the BMD density of each evaluation wafer and the selected threshold (Step S27: Evaluation process). For example, if the BMD density at at least one of the three locations on the evaluation wafer exceeds a threshold, a predetermined area including the acquisition location of the evaluation wafer in the straight body is determined to be a failure area. On the other hand, if the BMD density at all three locations on the evaluation wafer is below the threshold, the predetermined area including the acquisition location of the evaluation wafer is determined to be a passing region. The rejection area and the pass area may be an area centered on the acquisition position of the evaluation wafer, or an area not centered on it. For example, if the acquisition position of the evaluation wafer is the position with a solidification rate of X%, the rejection area and the pass area may be from the position with a solidification rate of (X-5)% to the position with a solidification rate of (X+5)%. Also, the evaluation process in step S27 may be performed by a computer.

[0069] Furthermore, if there are multiple locations for acquiring the evaluation wafer in step S22, the evaluation process in step S27 may be performed as follows. If there are evaluation wafers whose BMD density is determined to be below the threshold, then among the evaluation wafers with a BMD density below the threshold, the entire region on the upper side, with the acquisition position of the evaluation wafer acquired from the uppermost edge as the lower end, may be determined to be an acceptable region, and all regions below this acceptable region may be determined to be unacceptable regions. The reason for determining the upper region as an acceptable region based on the BMD density of the evaluation wafer acquired only from the lower end of the straight body is that, as shown in Figure 1(B), for example, the formation temperature residence time of the upper region is shorter than the formation temperature residence time of the acquisition position at the uppermost edge of the evaluation wafer, and therefore the BMD density is also considered to be lower.

[0070] [Effects of the Embodiment] The evaluation process for silicon single crystals involves a resistivity of 10 mΩ·cm or less with boron doping, and an oxygen concentration of 12 × 10⁻¹⁰. 17 atoms / cm 3 The above 16 x 10 17 atoms / cm 3 An evaluation epiwafer is prepared using an evaluation wafer obtained from the following silicon single crystal. The evaluation epiwafer is then subjected to oxygen precipitation nucleation growth treatment in an oxidizing atmosphere in the oxygen precipitation nucleation growth temperature range that satisfies equations (3) and (4) above, without performing heat treatment in the oxygen precipitation nucleation temperature range. In the evaluation process of the silicon single crystal, if the BMD density of the evaluation wafer is above a threshold, a predetermined region including the acquisition position of the evaluation wafer in the straight body is determined to be a failure region. If it is below the threshold, the predetermined region is determined to be a passing region. Thus, by performing oxygen deposition nucleation growth treatment instead of 2-step deposition heat treatment on the evaluation epiwafer, BMDs can be generated such that, as shown in Figure 4(A), the BMD density of the evaluation wafer obtained from a predetermined region including the position with the longest residence time at the formation temperature is locally higher than the BMD density of the evaluation wafer obtained from other locations. Therefore, based on the relationship between the BMD density of the evaluation wafer subjected to oxygen deposition nucleation growth treatment and the threshold, it is possible to determine whether the predetermined region including the acquisition position of the evaluation wafer is a passable region in which an epiwafer with high gettering capability and suppression of semiconductor device quality defects can be produced. Furthermore, the evaluation process for silicon single crystals is such that the resistivity of the straight body of the silicon single crystal is between 1 mΩ·cm and 10 mΩ·cm (boron concentration is 8.5 × 10⁻⁶). 18 atoms / cm 3 The above 1.18 × 10 20 atoms / cm 3 (See below), and the oxygen concentration in the straight section is 12 × 10 17 atoms / cm 3 The above 16 x 10 17 atoms / cm 3 The following silicon single crystals will be evaluated. In this way, by evaluating silicon single crystals with high boron and oxygen concentrations, it is possible to create epiwafers with excellent BMD density and high gettering capability from the regions determined to be acceptable.

[0071] The evaluation process for silicon single crystals involves performing a 2-step deposition heat treatment on wafers obtained from the entire straight body, resulting in a BMD density of 3 × 10⁻¹⁶. 9 pieces / cm 3 The above 1 x 10 11 pieces / cm 3 The silicon single crystals formed as described below will be evaluated. As shown in Figure 2(A), even in silicon single crystals where the BMD density of the wafer obtained from the position with the longest residence time at the formation temperature does not increase locally in the 2-step deposition heat treatment, the evaluation process can increase the BMD density of the wafer obtained from the position with the longest residence time at the formation temperature, as shown in Figure 4(A), allowing for appropriate evaluation of the silicon single crystal.

[0072] The evaluation process for silicon single crystals targets silicon single crystals that do not contain OSF regions in the straight body. Therefore, it is possible to suppress the occurrence of defects caused by OSF in epiwafers made from wafers obtained from the approved region.

[0073] The evaluation process for silicon single crystals involves selecting a threshold from a set of thresholds associated with the growth conditions of the silicon single crystal and the growth conditions of the epitaxial layer, where at least one of the growth conditions is different from the other. The threshold is then selected to be associated with the same growth conditions as the silicon single crystal being evaluated and the same growth conditions as the epitaxial layer of the evaluation epiwafer. The evaluation is then performed based on the selected threshold. Here, the BMD density, which is one of the causes of dislocation defect generation, is thought to depend on the silicon single crystal growth conditions. Furthermore, since BMD disappears due to heating during epitaxial layer growth, the BMD density is thought to depend on the epitaxial layer growth conditions. In addition, the likelihood of dislocation defect generation is thought to depend on the thickness of the epitaxial layer. Moreover, if the threshold setting criteria in step S14 are the same, the threshold is thought to be a value corresponding to the silicon single crystal growth conditions and the epitaxial layer growth conditions. Therefore, if evaluation is performed using a threshold associated with at least one of the following: growth conditions different from those of the silicon single crystal being evaluated, and growth conditions different from those of the epitaxial layer of the evaluation epiwafer, there is a risk that a failing region may be judged as a passing region. As in this embodiment, by using thresholds associated with the same growth conditions as the silicon single crystal under evaluation and the same growth conditions as the epitaxial layer of the evaluation epiwafer, the silicon single crystal can be evaluated more appropriately.

[0074] The threshold setting process creates a first epiwafer and a second epiwafer by growing epitaxial layers on adjacent first and second wafers obtained from the straight body portion of a silicon single crystal used for threshold setting. Next, the threshold setting process performs oxygen deposition nucleation growth on the first and second epiwafers, and then measures the BMD density of the first wafer. It also performs flash lamp heat treatment on the second epiwafer to determine whether or not dislocation defects have occurred in the epitaxial layer. If dislocation defects have occurred in the epitaxial layer, the threshold setting process sets thresholds based on the BMD density of the first wafer, associated with the growth conditions for the silicon single crystal used for threshold setting and the growth conditions for the epitaxial layers of the first and second epiwafers. In this way, by using experimentally determined thresholds in the evaluation process of silicon single crystals, it is possible to evaluate silicon single crystals more appropriately.

[0075] The threshold setting process determines whether or not dislocation defects have occurred in the epitaxial layer by observing the epitaxial layer side surface of the second epiwafer to identify the location of the defect, and by observing the cross-section of the defect location on the second epiwafer to determine whether or not the defect is a dislocation defect caused by BMD in the second wafer. In this way, by pre-identifying the cross-sectional locations for confirming the presence of dislocation defects in the second epiwafer through surface observation, the presence or absence of dislocation defects can be efficiently determined.

[0076] [Differentiation] Although embodiments of the present invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments, and various improvements and design changes, etc., that do not depart from the spirit of the present invention are also included.

[0077] For example, a plurality of thresholds associated with the growth conditions of the silicon single crystal and the growth conditions of the epitaxial layer may be set in advance as thresholds used in the evaluation process of a silicon single crystal, wherein at least one of the growth conditions and the growth conditions is different from each other. Alternatively, a single threshold may be set regardless of the growth conditions of the silicon single crystal and the growth conditions of the epitaxial layer.

[0078] The threshold used in the evaluation process for silicon single crystals is selected from a plurality of thresholds associated with the growth conditions of the silicon single crystal and the growth conditions of the epitaxial layer, where at least one of the growth conditions is different from the other. Alternatively, a threshold associated with the same growth conditions as the silicon single crystal being evaluated may be selected from a plurality of thresholds associated only with the growth conditions of silicon single crystals that are different from each other. In this case, in step S14 of the threshold setting process described above, the threshold only needs to be associated with the growth conditions of the silicon single crystal. Thus, by using a threshold associated with the same growth conditions as the silicon single crystal being evaluated, it is possible to evaluate the silicon single crystal more appropriately compared to using a threshold set independently of the growth conditions.

[0079] In the threshold setting process, if a defect is confirmed to have occurred by surface observation of the second epiwafer, the defect may be considered a dislocation defect caused by BMD. Furthermore, if the locations where dislocation defects may occur are known empirically, cross-sectional observation of these locations may be performed without performing surface observation of the second epiwafer.

[0080] As a method for measuring BMD density, the laser scattering tomography (LST) method may be used, in which an infrared laser beam is incident from the surface of the wafer and the scattered light is detected. In this case, the threshold value may be set to a different value than that of the embodiment described above.

Claims

1. Grown using the Czochralski method, with boron added, the resistivity is 10 mΩ·cm or less, and the oxygen concentration is 12 × 10⁻¹⁰ １７ atoms / cm ３ The above 16 x 10 １７ atoms / cm ３ The following is a method for evaluating silicon single crystals, An evaluation wafer acquisition step is to acquire an evaluation wafer from the bottom region of the straight body portion of the silicon single crystal, A process for creating an evaluation epitaxial wafer by growing an epitaxial layer on the surface of the aforementioned evaluation wafer, The evaluation epitaxial wafer is subjected to an oxygen precipitation nucleation growth process, which involves heat treatment under conditions satisfying the following formulas (1) and (2) in an oxidizing atmosphere, A BMD density measurement step for measuring the BMD density of the evaluation wafer that constitutes the evaluation epitaxial wafer after the oxygen deposition nucleation growth treatment step, The evaluation step includes determining that if the BMD density measured in the BMD density measurement step is equal to or greater than a threshold, a predetermined region including the acquisition position of the evaluation wafer in the straight body is a failure region, and if it is less than the threshold, the predetermined region is a passing region. A method for evaluating a silicon single crystal, wherein the bottom region is a region with a solidification rate of 85% or more. H ≥ -0.06 × T + 70 … (1) 900 ≤ T ≤ 1100 … (2) H: Heat treatment time (hours) T: Heat treatment temperature (°C)

2. A method for evaluating a silicon single crystal according to Claim 1, comprising: creating a threshold-setting epitaxial wafer by growing an epitaxial layer on a wafer obtained from the bottom region of the straight body portion of the silicon single crystal for threshold setting; measuring the BMD density of the wafer constituting the threshold-setting epitaxial wafer when dislocation defects occur in the epitaxial layer when the threshold-setting epitaxial wafer is subjected to heat treatment and flash lamp heat treatment under the same conditions as the oxygen deposition nucleation growth treatment process; and setting the threshold to be less than or equal to the measured value.

3. When the straight body portion of the silicon single crystal is subjected to heat treatment at 800°C for 3 hours in an oxidizing atmosphere, followed by heat treatment at 1000°C for 16 hours, the BMD density in the silicon wafer is 3 × 10⁻¹⁶. ９ pieces / cm ３ The above 1 x 10 １１ pieces / cm ３ A method for evaluating a silicon single crystal according to claim 1, wherein the straight body portion is formed to satisfy the following conditions.

4. The method for evaluating a silicon single crystal according to claim 1, wherein the straight body portion of the silicon single crystal does not include an OSF region.

5. The system includes a residency time calculation step, in which the oxygen precipitation nucleation temperature range is determined in advance by measuring the temperature inside the lifting furnace or by simulation, and the time that the silicon single crystal being grown stays in the oxygen precipitation nucleation temperature range is calculated. The method for evaluating a silicon single crystal according to claim 1, wherein in the evaluation wafer acquisition step, the evaluation wafer is acquired from the bottom region where the residence time calculated in the residence time calculation step exceeds a predetermined residence time.

6. A method for manufacturing an epitaxial silicon wafer, comprising performing an epitaxial growth process on a wafer obtained from a region determined to be a passable region using the silicon single crystal evaluation method described in any one of claims 1 to 5.

Images