Method for determining defect region of silicon single-crystal substrate

Abstract

The present invention is a method for determining a defect region of a silicon single-crystal substrate having a mirror-polished (100) surface orientation, using a laser scattering surface inspection apparatus with a rotary stage, the method being characterized in that: crystal defects present in regions of a main surface of the silicon single-crystal substrate in which the crystal orientation includes the <010> direction and the <011> direction are measured using the oblique incidence mode of the surface inspection apparatus to determine the number of defects or the defect density of the crystal defects, and it is determined whether an I-rich region is included in the silicon single-crystal substrate on the basis of the difference between the number of defects or the defect density of the crystal defects present in the region in which the crystal orientation includes the <010> direction and the number of defects or the defect density of the crystal defects present in the region in which the crystal orientation includes the <011> direction. Thus, a non-destructive and simple method for I-rich determination for a defect region of a semiconductor substrate is provided.

Description

Method for determining defective areas in silicon single crystal substrates

[0001] The present invention relates to a method for determining a defect region in a silicon single crystal substrate.

[0002] In recent years, with the advancement of device technology, highly accurate control and evaluation of defect regions in semiconductor substrates have become increasingly important. Defect regions in semiconductor substrates that differ from desired defect regions can cause device failure. Examples of such defect regions include regions with an excess of interstitial silicon (hereinafter referred to as "I-rich regions"), regions prone to the generation of cavities due to vacancy aggregation (hereinafter referred to as "V-rich regions"), regions prone to the generation of oxygen precipitates (OSF regions), and regions with few defects (N regions). The types of defects present in each defect region vary. It is known that the type of defect region a substrate contains varies depending on the manufacturing conditions of the single crystal and the position of the single crystal. Therefore, it is necessary to determine the defect regions in the substrate during the manufacturing stage of the substrate, and if undesirable crystal defects are present, to quickly provide feedback to the manufacturing process of the single crystal to prevent the generation of such defect regions.

[0003] As a technique for determining defective regions, for example, Patent Document 1 discloses a technique for evaluating crystal defects in a semiconductor substrate by subjecting the semiconductor substrate to heat treatment, polishing the substrate, and then inspecting the resultant substrate with a surface inspection device.

[0004] Patent Document 2 discloses a technique for evaluating crystal defects in a semiconductor substrate by applying an impurity metal to the semiconductor substrate, subjecting it to heat treatment, polishing it, and then inspecting the polished surface with a surface inspection device.

[0005] Patent Document 3 discloses a technique for measuring the number and / or density of defects on the surface of a silicon single crystal substrate using a surface inspection device, and determining defective regions of the silicon single crystal substrate from the number and / or density of defects.

[0006] JP 2017-220587 A JP 2021-172573 A JP 2017-092400 A

[0007] The techniques for revealing defect regions caused by heat treatment or metal contamination, as typified by Patent Documents 1 and 2, are destructive inspections. Therefore, generally, a portion of a substrate cut out from a crystal is sampled and inspected. However, this method leaves the defect regions in most of the crystal unknown. Therefore, if an undesired defect region exists only in a portion of the crystal, there is a risk that a substrate containing undesired crystal defects may be shipped as a product.

[0008] Furthermore, as a result of a detailed investigation into the technology described in Patent Document 3, it was found that there is a problem in that it may not be possible to determine whether a silicon single crystal substrate contains an I-rich region.

[0009] The present invention has been made to solve the above problems, and an object of the present invention is to provide a method for non-destructively and simply determining the presence or absence of an I-rich region in a semiconductor substrate.

[0010] The present invention has been made to achieve the above-mentioned object, and provides a method for determining a defect region in a silicon single crystal substrate, which determines a defect region in a mirror-polished silicon single crystal substrate with a plane orientation of (100) using a laser scattering type surface inspection device equipped with a rotary stage, the method comprising: measuring crystal defects present in a region of a main surface of the silicon single crystal substrate whose crystal orientation includes the <010> direction and a region of a main surface of the silicon single crystal substrate whose crystal orientation includes the <011> direction using an oblique incidence mode of the surface inspection device; determining the number of defects or the defect density of the crystal defects; and determining whether or not the silicon single crystal substrate includes an I-rich region based on the difference between the number of defects or the defect density of the crystal defects present in the region whose crystal orientation includes the <010> direction and the number of defects or the defect density of the crystal defects present in the region whose crystal orientation includes the <011> direction.

[0011] According to this method for determining a defect region in a silicon single crystal substrate, it is possible to non-destructively and simply determine whether a silicon single crystal substrate with a (100) plane orientation contains an I-rich region.

[0012] In this case, the region for measuring the crystal defects can be within an arbitrary angle range of ±22.5° from the center line of the axis of the crystal orientations of the <010> direction and the <011> direction on the main surface of the silicon single crystal substrate.

[0013] By setting the measurement area to such a range, it is possible to include a sufficient number of crystal defects in the area including each of the directions in order to determine the defect area.

[0014] In this case, the region for measuring the crystal defects can be a region outside a radius of 10 mm from the center of the main surface of the silicon single crystal substrate.

[0015] By setting the measurement region to such a range, it is possible to reduce measurement errors of crystal defects that may occur at the center of the main surface of the silicon single crystal substrate.

[0016] Furthermore, when measuring the crystal defects, the detection size of the crystal defects is also acquired, and the judgment can be made based on the number or density of crystal defects within a detection size range of a portion of the measured crystal defects.

[0017] This makes it possible to determine the defect area with high accuracy even when there is a difference in the number of defects or defect density between the area containing the <010> direction and the area containing the <011> direction of the silicon single crystal substrate only within a specific detection size range.

[0018] At this time, the number or density of the crystal defects can be determined by accumulating the measurement results for a plurality of silicon single crystal substrates having the crystal defects of the same size and density.

[0019] This makes it possible to determine the defect area with high accuracy even when the number of defects per silicon single crystal substrate is small.

[0020] In this case, the main surface of the silicon single crystal substrate can be re-polished and / or cleaned before the measurement.

[0021] This removes defects and particles caused by processing that are present on the main surface of the silicon single crystal substrate, further improving the accuracy of determining defective regions in the silicon single crystal substrate.

[0022] As described above, according to the method for determining a defect region in a silicon single crystal substrate of the present invention, it is possible to non-destructively and simply determine whether a silicon single crystal substrate with a (100) plane orientation contains an I-rich region.

[0023] 1 is a flowchart showing an example of an embodiment of the present invention; and FIG. 2 is a schematic diagram showing an example of a surface inspection apparatus for carrying out an embodiment of the present invention. It is a schematic diagram showing the distribution of defects with a detection size of 13 nm or more in substrate C including an I-rich region, as measured in Example 2. It is a schematic diagram showing the distribution of defects with a detection size of 13 nm or more in substrate B not including an I-rich region, as measured in Example 2. It is a schematic diagram showing the defect distribution in substrate C including an I-rich region. It is a schematic diagram showing the defect distribution in substrate B not including an I-rich region. It is a schematic diagram showing an example of the relationship between defects existing in the

[010] direction of a substrate, whose major axis is aligned with the

[011] direction, and the laser incident surface of a surface inspection apparatus. It is a schematic diagram showing an example of the relationship between defects existing in the

[011] direction of a substrate, whose major axis is aligned with the

[011] direction, and the laser incident surface of a surface inspection apparatus.

[0024] The present invention will be described in detail below, but the present invention is not limited thereto.

[0025] As described above, there has been a demand for a method for non-destructively and simply determining whether or not an I-rich region exists in a silicon single crystal substrate.

[0026] As a result of extensive research into the above-mentioned problems, the present inventors have discovered a method for determining defect regions in a silicon single crystal substrate (hereinafter also simply referred to as a "substrate"), which uses a laser scattering type surface inspection device equipped with a rotary stage to determine defect regions in a mirror-polished silicon single crystal substrate with a (100) crystal orientation, by measuring crystal defects present in a region of the main surface of the silicon single crystal substrate whose crystal orientation includes the <010> direction (hereinafter also simply referred to as the "<010> direction"; the same applies to other orientations) and a region including the <011> direction using an oblique incidence mode of the surface inspection device, and determining the number of defects or the defect density of the crystal defects. The inventors have found that it is possible to non-destructively and simply determine whether a silicon single crystal substrate with a (100) crystal orientation contains an I-rich region by determining the density of the crystal defects present in the region where the crystal orientation includes the <010> direction and judging whether the silicon single crystal substrate contains an I-rich region based on the difference between the number of defects or the defect density of the crystal defects present in the region where the crystal orientation includes the <010> direction (hereinafter simply referred to as "defects / number of defects / defect density in the <010> direction"; the same applies to other orientations) and the number of defects or the defect density of the crystal defects present in the region where the crystal orientation includes the <011> direction, and have completed the present invention.

[0027] The method for determining defect regions in a silicon single crystal substrate of the present invention (hereinafter also simply referred to as the "determination method of the present invention") is, in other words, a determination method comprising: a first step of measuring defects on a main surface of a substrate having a (100) surface in oblique incidence mode using a surface inspection device equipped with a rotary stage; and a second step of deriving the number or defect densities in the <010> and <011> directions of the substrate based on information such as defect positions obtained in the first step, and determining whether or not the substrate contains an I-rich region based on the difference between the two. Figure 1 shows an example of a process flow of the determination method of the present invention.

[0028] [First Step] Figure 2 shows an example of a laser scattering type surface inspection device equipped with a rotary stage, which is used to carry out the determination method of the present invention. As shown in Figure 2, the surface inspection device 1 may include laser light irradiation means 11 that irradiates the surface of the substrate W to be evaluated with laser light at an oblique angle, a detector 12 that detects scattered light of the irradiated laser light that is scattered by defects D or the like on the substrate surface, and a rotary R-θ stage 13 that scans the substrate W with the laser light. An example of such a surface inspection device 1 is the Surfscan SP7XP manufactured by KLA Corporation.

[0029] First, a silicon single crystal substrate having a (100) surface for determining defect regions is prepared, and the surface of the substrate is polished to a mirror finish.

[0030] In this case, if processing-induced defects exist on the substrate surface, they may be removed by re-polishing before measurement. Since processing-induced defects do not depend on the crystal-derived defect region, removing the processing-induced defects improves the accuracy of determining the I-rich region using the determination method of the present invention. Note that the re-polishing method may be a known method.

[0031] After mirror polishing the substrate, it may be cleaned. Cleaning before measurement allows particles adhering to the substrate surface to be removed. Since particle adhesion does not depend on crystal-derived defect regions, removing the particles further improves the accuracy of determining the I-rich region using the determination method of the present invention. Note that cleaning may be performed by a known method.

[0032] Next, the substrate is measured in oblique incidence mode by the surface inspection device 1 having a rotary R-θ stage 13, and the positions of defects on the substrate are obtained.

[0033] The surface inspection device 1 detects foreign matter such as defects by irradiating a substrate with a laser and detecting scattering caused by the foreign matter. In this case, particles may be detected as foreign matter in addition to defects. However, as described above, particles can be sufficiently reduced by cleaning.

[0034] The surface inspection device 1 scans the entire measurement surface of the substrate with a laser by moving the substrate or the laser irradiation position in the radial direction while rotating the substrate using a rotary stage. Position information of the defect on the substrate is obtained from the laser irradiation position on the substrate when the defect is detected.

[0035] The substrate on the rotating stage is measured in oblique incidence mode. In this case, when detecting an anisotropic defect with a distinguishable major and minor axis, the orientation of the defect's major axis relative to the incident light depends not only on the orientation of the defect's major axis on the substrate but also on the defect's circumferential position on the substrate.

[0036] For example, consider a case where the installation direction of the laser light irradiation means 11 in the surface inspection device 1 shown in Figure 2 is set to the 9 o'clock direction, and a silicon single crystal substrate with a (100) surface orientation is measured in oblique incidence mode using the surface inspection device 1 equipped with a rotary stage to detect defects in the 9 o'clock direction.

[0037] When detecting a defect that exists in the

[010] direction of the substrate and whose major axis is aligned with the

[011] direction, the defect is detected when the

[010] direction of the substrate is the 9 o'clock direction of the device, and the major axis at that time is oriented in the 7:30 direction of the device, and the defect is oblique to the direction of laser incidence, as shown in Figure 7.

[0038] On the other hand, when detecting a defect that exists in the

[011] direction of the substrate and whose major axis is aligned with the

[011] direction, the defect is detected when the

[011] direction of the substrate is the 9 o'clock direction of the device, and the orientation of the major axis at that time is the 9 o'clock direction of the device, and the defect is parallel to the incident direction of the laser, as shown in Figure 8.

[0039] Thus, even if the major axis of a defect is aligned with a particular orientation of the substrate, the orientation of the major axis of the defect relative to the incident light when detected will vary depending on the circumferential position of the defect on the substrate.

[0040] Surface inspection devices can obtain the detected size of a foreign particle based on the intensity of scattered light from the foreign particle. The detected size is determined by measuring the intensity of scattered light from a standard particle of known size and determining the relationship between the size of the standard particle and the scattered light intensity. Therefore, depending on the shape, orientation, and composition of the actual foreign particle, the actual size and the detected size may not necessarily match.

[0041] Generally, surface inspection devices determine whether the detected scattered light is a foreign object based on some threshold value. Here, the threshold value can be, for example, the actual scattered light intensity or the minimum size of detection, but for commercially available surface inspection devices, the threshold value can be set based on the specifications of each device.

[0042] By setting a threshold value, it is possible to prevent the scattered light caused by the roughness of the substrate surface from being mistakenly detected as scattered light from a foreign object. On the other hand, if the intensity of the scattered light from a foreign object is lower than the threshold value, for example, when the actual size of the foreign object is small or the foreign object is not oriented appropriately relative to the incident light, the detection accuracy of the foreign object may decrease.

[0043] When measuring a substrate using the surface inspection device 1, not only the defect positions on the substrate but also the detected size of the defects may be acquired. By acquiring the detected size, it is possible to limit the detected size to a specific range when deriving the number of defects or the defect density, and it is possible to determine the defective area with higher accuracy.

[0044] [Second step] Next, based on the position information of the defects on the substrate obtained in the first step, the crystal orientation of the substrate is classified as to whether it exists in a region including the <010> direction, a region including the <011> direction, or a region other than the above.

[0045] Here, the <010> direction of the substrate refers to all four equivalent directions,

[010] ,

[001] , [0-10], and [00-1], relative to the center of the substrate, and the <011> direction refers to all four equivalent directions,

[011] , [0-11], [0-1-1], and [01-1], relative to the center of the substrate.

[0046] Instead of the above, one of the equivalent orientations (for example,

[010] and

[011] ) may be used as a representative for the <010> direction of the substrate and the <011> direction of the substrate.

[0047] The region including the <010> direction and the region including the <011> direction on the main surface of the substrate, where defects are measured, can be not only on the axes strictly corresponding to the <010> direction and the <011> direction, but also in any angle range within ±22.5° from the center line, with the axes as the reference center line. By taking such a range, a sufficient number of defects can be included in the region including each direction to determine the defect region.

[0048] Here, when both the region including the <010> direction of the substrate and the region including the <011> direction of the substrate are set within a range of ±22.5° from the center line, all defects, except for defects located exactly at the center of the substrate surface, are included in the region including the <010> direction of the substrate or the region including the <011> direction.

[0049] The region of the substrate where defects are measured may be a specific radial range. For example, such a radial range may be, but is not limited to, a region outside a radius of 10 mm from the center of the main surface of the substrate. Because the distance between the <010> direction and the <011> direction is short near the center of the substrate, a slight deviation in the defect position on the substrate measured by the surface inspection device may cause an error in which the defect is recorded in a region including a direction different from the actual defect. However, by setting the measurement region to such a range, the error can be reduced.

[0050] Furthermore, the specific radial range may be a range excluding the outermost peripheral region of the substrate. Because the outer periphery of the substrate is susceptible to shape changes due to chamfering and particles that were not completely removed by cleaning, excluding the outermost peripheral region of the substrate can further reduce judgment errors. Such a radial range may be, for example, within 2 mm from the outer periphery, but is not particularly limited thereto.

[0051] Next, the number of defects classified as existing in the region containing the <010> direction of the substrate is compared with the number of defects classified as existing in the region containing the <011> direction of the substrate.

[0052] In the determination method of the present invention, it is possible to determine whether or not a substrate contains an I-rich region based on the difference in the number of defects in the <010> direction and the <011> direction of the substrate.

[0053] As a result of the investigation, it was found that the surface of a substrate with a (100) plane orientation of silicon single crystal containing an I-rich region contains multiple types of high-aspect-ratio anisotropic defects, with the ratio of the length of the long axis to the length of the short axis exceeding 1:20, and that the long axes of these defects are aligned along specific directions for each type of defect (hereinafter, the high-aspect-ratio anisotropic defects may be collectively referred to as "I-rich linear defects"). Furthermore, it was found that the detection rate of at least some I-rich linear defects when measured in oblique incidence mode with a surface inspection device having a rotation stage differs depending on whether they are present in the <010> direction or the <011> direction of the substrate.

[0054] Even if the long axis of a defect is aligned with a specific orientation of the substrate, the orientation of the long axis of the defect relative to the incident light when it is detected changes depending on the circumferential position of the defect on the substrate. As a result, it is thought that the scattered light intensity and the detection rate of certain I-rich linear defects change depending on whether they are present in the <010> direction or the <011> direction of the substrate.

[0055] On the other hand, in defect regions other than the I-rich region, no crystal defects with an aspect ratio exceeding 1:20 were detected, and therefore it was found that there was no difference in the number of defects and defect density in the <010> direction and the <011> direction of the substrate. Therefore, by comparing the number of defects or defect density in the <010> direction and the <011> direction of the substrate, if there is a difference in the number of defects or defect density, it can be determined that the substrate contains an I-rich region.

[0056] Instead of comparing the numbers of all defects classified into each direction, the detection size of the defects may also be acquired in the first step, and the defect area may be determined based on the number of defects within a certain detection size range among the measured defects. By limiting the detection size, it is possible to determine with high accuracy whether an I-rich area is included, even if a difference in the number of defects in the <010> direction and the <011> direction of the substrate is observed only within a specific detection size range.

[0057] The specific range of detection sizes can be determined by comparing the number of defects in a certain range of detection sizes in the <010> direction with the number of defects in the same range of detection sizes in the <011> direction for a substrate that actually includes an I-rich region, and determining any range of detection sizes for which a difference is found. The range of detection sizes determined in this way can also be applied to other substrates that are considered to be generally identical in terms of levels other than the defect regions.

[0058] As a method for comparing the number of defects, for example, a schematic diagram plotting defect positions on a substrate, as shown in Figure 3, is prepared, and the number of defects in the <010> direction of the substrate is compared with the number of defects in the <011> direction. In this case, if there is a significant difference between the two, a visual comparison can be made. In the case of Figure 3, it is visually clear that the number of defects in the <011> direction is greater than the number of defects in the <010> direction, and a difference is observed. Therefore, the substrate is determined to contain an I-rich region.

[0059] Alternatively, instead of the above comparison method, a specific threshold value relating to the difference in the number of defects for determining whether a substrate contains an I-rich region can be determined. It is also possible to determine this based on the presence or absence of a statistically significant difference. Furthermore, for substrates that actually contain I-rich regions and substrates that do not contain I-rich regions, a threshold value (both when b / a is greater than or smaller than 1) can be determined by referring to the ratio b / a of the number of defects b in the <011> direction to the number of defects a in the <010> direction.

[0060] For example, when b / a is greater than 1, the threshold value can be b / a≧1.1, but is not particularly limited and may be set appropriately depending on the purpose of the judgment and the total number of defects on the substrate.

[0061] For example, when it is desired to determine that a substrate contains an I-rich region without omitting all substrates that contain an I-rich region, it is preferable to set the threshold value closer to 1 (for example, b / a≧1.05). This makes it possible to determine with higher accuracy that a substrate contains an I-rich region even when the ratio of other regions to the I-rich region is high.

[0062] Conversely, if it is desired to reduce the possibility of a substrate that does not contain an I-rich region being erroneously determined to contain an I-rich region, which may occur due to variations in the number of defects in each direction, it is preferable to set the threshold value to be greater than 1.1 (for example, b / a≧1.2). In particular, when the number of defects is small, the substrate is more susceptible to the effects of variations, so it is more preferable to set the threshold value to be greater than 1.1.

[0063] Similarly, when b / a is smaller than 1, the threshold value can be set to b / a≦0.9, but is not particularly limited thereto and may be set appropriately depending on the purpose of the judgment and the total number of defects on the substrate.

[0064] For example, when it is desired to determine that a substrate contains an I-rich region without omitting all substrates that contain an I-rich region, it is preferable to set the threshold value closer to 1 (for example, b / a≦0.95). This makes it possible to determine with higher accuracy that a substrate contains an I-rich region even when the ratio of other regions to the I-rich region is high.

[0065] Conversely, if it is desired to reduce the possibility of a substrate that does not contain an I-rich region being erroneously determined to contain an I-rich region, which may occur due to variations in the number of defects in each direction, it is preferable to set the threshold value to be smaller than 0.9 (for example, b / a≦0.8). In particular, when the number of defects is small, the substrate is more susceptible to the effects of variations, so it is more preferable to set the threshold value to be smaller than 0.9.

[0066] Instead of the number of defects, if a difference in defect density between the <010> direction and the <011> direction of the substrate is found, it can also be determined that the substrate contains an I-rich region. This allows accurate determination of defect regions even when the angle ranges of the region containing the <010> direction and the region containing the <011> direction are different.

[0067] In this case, as a specific threshold value relating to the difference in defect density for determining whether a substrate contains an I-rich region, the ratio b' / a' of the defect density b' in the <011> direction to the defect density a' in the <010> direction can be referenced for a substrate that actually contains an I-rich region and a substrate that does not contain an I-rich region, and the threshold value (both when b' / a' is greater than 1 and when it is less than 1) can be determined. For example, the same value as the threshold value obtained from the number of defects can be used as the threshold value relating to the difference in defect density.

[0068] Alternatively, the number of defects or defect density may be determined by accumulating the measurement results for multiple substrates having defects of the same size and density. By making a determination based on the measurement results for multiple substrates of the same level, it is possible to determine the defect area with high accuracy even when the number of defects per substrate is small.

[0069] The present invention will be specifically explained below with reference to examples, but the present invention is not limited to these examples.

[0070] Example 1 First, two cleaned mirror-polished silicon single crystal substrates with a diameter of 300 mm and a plane orientation of (100), which were manufactured using the Czochralski method under different conditions, were prepared and designated as Substrate A and Substrate B, respectively.

[0071] Next, foreign particles present on the main surfaces of substrates A and B were measured using a surface inspection device (SP7XP, Oblique mode (oblique incidence), 11 nm up) equipped with a rotary stage.

[0072] Next, for substrates A and B, the number of defects present in the region including the <010> direction and the region including the <011> direction of the substrate was examined.

[0073] Here, the region including the <010> direction of the substrate was defined as all four equivalent orientations of

[010] ,

[001] , [0-10], and [00-1] relative to the center of the substrate, and a range of ±22.5° therefrom. Similarly, the region including the <011> direction of the substrate was defined as all four equivalent orientations of

[011] , [0-11], [0-1-1], and [01-1] relative to the center of the substrate, and a range of ±22.5° therefrom.

[0074] In this case, the threshold values ​​for determining whether a substrate contains an I-rich region, relating to the difference in the number of defects or defect density present in the region containing the <010> direction of the substrate and the region containing the <011> direction, were set to 0.9 and 1.1, and the substrate was determined to contain an I-rich region when the ratio of the number of defects or defect density present in the region containing the <011> direction of the substrate to the number of defects or defect density present in the region containing the <010> direction of the substrate was smaller than 0.9 or greater than 1.1.

[0075] As a result, the number of defects in the region of Substrate A containing the <010> direction was 707, while the number of defects in the region containing the <011> direction was 379. Therefore, the ratio of the number of defects in the <011> direction to the number of defects in the <010> direction of the substrate was 379 / 707 ≒ 0.536 < 0.9, and therefore Substrate A was determined to contain an I-rich region.

[0076] In substrate B, the number of defects present in the region including the <010> direction was 23,029, while the number of defects present in the region including the <011> direction was 23,575. Therefore, the ratio of the number of defects present in the region including the <011> direction to the number of defects present in the region including the <010> direction of the substrate was 23,575 / 23,029 ≈ 1.024, which was 0.9 < 1.024 < 1.1, and therefore substrate B was determined to not contain an I-rich region.

[0077] The defect areas of each adjacent substrate were determined using a known revealing etching method, and it was confirmed that substrate A contained an I-rich region, while substrate B was entirely V-rich and did not contain an I-rich region, confirming that the defect areas had been correctly evaluated.

[0078] The above results demonstrate that the method of determining defect regions in silicon single crystal substrates according to the present invention is effective in determining I-rich regions.

[0079] Example 2 A mirror-polished silicon single crystal substrate having a diameter of 300 mm, which was manufactured using the Czochralski method, was cleaned and used as substrate C.

[0080] Next, defects present on the main surfaces of substrate C and substrate B of Example 1 were measured using a surface inspection device equipped with a rotary stage (SP7XP, Oblique mode (low angle incidence), 11 nm up). The distribution of defects on substrate C is shown in FIG. 5, and the distribution of defects on substrate B is shown in FIG. 6.

[0081] Next, for substrates B and C, schematic diagrams were prepared in which the defect positions were plotted for defects with a detection size of 13 nm or larger, and the number of defects present in the region including the <010> direction of the substrate was visually compared with the number of defects present in the region including the <011> direction.

[0082] Here, the region including the <010> direction of the substrate was defined as all four equivalent orientations of

[010] ,

[001] , [0-10], and [00-1] relative to the center of the substrate, and a range of ±10° therefrom. Similarly, the region including the <011> direction of the substrate was defined as all four equivalent orientations of

[011] , [0-11], [0-1-1], and [01-1] relative to the center of the substrate, and a range of ±10° therefrom.

[0083] Furthermore, the threshold values ​​for determining whether a substrate contains an I-rich region, relating to the difference in the number of defects or defect density present in the region containing the <010> direction of the substrate and the region containing the <011> direction, were set to 0.9 and 1.1, and the substrate was determined to contain an I-rich region when the ratio of the number of defects or defect density present in the region containing the <011> direction of the substrate to the number of defects or defect density present in the region containing the <010> direction of the substrate was smaller than 0.9 or greater than 1.1.

[0084] The distribution of defects with a detection size of 13 nm or larger in Substrate C is shown in Figure 3. As can be seen from Figure 3, the number of defects present in the region including the <011> orientation was visually clearly greater than the number of defects present in the region including the <010> orientation, and a difference was observed, so Substrate C was determined to contain an I-rich region.

[0085] The defect density in the region including each direction was found to be 0.140 defects / cm . ２ , the defect density in the region including the <011> direction is 0.879 / cm ２The ratio of the defect density in the region including the <011> direction to the defect density in the region including the <010> direction of the substrate was 0.879 / 0.140=6.279>1.1, so it was determined that substrate C included an I-rich region.

[0086] In fact, when the defective regions of the substrate adjacent to substrate C were determined by a known revealing etching method, it was confirmed that substrate C was an I-rich region over the entire surface.

[0087] The distribution of defects with a detection size of 13 nm or larger for Substrate B is shown in Figure 4. From Figure 4, no apparent difference was visually observed between the number of defects present in the region containing the <010> orientation and the number of defects present in the region containing the <011> orientation.

[0088] The defect density in the region including each direction was calculated, and the defect density in the region including the <010> direction of the substrate was 45.35 / cm ２ , the defect density in the region including the <011> direction is 46.75 / cm ２ The ratio of the defect density present in the region including the <011> direction to the defect density present in the region including the <010> direction of the substrate was 46.75 / 45.35 = 1.031, which was 0.9 < 1.031 < 1.1, so it was determined that substrate B did not include an I-rich region. Note that in Example 1, substrate B was confirmed to have a V-rich region over the entire surface and not include an I-rich region by a known manifestation etching method.

[0089] From the above results, it is possible to determine the presence or absence of an I-rich region based on the difference in defect density between the region containing the <010> direction and the region containing the <011> direction of the substrate, demonstrating that the determination of an I-rich region using the method of the present invention for determining a defect region in a silicon single crystal substrate is effective.

[0090] Furthermore, for all detected defects in substrates C and B, the ratio of the defect density present in the region including the <011> direction to the defect density present in the region including the <010> direction of the substrate was calculated. As a result, the defect density ratio was 0.897 for substrate C and 1.018 for substrate B, and it was determined that substrate C contained an I-rich region, while substrate B did not. Furthermore, when the detection size was limited, the defect density ratio for substrate C was considerably larger than the threshold value, indicating that limiting the defect size is effective for determining I-rich regions. Note that the magnitudes of the defect density ratios for substrates C and B were reversed when the detection size was limited and when it was not. This is because substrate C contained two types of defects with large aspect ratios, each with a different average size and major axis orientation.

[0091] (Comparative Example) Defects present on the main surfaces of Substrate C and Substrate B used in Example 2 were measured using a surface inspection device (SP7XP, Oblique mode, 11 nm Up), and the defect positions were plotted from the obtained position information. The distribution of defects on Substrate C is shown in Figure 5, and the distribution of defects on Substrate B is shown in Figure 6.

[0092] An attempt was made to determine defective areas for substrates B and C based on the technology described in Patent Document 3, and although the total number of defects differed, both were determined to be entirely V-rich areas because defects were present across the entire surface of the substrate. Substrate B, which was entirely V-rich, was accurately determined, but substrate C, which included an I-rich area, resulted in a determination result that was different from the actual situation.

[0093] As described above, according to the examples of the present invention, it was possible to accurately determine whether or not a silicon single crystal substrate contains an I-rich region.

[0094] This specification includes the following aspects: [1]: A method for determining a defect region in a mirror-polished silicon single crystal substrate having a (100) plane orientation, using a laser scattering surface inspection device equipped with a rotary stage, the method comprising: measuring crystal defects present in a region of a main surface of the silicon single crystal substrate whose crystal orientation includes a <010> direction and a region of a main surface of the silicon single crystal substrate whose crystal orientation includes a <011> direction using an oblique incidence mode of the surface inspection device, determining the number of defects or the defect density of the crystal defects, and determining whether or not the silicon single crystal substrate includes an I-rich region based on the difference between the number of defects or the defect density of the crystal defects present in the region whose crystal orientation includes a <010> direction and the number of defects or the defect density of the crystal defects present in the region whose crystal orientation includes a <011> direction. [2]: The method for determining a defect region in a silicon single crystal substrate according to [1] above, including: the region for measuring crystal defects is within a range of any angle within ±22.5° from an axis of the main surface of the silicon single crystal substrate whose crystal orientations are <010> and <011>, with the axis serving as a centerline. [3]: The method for determining a defect region in a silicon single crystal substrate according to [1] above or [2] above, including: the region for measuring crystal defects is within a radius of 10 mm from the center of the main surface of the silicon single crystal substrate. [4]: ​​The method for determining a defect region in a silicon single crystal substrate according to [1] above, [2] or [3] above, including: when measuring the crystal defects, also obtaining detection sizes of the crystal defects, and making the determination based on the number or defect density of crystal defects within a detection size range of a portion of the measured crystal defects. [5]: The method for determining a defect area in a silicon single crystal substrate according to [1], [2], [3] or [4], comprising accumulating the measurement results for a plurality of silicon single crystal substrates having crystal defects of the same level of size and density to determine the number of defects or defect density of the crystal defects. [6]: The method for determining a defect area in a silicon single crystal substrate according to [1], [2], [3], [4] or [5], comprising re-polishing and / or cleaning the main surface of the silicon single crystal substrate before the measurement.

[0095] The present invention is not limited to the above-described embodiments. The above-described embodiments are merely examples, and anything that has substantially the same configuration as the technical idea described in the claims of the present invention and that exhibits similar effects is included within the technical scope of the present invention.

Claims

1. A method for determining a defective region of a silicon single crystal substrate, which uses a laser scattering type surface inspection apparatus equipped with a rotary stage to determine a defective region of a silicon single crystal substrate with a mirror-polished surface orientation (100), comprising: measuring crystal defects present in regions of the main surface of the silicon single crystal substrate that include the <010> direction and the <011> direction in the crystal orientation by an oblique incidence mode of the surface inspection apparatus, obtaining the number of defects or defect density of the crystal defects, and determining whether the silicon single crystal substrate contains an I-rich region based on the difference between the number of defects or defect density of the crystal defects present in the region where the crystal orientation includes the <010> direction and the number of defects or defect density of the crystal defects present in the region where the crystal orientation includes the <011> direction.

2. The method for determining a defective region of a silicon single crystal substrate according to claim 1, wherein the region for measuring the crystal defects is an arbitrary angular range within ±22.5° from a center line with axes in the <010> direction and the <011> direction of the main surface of the silicon single crystal substrate as the center line.

3. The method for determining a defective region of a silicon single crystal substrate according to claim 1, wherein the region for measuring the crystal defects is a region outside a radius of 10 mm from the center of the main surface of the silicon single crystal substrate.

4. The method for determining a defective region of a silicon single crystal substrate according to claim 1, wherein when measuring the crystal defects, the detection size of the crystal defects is also obtained, and the determination is made based on the number of defects or defect density of the crystal defects within a detection size range of a part of the measured crystal defects.

5. The method for determining a defective region of a silicon single crystal substrate according to claim 1, wherein the measurement results for a plurality of silicon single crystal substrates having the same level of size and density of the crystal defects are accumulated to obtain the number of defects or defect density of the crystal defects.

6. The method for determining a defective region of a silicon single crystal substrate according to any one of claims 1 to 5, wherein the main surface of the silicon single crystal substrate is re-polished and / or cleaned before the measurement.

Images