Method for evaluating oxide film breakdown voltage characteristics of silicon single crystal substrates and method for determining pass / fail of silicon single crystal substrates

Abstract

To provide an evaluation method for oxide film breakdown voltage characteristics of a silicon single crystal substrate and an acceptance determination method that can easily and highly accurately evaluate the oxide film breakdown voltage characteristics.SOLUTION: An evaluation method for oxide film breakdown voltage characteristics of a silicon single crystal substrate includes: a step 1 of detecting, by a surface defect inspection device, a defect of the silicon single crystal substrate and acquiring defect coordinates; a step 2 of observing the defect of the silicon single crystal substrate through a scanning type electron microscope based upon the defect coordinates to acquire a defect image, and classifying a defective seed; a step 3 of cleaning, with hydrofluoric acid, the silicon single crystal substrate; a step 4 of observing the defect through the scanning type electron microscope based upon the defect coordinates acquired in the first step after the hydrofluoric acid cleaning, and classifying the defective seed; a step 5 of comparing the defect images with the defective seeds acquired in the step 2 and step 4 to extract oxygen deposits; and a step 6 of extracting a defect which causes oxide film breakdown voltage deterioration through the steps 1 to 5, and generating a pseudo oxide film breakdown voltage deterioration map based upon coordinates of the oxide film breakdown voltage deterioration defect.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for evaluating the dielectric strength characteristics of an oxide film of a silicon single crystal substrate and a method for determining the acceptability of a silicon single crystal substrate. [Background technology]

[0002] In recent years, with the miniaturization and high integration of semiconductor elements, the control and accurate evaluation of crystal defects in semiconductor crystals has become more important. Semiconductor devices formed on silicon single crystal substrates use insulating films such as gate oxide films, but it is widely known that the presence of defects in the silicon single crystal substrate reduces the quality of the insulating film. Therefore, high-quality insulating films are necessary to form high-quality semiconductor devices.

[0003] One method for evaluating the reliability of insulating films is GOI (Gate Oxide Integrity). This involves forming an oxide film on a silicon single crystal substrate and then forming electrodes on the oxide film to form a MOS (Metal Oxide Semiconductor) structure. A high electric field is applied to this MOS to destroy the oxide film, and the breakdown field strength is measured to evaluate the oxide film reliability. GOI evaluation is characterized by its ability to detect and evaluate crystal defects (COPs (Crystal Originated Particles), oxygen precipitates, etc.) and processing-induced defects (scratches, etc.) with high accuracy. However, in order to evaluate the GOI, it is necessary to fabricate a MOS structure, which requires a complex and lengthy MOS formation process, resulting in the drawback of being time-consuming and costly.

[0004] Patent Document 1 discloses a method for estimating the oxide film breakdown voltage failure rate using a foreign matter inspection device and a scanning electron microscope. Patent Document 2 discloses a method for evaluating crystal defects inside a substrate by performing anisotropic etching and exposing etching residues caused by crystal defects. Patent Document 3 discloses a method for distinguishing between defects caused by processing and defects caused by metal contamination by immersing a silicon wafer for a long period of time in an etching solution consisting of ammonia and hydrogen peroxide solution. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2020-161555 [Patent Document 2] Japanese Patent Application Laid-Open No. 2005-257576 [Patent Document 3] Japanese Patent Application Laid-Open No. 2005-063984 Summary of the Invention [Problem to be solved by the invention]

[0006] However, Patent Document 1 does not disclose a method for distinguishing oxygen precipitates from particles, and it is not possible to distinguish between GOI defects caused by oxygen precipitates, which are crystal defects, and GOI defects caused by particles that adhere during the processing process, so the accuracy of estimating the cause of GOI defects is insufficient.In Patent Documents 2 and 3, because the silicon single crystal substrate is etched, it is not possible to determine whether defects that existed in the silicon single crystal substrate before etching were oxygen precipitates.

[0007] The present invention has been made to solve the above problems, and an object of the present invention is to provide a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate, which can easily and accurately evaluate the oxide film breakdown voltage characteristics, and a method for determining the pass / fail of a silicon single crystal substrate, which can easily and accurately determine the pass / fail of a silicon single crystal substrate. More specifically, an object of the present invention is to provide a method for determining the pass / fail of a silicon single crystal substrate by distinguishing between crystal defects, particularly oxygen precipitates, and processing-induced particles on the surface of the silicon single crystal substrate and easily and accurately evaluating the reliability of an oxide film formed on the surface of the silicon single crystal substrate. [Means for solving the problem]

[0008] The present invention has been made to achieve the above-mentioned object, and provides a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate, comprising: a first step of detecting defects in the silicon single crystal substrate using a surface defect inspection device and acquiring defect coordinates; a second step of observing the defects in the silicon single crystal substrate using a scanning electron microscope based on the defect coordinates, acquiring defect images, and classifying the defect types; a third step of cleaning the silicon single crystal substrate with hydrofluoric acid; a fourth step of observing the defects using a scanning electron microscope after the hydrofluoric acid cleaning, based on the defect coordinates acquired in the first step, acquiring defect images, and classifying the defect types; a fifth step of comparing the defect images and defect types acquired in the second step and the fourth step to extract oxygen precipitates; and a sixth step of extracting defects that cause oxide film breakdown voltage degradation from the first to fifth steps, and creating a pseudo oxide film breakdown voltage degradation map based on the coordinates of the oxide film breakdown voltage degradation defects.

[0009] According to such a method for evaluating the dielectric strength of an oxide film of a silicon single crystal substrate, oxygen precipitates can be extracted easily and with high precision, and the dielectric strength of an oxide film can be evaluated easily and with high precision.

[0010] At this time, the defect types classified in the second and fourth steps can be particles, pits, scratches, and PIDs (Polished Induced Defects).

[0011] This allows oxygen precipitates to be extracted in the fifth step. The defect classification in the second step may be performed by visually inspecting the scanning electron microscope image or by automatic classification using Rule-Based Binning (RBB). More preferably, automatic classification can be performed with high accuracy and high throughput using a Convolutional Neural Network (CNN).

[0012] At this time, in the fifth step of extracting the oxygen precipitates, defects classified as particles in the second step and as pits in the fourth step are reclassified as oxygen precipitates, and among the defects classified as pits in the second step and as pits in the fourth step, defects whose pit depth in the fourth step is deeper than in the second step or whose inner diameter in the fourth step is larger than in the second step can be reclassified as oxygen precipitates.

[0013] This allows the oxygen precipitates to be reclassified. Because oxygen precipitates are selectively etched by hydrofluoric acid to form pits, in the fifth step, defects classified as particles in the second step and as pits in the fourth step (after hydrofluoric acid cleaning) are reclassified as oxygen precipitates, thereby making it possible to distinguish between particles and oxygen precipitates. Although some oxygen precipitates are etched to form pits during the silicon single crystal substrate manufacturing process, some unetched oxygen precipitates may remain at the bottom of the pits. Therefore, among the defects classified as pits in the second step and as pits in the fourth step, defects whose pit depth after hydrofluoric acid cleaning is greater than before cleaning or whose inner diameter after hydrofluoric acid cleaning is greater than before cleaning can be reclassified as oxygen precipitates.

[0014] In this case, the defects that cause the deterioration of the oxide film breakdown voltage can be pits, oxygen precipitates, or scratches.

[0015] This allows for highly accurate evaluation of oxide breakdown voltage characteristics by creating a pseudo oxide breakdown voltage degradation map based on the coordinates of the defects that cause the oxide breakdown voltage degradation. The format of the oxide breakdown voltage degradation map is not particularly limited, but it is preferably a map that imitates the oxide breakdown voltage degradation map obtained by GOI evaluation.

[0016] The present invention has been made to achieve the above-mentioned object, and provides a method for determining the acceptability of a silicon single crystal substrate using the above-described method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate, characterized in that an oxide film breakdown voltage defect rate is calculated from the pseudo oxide film breakdown voltage degradation map created in the sixth step, and when the oxide film breakdown voltage defect rate exceeds a predetermined predetermined value, the silicon single crystal substrate is determined to be unacceptable.

[0017] According to such a method for determining the acceptability of a silicon single crystal substrate, it is possible to evaluate the oxide film breakdown voltage characteristics with high accuracy and high throughput, and to prevent silicon substrates with poor oxide film breakdown voltage from being shipped with high accuracy. [Effects of the Invention]

[0018] As described above, according to the method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate of the present invention, oxygen precipitates can be extracted easily and with high accuracy, and the oxide film breakdown voltage characteristics can be evaluated easily and with high accuracy. According to the method for determining the acceptability of a silicon single crystal substrate of the present invention, it is possible to easily and highly accurately determine the acceptability of a silicon single crystal substrate. The oxide film breakdown voltage characteristics can be evaluated with high accuracy and high throughput, and it is possible to accurately prevent the shipment of silicon substrates with poor oxide film breakdown voltage characteristics. By using the method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate and the method for determining the acceptability of a silicon single crystal substrate of the present invention, it is possible to evaluate the oxide film breakdown voltage characteristics with high precision and high throughput, determine the acceptability of a silicon single crystal substrate, and prevent the shipment of silicon single crystal substrates with poor oxide film breakdown voltage characteristics with high precision. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a flowchart showing an example of a method for determining the acceptability of a silicon single crystal substrate according to the present invention. [Figure 2A] 1 is an example of the experimental results obtained in Example 1, and is an oxide film breakdown voltage degradation map obtained by actual GOI measurement. [Figure 2B] 1 is an example of the experimental results obtained in Example 1, showing a pseudo oxide film breakdown voltage degradation map obtained by the process of the present invention. [Figure 3A] 10 is an example of the experimental results obtained in Example 2, and is an oxide film breakdown voltage degradation map obtained by actual GOI measurement. [Figure 3B] 10 is an example of the experimental results obtained in Example 2, showing a pseudo oxide film breakdown voltage degradation map obtained by the process of the present invention. [Figure 4] 1 is a table showing defect examples 1 to 3. [Figure 5] 10 is a table showing the experimental results of Comparative Example 1, including defect types (defect types), weighted average values ​​of GOI defect rates, the number of defects, the number of GOI defect defects, and the (estimated) GOI defect rate. [Figure 6] 1 is a table showing the GOI defect rate calculated from the present technology and GOI measurement results in Examples 1 and 2, and the GOI defect rate calculated using the conventional technology (Comparative Example 1). DETAILED DESCRIPTION OF THE INVENTION

[0020] The present invention will be described in detail below, but the present invention is not limited thereto.

[0021] As described above, there has been a demand for a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate, which can easily and accurately evaluate the oxide film breakdown voltage characteristics, and a method for determining the pass / fail of a silicon single crystal substrate, which can easily and accurately determine the pass / fail of a silicon single crystal substrate.

[0022] As a result of extensive research into the above-mentioned problems, the present inventors have discovered a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate, comprising: a first step of detecting defects in a silicon single crystal substrate with a surface defect inspection device and acquiring defect coordinates; a second step of observing the defects in the silicon single crystal substrate with a scanning electron microscope based on the defect coordinates, acquiring defect images, and classifying the defect types; a third step of cleaning the silicon single crystal substrate with hydrofluoric acid; and a third step of observing the defects with a scanning electron microscope based on the defect coordinates acquired in the first step after the hydrofluoric acid cleaning, a fourth step of acquiring an image of the oxide film and classifying the defect types; a fifth step of comparing the defect images and defect types acquired in the second step and the fourth step to extract oxygen precipitates; and a sixth step of extracting defects that cause oxide film breakdown voltage degradation from the first to fifth steps and creating a pseudo oxide film breakdown voltage degradation map based on the coordinates of the oxide film breakdown voltage degradation defects, thereby enabling the oxide film breakdown voltage characteristics to be extracted easily and with high accuracy, and The present inventors have found that a method for determining the acceptability of a silicon single crystal substrate using the above-described method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate, characterized in that an oxide film breakdown voltage defect rate is calculated from the pseudo oxide film breakdown voltage degradation map created in the sixth step, and when a predetermined oxide film breakdown voltage defect rate is exceeded, the silicon single crystal substrate is determined to be unacceptable, thereby enabling a simple and highly accurate determination of the acceptability of a silicon single crystal substrate, and have completed the present invention.

[0023] Hereinafter, a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate and a method for determining the acceptability of a silicon single crystal substrate according to the present invention will be described with reference to FIG.

[0024] (Method for evaluating oxide film breakdown voltage characteristics of silicon single crystal substrates and method for determining pass / fail of silicon single crystal substrates according to the present invention) FIG. 1 is a flowchart showing an example of a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to the present invention and a method for determining the acceptability of a silicon single crystal substrate according to the present invention. 1 are an example of a method for determining the pass / fail of a silicon single crystal substrate according to the present invention. Among these, steps 1 to 6 are an example of a method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to the present invention. Therefore, the method for determining the pass / fail of a silicon single crystal substrate according to the present invention can be said to be a method for determining the pass / fail of a silicon single crystal substrate using the method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to the present invention.

[0025] (1st step) The first step is to detect defects in a silicon single crystal substrate using a surface defect inspection device and obtain defect coordinates. First, a silicon single crystal substrate to be evaluated is prepared. The silicon single crystal substrate to be evaluated may be manufactured by the Czochralski method or the floating zone method. Furthermore, there are no particular restrictions on the crystal orientation. Defects on this silicon single crystal substrate are detected and the defect coordinates are obtained using a surface defect inspection device. The surface defect inspection device can be a particle counter, for example, a KLA Surfscan SP5, which can be used to measure in oblique mode at 19 nm up.

[0026] (2nd process) The second step is a step of observing the defects with a scanning electron microscope based on the defect coordinates obtained in the first step, obtaining scanning electron microscope images of the defects (defect images), and classifying the defect types. Since defect types are classified based on scanning electron microscope images, it is preferable to classify them based on defect shape and height information, and they can be classified into particles (convex defects), pits (concave defects), scratches (linear concave defects), and PIDs (Polished Induced Defects, which are wide and short defects). In this case, since particles and pits are defect types that may be classified as oxygen precipitates, when identifying classified defect types as oxygen precipitates, particles and pits must be included in the classification of defect types.

[0027] (3rd step) The third step is to clean the silicon single crystal substrate with hydrofluoric acid. In this step, the silicon single crystal substrate is cleaned with hydrofluoric acid in order to etch away oxygen precipitates present on the surface of the silicon single crystal substrate and turn them into pits. The concentration of hydrofluoric acid and the cleaning time are not particularly limited. The hydrofluoric acid concentration and cleaning time may be selected so long as they sufficiently etch away oxygen precipitates present on the silicon single crystal substrate. However, from the viewpoint of safety and operation, the hydrofluoric acid concentration is preferably 50 wt% or less and the cleaning time is 10 minutes or less. It is more preferable that the mass percent concentration of hydrofluoric acid is 0.5 to 10 wt% and the cleaning time is 1 to 5 minutes. Furthermore, in order to prevent particle adhesion due to dust generation during transportation, a thin oxide film may be formed using hydrogen peroxide water or ozone water after the hydrofluoric acid cleaning.

[0028] (4th step) The fourth step is a step in which, after the hydrofluoric acid cleaning, the defects are observed using a scanning electron microscope based on the defect coordinates obtained in the first step, a scanning electron microscope image of the defect (defect image) is obtained, and the defect type is classified using the same criteria as in the second step.

[0029] (5th step) The fifth step is a step of extracting oxygen precipitates by comparing the defect images and defect types obtained in the second and fourth steps. Oxygen precipitates are selectively etched by hydrofluoric acid and become pits, so they are classified as particles in the second step, and defects classified as pits after hydrofluoric acid cleaning in the fourth step are reclassified as oxygen precipitates, making it possible to distinguish between particles and oxygen precipitates. Furthermore, while some of the oxygen precipitates present on the silicon single crystal substrate are etched to form pits during the silicon single crystal substrate manufacturing process, some oxygen precipitates that were not etched may remain at the bottom of the pits. Selective etching of these remaining oxygen precipitates with hydrofluoric acid deepens the depth of the pits or increases the inner diameter of the pits after hydrofluoric acid cleaning. Therefore, among the defects classified as pits in the second step and those classified as pits in the fourth step, defects whose pit depth after hydrofluoric acid cleaning is deeper than before cleaning and defects whose pit inner diameter after hydrofluoric acid cleaning is larger than before cleaning are reclassified as oxygen precipitates. Defects other than oxygen precipitates are not etched by hydrofluoric acid, so there is no distinctive change in the defect images obtained before and after hydrofluoric acid cleaning. Therefore, oxygen precipitates can be extracted by reclassifying them. Extracting oxygen precipitates makes it possible to separate particles attached during the processing process from oxygen precipitates, as well as pits caused by COPs or processing from pits caused by oxygen precipitates, improving the accuracy of estimating the cause of GOI defects and the accuracy of oxide film breakdown voltage evaluation.

[0030] (6th step) The sixth step is a step of extracting defects that cause degradation of the oxide film breakdown voltage from the first to fifth steps, and creating a pseudo oxide film breakdown voltage degradation map based on the coordinates of the defects that cause degradation of the oxide film breakdown voltage. At this time, a preliminary investigation into the influence of defects present on the silicon single crystal substrate on the oxide film breakdown voltage revealed that pits, oxygen precipitates, and scratches have a significant effect on the oxide film breakdown voltage. Therefore, the defects that cause deterioration in the oxide film breakdown voltage can be determined to be pits, oxygen precipitates, and scratches. The format of the oxide breakdown voltage degradation map is not particularly limited, but it is preferable that the map be a map that imitates the oxide breakdown voltage degradation map obtained by GOI evaluation. In this case, if it is desired to evaluate the failure mode of GOI, which is more strongly influenced by oxygen precipitates, it is possible to evaluate the oxide film breakdown voltage characteristics with even higher accuracy by extracting and evaluating only the oxygen precipitates.

[0031] (7th step) The seventh step is a step of calculating the oxide film breakdown voltage defect rate from the pseudo oxide film breakdown voltage degradation map created in the sixth step, and judging the silicon single crystal substrate as a reject if the oxide film breakdown voltage defect rate exceeds a predetermined predetermined value. [Example]

[0032] The present invention will be specifically explained below with reference to examples, but the present invention is not limited to these examples.

[0033] Example 1 First, two silicon single crystal substrates with a diameter of 300 mm, which were manufactured using the Czochralski method under conditions such that the entire surface of the silicon single crystal substrate was defect-free, were prepared as samples to be evaluated. For one of the two substrates, defects present on the surface of the silicon single crystal substrate were detected using a surface defect inspection device (SP5, Oblique mode, 19 nm Up). Next, the defects were observed using a scanning electron microscope and the defect types were classified. Next, the substrate was washed with hydrofluoric acid having a mass percent concentration of 0.5 wt % for 3 minutes. Next, the defects were observed at the same point using a scanning electron microscope, and the defect types were classified. Next, the defect images and the defect classification results before and after the hydrofluoric acid cleaning were compared, and defects that were classified as particles before the hydrofluoric acid cleaning and as pits after the hydrofluoric acid cleaning were reclassified as oxygen precipitates. Also, for defects that were classified as pits before the hydrofluoric acid cleaning and as pits after the hydrofluoric acid cleaning, defects whose inner diameter after the hydrofluoric acid cleaning was larger than before the hydrofluoric acid cleaning were reclassified as oxygen precipitates. Next, defects that degrade the oxide breakdown voltage (pits, oxygen precipitates, scratches) were extracted, and a pseudo oxide breakdown voltage degradation map (Fig. 2B) was created. If the GOI defect rate is less than 2%, there is no particular problem, so the pass / fail criterion was set at 2%. The GOI defect rate was calculated to be 0.1%, which is less than 2% and was therefore judged to be acceptable. For the other wafer, actual GOI measurements were performed, and the GOI failure rate was calculated from the resulting oxide film breakdown voltage degradation map (Fig. 2A), resulting in a result of 0.2%. From this, the evaluation results of the present invention and the actual GOI measurement results were almost the same, proving that the pass / fail judgment of silicon single crystal substrates based on the evaluation results of the present invention is correct.

[0034] Example 2 First, two silicon single crystal substrates with a diameter of 300 mm, which were manufactured using the Czochralski method under conditions such that the entire surface of the silicon single crystal substrate was defect-free, were prepared as samples to be evaluated. For one of the two substrates, defects present on the surface of the silicon single crystal substrate were detected using a surface defect inspection device (SP5, Oblique mode, 19 nm Up). Next, the defects were observed using a scanning electron microscope and the defect types were classified. Next, the substrate was washed with hydrofluoric acid having a mass percent concentration of 0.5 wt % for 3 minutes. Next, the defects were observed at the same point using a scanning electron microscope, and the defect types were classified. Next, the defect images and defect type classification results before and after hydrofluoric acid cleaning were compared. FIG. 4 is a table showing defect examples 1 to 3. Defect example 1 is a defect that was classified as a particle before the hydrofluoric acid cleaning and classified as a pit after the hydrofluoric acid cleaning, and was reclassified as an oxygen precipitate. Defect example 2 is a defect that was classified as a pit before the hydrofluoric acid cleaning and was also classified as a pit after the hydrofluoric acid cleaning, and the inner diameter of the pit after the hydrofluoric acid cleaning was larger than before the cleaning, and this defect was also reclassified as an oxygen precipitate. Defect example 3 is a defect that was classified as a particle before cleaning with hydrofluoric acid and was classified as a particle after cleaning with hydrofluoric acid, and was determined to be a particle. Next, oxide breakdown voltage degradation defects (pits, oxygen precipitates, scratches) were extracted, a pseudo oxide breakdown voltage degradation map (Fig. 3B) was created, and the GOI failure rate was calculated to be 3.3%. Since the failure rate was 2% or more, the sample was deemed to be unacceptable. For the other wafer, we actually performed GOI measurements and calculated the GOI failure rate from the resulting oxide film breakdown voltage degradation map (Figure 3A), which resulted in a result of 3.5%. From this, the evaluation results of the present invention and the actual GOI measurement results were almost the same, proving that the pass / fail judgment of silicon single crystal substrates based on the evaluation results of the present invention is correct.

[0035] (Comparative Example 1) Based on the results of the scanning electron microscope observation in Example 2, the defect types were classified into COP, processing pit, PID, and particle, and the number of defects of each defect type was counted. In addition, the weighted average value of the GOI defect rate described in Patent Document 1 was multiplied by the number of defects of each defect type to calculate the number of GOI defects. Finally, the number of GOI failure defects was added up and divided by the number of GOI measurement cells (3,000 points), resulting in an (estimated) GOI failure rate of 1.3%. Figure 5 is a table showing the experimental results of Comparative Example 1, including defect types, weighted average GOI failure rates, number of defects, number of GOI failure defects, and (estimated) GOI failure rate. FIG. 6 is a table showing the GOI defect rate calculated from the GOI measurement results using the present technology in Examples 1 and 2, and the GOI defect rate calculated using the conventional technology (Comparative Example 1). As can be seen from FIG. 6, the GOI defect rate calculated using the conventional technology was smaller than the GOI defect rate calculated from the GOI measurement results of Example 2.

[0036] As described above, according to the examples of the present invention, it is possible to simply and accurately evaluate the oxide film breakdown voltage characteristics, and to simply and accurately determine the pass / fail of a silicon single crystal substrate.

[0037] The present specification includes the following aspects. [1]: A method for evaluating oxide film breakdown characteristics of a silicon single crystal substrate, comprising: a first step of detecting defects in the silicon single crystal substrate using a surface defect inspection device and acquiring defect coordinates; a second step of observing the defects in the silicon single crystal substrate using a scanning electron microscope based on the defect coordinates, acquiring defect images, and classifying the defect types; a third step of cleaning the silicon single crystal substrate with hydrofluoric acid; a fourth step of observing the defects using a scanning electron microscope based on the defect coordinates acquired in the first step after the hydrofluoric acid cleaning, acquiring defect images, and classifying the defect types; a fifth step of comparing the defect images and defect types acquired in the second step and the fourth step to extract oxygen precipitates; and a sixth step of extracting defects that cause oxide film breakdown degradation from the first to fifth steps, and creating a pseudo oxide film breakdown degradation map based on the coordinates of the oxide film breakdown degradation defects. [2]: The method for evaluating the dielectric strength characteristics of an oxide film of a silicon single crystal substrate according to [1] above, characterized in that the defect types classified in the second and fourth steps are particles, pits, scratches, and PIDs. [3]: The method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to [1] or [2] above, characterized in that in the fifth step of extracting the oxygen precipitates, defects classified as particles in the second step and as pits in the fourth step are reclassified as oxygen precipitates, and among the defects classified as pits in the second step and as pits in the fourth step, defects whose pit depth in the fourth step is deeper than in the second step or whose inner diameter in the fourth step is larger than in the second step are reclassified as oxygen precipitates. [4]: A method for evaluating the dielectric strength characteristics of an oxide film of a silicon single crystal substrate according to any one of [1] to [3] above, characterized in that the defects causing degradation of the dielectric strength characteristics of the oxide film are pits, oxygen precipitates, or scratches. [5]: A method for determining the acceptability of a silicon single crystal substrate using the method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to any one of [1] to [4] above, characterized in that an oxide film breakdown voltage defect rate is calculated from the pseudo oxide film breakdown voltage degradation map created in the sixth step, and when the oxide film breakdown voltage defect rate exceeds a predetermined predetermined value, the silicon single crystal substrate is determined to be unacceptable.

[0038] 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 evaluating oxide film breakdown voltage characteristics of a silicon single crystal substrate, comprising: a first step of detecting defects in a silicon single crystal substrate using a surface defect inspection device and acquiring defect coordinates; a second step of observing the defects in the silicon single crystal substrate with a scanning electron microscope based on the defect coordinates, acquiring defect images, and classifying the defect types; a third step of cleaning the silicon single crystal substrate with hydrofluoric acid; a fourth step of observing the defects with a scanning electron microscope based on the defect coordinates obtained in the first step after the hydrofluoric acid cleaning, obtaining defect images, and classifying the defect types; a fifth step of comparing the defect images and defect types acquired in the second step and the fourth step to extract oxygen precipitates; a sixth step of extracting defects that cause degradation of the oxide film breakdown voltage from the first to fifth steps, and creating a pseudo oxide film breakdown voltage degradation map based on the coordinates of the defects that cause the degradation of the oxide film breakdown voltage.

2. 2. The method for evaluating the dielectric strength characteristics of an oxide film of a silicon single crystal substrate according to claim 1, wherein the defect types classified in the second and fourth steps are particles, pits, scratches, and PIDs.

3. In the fifth step of extracting the oxygen precipitates, Reclassifying the defects classified as particles in the second step and as pits in the fourth step into oxygen precipitates; 2. The method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to claim 1, wherein, among the defects classified as pits in the second step and as pits in the fourth step, defects whose pit depth in the fourth step is deeper than in the second step or whose inner diameter in the fourth step is larger than in the second step are reclassified as oxygen precipitates.

4. 2. The method for evaluating the dielectric strength characteristics of an oxide film of a silicon single crystal substrate according to claim 1, wherein the defects causing the deterioration of the dielectric strength of the oxide film are pits, oxygen precipitates, or scratches.

5. A method for determining whether a silicon single crystal substrate is acceptable or not, using the method for evaluating the oxide film breakdown voltage characteristics of a silicon single crystal substrate according to any one of claims 1 to 4, calculating a failure rate of oxide film breakdown voltage from the pseudo oxide film breakdown voltage degradation map created in the sixth step; A method for determining whether a silicon single crystal substrate is acceptable or not, comprising determining that the silicon single crystal substrate is unacceptable when a predetermined oxide film breakdown voltage defect rate is exceeded.

Images