Method for evaluating a silicon single-crystal ingot

Abstract

The invention provides a method for evaluating a silicon single-crystal ingot that reduces the time required to assess the change in resistivity along the crystal length direction and improves wafer yield. In the method for evaluating a silicon single-crystal ingot according to the present embodiment, counter-doping is performed by adding a subdoping agent during the pulling of the silicon single-crystal ingot using the CZ method. The resistivity at a side face along a crystal length direction is measured by a four-point probe method in a state of the silicon single-crystal ingot 1 after the silicon single-crystal ingot has been pulled up. A reject area is determined based on the distribution of the resistivity along the crystal length direction.

Description

Technical area

[0001] The present invention relates to a method for evaluating silicon single crystal ingots produced by the Czochralski process (CZ process). Background technology

[0002] The Czochralski process produces silicon single-crystal ingots (hereinafter referred to as "single-crystal ingots") by contacting a seed crystal with a silicon melt and slowly pulling the seed crystal while rotating it. The Czochralski process is typically used to produce large-diameter single-crystal ingots.

[0003] It is now known that the resistivity of single-crystal ingots produced by the CZ process varies along the crystal growth direction (crystal length direction). In recent years, due to the quality requirements for semiconductor wafers (hereinafter referred to as wafers) produced from single-crystal ingots, it has become important to maintain the resistivity within a desired range.

[0004] For example, when measuring the resistivity of a wafer, a single-crystal ingot grown by the CZ method is usually ground around the circumference to a predetermined size (diameter), the part of the head and tail cones that cannot be used as a product is cut off, and the resulting single-crystal ingot is cut at a predetermined position and processed into an ingot with a length that fits into a cutting device such as an inner hole saw or wire saw. At this point, sample wafers are simultaneously cut out for a test of the resistivity and other parameters. The cut sample wafers can then be used to measure the resistivity. Each ingot is then cut into wafers of a predetermined thickness. By removing the cut wafers and measuring the resistivity, it is then possible to measure the resistivity in the crystal length direction.

[0005] When growing single-crystal ingots using the CZ method, a phenomenon occurs in which the resistivity changes in the direction of crystal growth when dopants are added. This is due to the segregation of dopants. As the silicon melt in the crucible decreases with single-crystal growth, the concentration of dopants in the residual liquid gradually increases, and the resistivity of the single crystal continuously decreases. The segregation coefficient of phosphorus P is 0.35, which is lower than that of boron B (0.8), which is commonly used as a dopant in p-type crystals. The decrease in resistivity from top to bottom is more pronounced than that of p-type crystals. This creates the problem that the proportion usable as the product becomes small, and it is difficult to improve the yield.

[0006] As a countermeasure to the above-described problem that the resistivity in the crystal length direction deviates from the desired resistivity range, a method for measuring the resistivity in the state of a single crystal ingot is disclosed in the following Patent Literature 1 (PTL 1).

[0007] Specifically, the resistivity along the crystal length direction is measured on the side surface of the single-crystal ingot, the position where the desired resistivity is obtained is specified, a block of a predetermined length is cut, and wafers are cut from the block. The method for measuring resistivity along the crystal length direction uses the four-point probe method for measuring resistivity. This makes it possible to produce wafers with a resistivity of 1 Ω cm or less.

[0008] It is known that it is difficult to achieve a uniform resistivity distribution along the crystal length direction in single-crystal ingots grown by the CZ method due to segregation. For this reason, single-crystal ingots have traditionally been grown by the so-called counter-doping method, in which a main dopant and a sub-dopant of opposite polarity are added to the main dopant (see Patent Literature 2 below).

[0009] According to Patent Literature 2 below, counter-doping consists of the process of doping the main dopant (e.g., phosphorus) that causes an n-type conductivity and continuously or intermittently doping an additional sub-dopant (e.g., boron) that causes a p-type conductivity opposite to the n-type conductivity while growing the single crystal ingot, according to the solidification rate expressed as (crystallized weight) / (initial weight of the silicon raw material). In some cases, a single crystal grown by counter-doping is hereinafter referred to as a counter-doped crystal. Citation listPatent literature PTL 1: JP-A-2012-129308 PTL 2: JP-A-2016-060667 Brief description of the inventionTechnical problem

[0010] According to PTL 1, wafer yield can be improved because an ingot can be cut to include more of the desired resistivity portion when producing wafers with a low resistivity of less than 1 Ω cm. Furthermore, since the resistivity is measured in a single-crystal ingot state, the time required to evaluate the resistivity can be significantly reduced compared to the method in which the resistivity is measured after cutting a sample wafer.

[0011] The manufacturing process described in PTL 1 is problematic in the following respects.

[0012] For example, for heavily doped wafers (low resistivity of 1 Ω cm or less), since the dopant concentration of the main dopant is sufficiently high compared to the thermal donor, and the effects of the thermal donors are less even without heat treatment, the resistivity can be measured in the single-crystal ingot state. In contrast, for a lightly doped crystal with a desired resistivity of, for example, 10 Ω cm or more, since it is affected by the thermal donor, the donor-killing process of heat treatment is required to achieve the desired resistivity, but the donor-killing process is difficult to perform in the single-crystal ingot state.That is, in the case of a lightly doped crystal, it is difficult to evaluate the true resistivity, hereinafter referred to as true resistivity, in the crystal length direction, from which the effect of the thermal donor is excluded, even if the resistivity is measured in the single-crystal ingot state.

[0013] A counterdoped crystal has a region where the resistivity increases rapidly and then decreases rapidly near a position along the crystal length where a subdopant is added (hereinafter referred to as the high resistivity region). Therefore, the high resistivity region described above should be determined in the counterdoped crystal.

[0014] The present invention has been developed in consideration of the above-mentioned problem, and an object of the invention is to provide an evaluation method that enables the determination of the high resistivity region in the crystal length direction in the single crystal ingot state in a lightly doped, counter-doped crystal. Solution to the problem

[0015] The method according to the present invention is for evaluating a single-crystal ingot that has been counter-doped by adding a sub-dopant during silicon single-crystal pulling by the CZ method. The method includes a step of measuring the resistivity in a crystal length direction using a four-point probe method on a side surface of a single-crystal ingot in a single-crystal ingot state, and a step of determining a reject region of the single-crystal ingot (non-product region) based on a resistivity distribution in the crystal length direction.

[0016] In the single-crystal ingot, there is a region where a predetermined threshold value or less of the change in resistivity is stable (e.g., 1 to 3%), hereinafter referred to as the stable region. In the counter-doped crystal according to the present invention, the reject region is determined with reference to a position, hereinafter also referred to as the reference position, at which there is a peak resistivity value that is more than 10% higher than the resistivity of the stable region on the head side. That is, the reject region is defined as a region between positions separated by predetermined lengths from the peak position (the reference position is defined as the position where the resistivity exceeds 10% for the first time when there are multiple data exceeding 10%) to the head side and the tail side, e.g.,8 to 12 mm toward the tip side and 10 to 20 mm toward the tail side. The reject area on the tail side may extend to the position where the change in resistivity in the crystal length direction falls below a threshold value (e.g., 1% or less) and begins to become subject to segregation (hereinafter referred to as the stable position).

[0017] With the above configuration, a high resistivity region due to counter-doping can be specified from the resistivity distribution on the side surface of the single-crystal ingot, the reject area in the single-crystal ingot state is determined, and the wafer yield can be significantly increased compared to a method (conventional method) in which resistivity is evaluated using test sample wafers cut from the single-crystal ingot. Furthermore, the time required for resistivity evaluation can be shortened compared to the conventional method.

[0018] In the evaluation method of the single crystal ingot according to the present invention, the measurement range of the resistivity preferably includes at least a range from the position where the sub-dopant is added to the position where the resistivity becomes the stable desired resistivity again after the rise and rapid fall of the resistivity.

[0019] Furthermore, in the evaluation method of the single crystal ingot according to the present invention, the specific resistance of the single crystal ingot should be 10 Ω·cm or more. Advantageous effects of the invention

[0020] With the present invention, the time required for evaluating the resistivity in the crystal length direction can be shortened, and also the yield of wafers can be improved. Short description of the drawings Fig.1 is a view showing an overview of the measurement of the resistivity on the side surface of a single crystal ingot; Fig. 2 is a view showing the single crystal ingot of an example; Fig. 3A and Fig. 3B shows views illustrating the measurement results of the resistivity on the side surface of a single crystal ingot; Fig. 4A and Fig. 4B show views for illustrating a reject area; Fig. 5A and Fig. 5B are views showing the measurement results of the resistivity at the center and outer periphery of a wafer; and Fig. 6A to 6D are views showing the measurement results of the resistivity in the radial direction of the wafer. Description of embodiments

[0021] An embodiment of the method for evaluating the resistivity of a single-crystal ingot is described below. The present invention is not limited to the above-described embodiments. Although the embodiment is described using an n-type single-crystal ingot, it is not limited thereto and can also be applied to a p-type single-crystal ingot.

[0022] In the present embodiment, when growing lightly doped (10 Ω cm or more) n-type single-crystal ingots by the CZ method, counter-doping is performed by adding a p-type dopant as a sub-dopant during pulling. For example, due to segregation, it is difficult to achieve a uniform distribution of resistivity in the crystal growth direction (crystal length direction) for single-crystal ingots grown by the CZ method. However, this problem can be solved by counter-doping with a sub-dopant of the opposite polarity to the main dopant.

[0023] For example, when silicon crystallizes, the concentration of the dopant introduced into the crystal is lower than the concentration of the dopant in the melt. As the single-crystal ingot continuously grows, more dopant remains in the melt, and the dopant concentration in the melt gradually increases. As a result, the dopant concentration in the crystal gradually increases, and the resistivity decreases. To prevent the resistivity from falling outside the desired resistivity range due to the decrease in resistivity, the sub-dopant of opposite polarity is added to the melt in the appropriate amount (counter-doping). The resistivity increases at the position of the crystal length where the sub-dopant is added, and then the resistivity decreases and stabilizes depending on the segregation.

[0024] In the present embodiment, for example, phosphorus (P), arsenic (As), and antimony (Sb) can be used as the n-type dopant (for the main dopant), and boron (B), aluminum (Al), and gallium (Ga), for example, can be used as the p-type dopant (for the sub-dopant). They can be used when the main dopant is p-type and the sub-dopant is n-type.

[0025] In the present embodiment, after pulling a counter-doped crystal, it is subjected to circumferential grinding to the predetermined dimension (diameter), and the head and tail cone portions are cut off. Then, the resistivity in the crystal length direction is measured by the four-point probe method on the side surface in the single-crystal ingot state, and the relative change in the resistivity in the crystal length direction is evaluated based on the resistivity distribution. Specifically, based on the resistivity distribution in the crystal length direction (relative resistivity distribution), a range is detected from a position that is a predetermined length (e.g.,8 to 12 mm) back toward the head side from the peak position (reference position) to the stable position, where the peak position is a position where the resistivity change has a resistivity 10% or more higher than the resistivity of the stable region from the head side. Alternatively, a reject region is determined by the length between the position a predetermined length (e.g., 8 to 12 mm) back from the reference position toward the head side and the position a predetermined length (e.g., 10 to 20 mm) forward from the reference position toward the end side. It should be noted that the single-crystal ingot is not necessarily ground circumferentially.

[0026] Fig.Figure 1 shows an overview of the measurement of the resistivity on the side surface of a single-crystal ingot. After the counter-doped crystal is pulled up, a single-crystal ingot 1 is formed as shown in Fig. 1, obtained by circumferential grinding of the counter-doped crystal and cutting off its head and tail cones. In addition, since the positions in the crystal length direction where the sub-dopant is added (sub-dopant addition positions) are recorded in advance when the crystal is pulled up, the resistivity at the side surface of the single-crystal ingot 1, including the sub-dopant addition positions, is measured by the four-point probe method. Specifically, as shown in Fig.As shown in Figure 1, the resistivity on the side surface of the single crystal ingot 1 was measured at thirty (30) points along the crystal growth direction at an interval of 1 mm as a starting point, which is the position 10 mm from the position where the sub-dopant is added, backward toward the head side. At this time, the resistivity near the position where the sub-dopant was added sharply increases and then drops sharply, and the resistivity gradually decreases depending on segregation. The measurement range of the resistivity should only cover a range from the position where the sub-dopant is added to the position where the resistivity stabilizes again to a desired resistivity after an increase, and may not necessarily be limited to the thirty points at intervals of 1 mm.

[0027] In the present embodiment, the resistivity is measured from the stable region on the head side in the crystal length direction, and the reference position is defined by a peak position where the peak value of the resistivity becomes not less than 10% higher than that of the stable region. In addition, the stable position is referred to as the position where the change rate of the resistivity on the end side in the crystal length direction stabilizes and the change rate of the resistivity falls below the threshold value (e.g., 1% to 3%) and the resistivity starts to be subject to segregation, and the reject region may be a range from a position 10 mm backward toward the head side from the reference position to the stable position.This can significantly reduce the processing loss in the wafer processing and the time required for evaluating the change in resistivity, compared to the method of cutting test sample wafers from a single-crystal ingot and measuring their resistivity (conventional method). Note that the reject area on the end side from the peak position is not limited to the area up to the stable position, but can have a predetermined length. For example, the reject area on the end side can be 10 mm or more and no more than 20 mm from the reference position. The reject area on the head side from the reference position can be 8 mm or more and no more than 12 mm from the peak position (reference position).

[0028] In a counterdoped n-type crystal grown by the CZ method as described above, a certain amount of oxygen is dissolved by the above manufacturing process; the dissolved oxygen partially becomes a thermal donor, which acts as an n-type dopant and reduces the resistivity. By subjecting the counterdoped crystal grown by the CZ method to a heat treatment (a donor killing process), the original resistivity of the crystal, which is determined by the main dopant, can be retained. Thermal donors are generated in large numbers when the thermal history is long at around 450°C and the oxygen concentration is high during crystal growth.

[0029] Furthermore, the resistivity peaks formed by the resistivity change on the side surface of the single-crystal ingot 1 vary depending on the dopant concentration, the number of thermal donors, and the degree of thermal donor change at the measurement position. In the present embodiment, when the resistivity is 10 Ω cm or more regardless of the oxygen concentration, peaks with a resistivity change of 10% or more are formed near the position in the crystal length direction where the sub-dopant is added.

[0030] The change in resistivity at the side surface of the single crystal ingot 1 due to the thermal donors can be represented by the following equation: Change in resistivity due to thermal donors = (resistivity corresponding to the amount of change in thermal donor) / (resistivity of single crystal ingot + resistivity due to the amount of thermal donors).

[0031] For example, when the oxygen concentration in single-crystal ingot 1 is high, since the resistivity on the side surface is mainly determined by the number of thermal donors, and the denominator becomes small due to the effect of the thermal donors, the ratio (the change in resistivity) becomes large. On the other hand, when the oxygen concentration is low, the numerator becomes large because the amount of change in the thermal donors becomes small (the resistivity increases due to the thermal donors). This forms a peak where the resistivity increases by 10% or more from that of the stable region due to the changes in the denominator or numerator due to the change in the oxygen concentration, and the reference position (the position where the change in resistivity due to counter-doping is large) can be detected.

[0032] It should be noted that the specific resistance is measured by the four-point probe method in the present embodiment, but the method is not limited to this, and any measurement method can be used as long as the specific resistance is measured in the crystal length direction on the side surface in the single-crystal ingot state. Effects

[0033] As described above, the time required for evaluating the resistivity in the evaluation process of a single crystal ingot can be reduced in the present embodiment. Examples

[0034] Next, examples of the evaluation method of a silicon single-crystal ingot according to the present invention will be described. Note that the present invention is not limited to the examples described below. Example 1

[0035] A counterdoped, lightly doped n-type crystal was grown by counterdoping with the main dopant phosphorus and the subdopant boron. The subdopant boron was added at positions of 500 mm and 800 mm along the crystal length during the growth of a counterdoped crystal under the conditions of a crystal rotation speed of 10 rpm, a crucible rotation speed of 1 rpm, a pulling speed of 1 mm / min, and a magnetic field strength of 2000 G. In Example 1, the desired resistivity was 50 Ω cm (phosphorus concentration: approximately 8.64 10 13 / cm 3 ) and the desired oxygen concentration 0.55·10 18 / cm 3 .

[0036] Using the above-mentioned settings, a counter-doped crystal was pulled up and finished by circumferential grinding to a diameter of 300 mm, the head and end cones were cut off, and a single-crystal ingot was prepared as in Fig. 2 is shown.

[0037] Subsequently, the resistivity in the crystal length direction at the side surface in the single-crystal ingot state was measured using the four-point probe method. Specifically, the resistivity of the side surface of the single-crystal ingot was measured at 30 points spaced 1 mm apart along the crystal growth length, starting from the position 10 mm behind the position where the sub-dopant is added, i.e., at the positions 490 mm and 790 mm from the head side (see Figure 1). Fig. 2).

[0038] The Fig. 3A and Fig.3B shows views illustrating the measurement results of the resistivity (relative values) on the side surface of a single crystal ingot; Fig. Figure 3A shows a change in the relative resistivity in the crystal length direction from a position 490 mm away from the head side as a starting point, and Fig. Figure 3B shows a change in the relative resistivity in the crystal length direction from a position 790 mm away from the head side as a starting point. The measurement of the resistivity on the side surface of the single-crystal ingot by the four-point probe method at 30 points spaced 1 mm apart shows that the relative resistivity increases sharply near the position where the sub-dopant was added; the maximum peak height is 42% at the Fig. 3A or 21% of the position shown in Fig.3B. After that, the relative resistivity decreased abruptly and stabilized at the specified resistivity.

[0039] Near the position of 500 mm in the crystal length direction, the peak position with 42% of the change in resistivity in the crystal length direction is defined as the reference position, and a position where the change in resistivity stabilizes and the change in relative resistivity in the crystal length direction becomes 1% or less and begins to depend on segregation is defined as a stable position. Then, as in Fig.As shown in Figure 4A, the reject area (16 mm) is defined as a range from a position 10 mm from the reference position back to the head side to the stable position. Furthermore, near the position of 800 mm in the crystal length direction, the peak position with a change of 21% in the specific resistance in the crystal length direction is defined as a reference position, and a position where the change in the specific resistance stabilizes and the change in the relative specific resistance in the crystal length direction is 1% or less and begins to depend on segregation is defined as a stable position. As shown in Fig. As shown in Figure 4B, the reject area (15 mm) is defined as a range from a position 10 mm from the reference position back to the head side to the stable position. Verification 1

[0040] Next, as described above, after measuring the resistivity on the side surface of the single-crystal ingot, wafers in the above-described reject area are obtained by processing the ingot in the reject area at a distance of 1 mm into wafers. Then, the effects on the evaluation of the change in resistivity on the side surface of the single-crystal ingot are verified by calculating the oxygen concentration and the number of thermal donors of the respective wafers. The oxygen concentration at the resistivity measurement positions was 0.55 10 18 / cm 3 , the number of thermal donors obtained from the resistivity before and after the donor killing process was 3.9·10 13 / cm 3 , and the change of thermal donors in the length direction of the single crystal was 3.18·10 12 / cm 3. From the results described above, it is clear that the number of thermal donors is the quantity that allows the evaluation of the peak value of the resistivity on the side surface of the single-crystal ingot, as shown in the Fig. 2A and Fig. 2B is shown. Verification 2

[0041] From the total processed wafers described in Verification 1, wafers from the position 490 mm to the position 519 mm in the crystal length range and wafers from the position 790 mm to the position 819 mm in the crystal length range were sampled, and the resistivity at the center and periphery of each sampled wafer was measured by the four-point probe method.

[0042] The Fig. 5A and Fig. 5B are views showing the measurement results of the resistivity (relative values) at the center and periphery of the wafers; Fig.Figure 5A shows the change in the relative resistivity of thirty (30) wafers taken from the range from the position 490 mm to the position 519 mm in the crystal length, and Fig. Figure 5B shows the change in the relative resistivity of thirty (30) wafers taken from the area from the position 790 mm to the position 819 mm in the crystal length. As shown in the Fig. 5A and Fig. As shown in Figure 5B, in both cases, an increase in the relative resistivity (increase in the specific resistance) occurs in the center of the wafers at the positions where the subdopant was added. Furthermore, in both cases, the Fig. 5A and Fig.5B shows an increase in the relative resistivity (increase in specific resistance) at the outer periphery of the wafer, a few millimeters from the peak position to the end side in the center of the wafer. From these results, it was confirmed that the changes in specific resistance at the side of the single-crystal ingot and the changes in specific resistance at the outer periphery of the wafer are generally consistent. In addition, as shown in the Fig. 5A and Fig. 5B, the peak positions of the outer periphery of the wafer shifted by 4 to 6 mm from the peak positions of the wafer center toward the end side; consequently, this verification confirms the need to discard wafers from the "position 10 mm from the reference position toward the head side" determined in Example 1. Verification 3

[0043] Next, the resistivity of each processed wafer was measured in the radial direction.

[0044] The Fig. For example, Figures 6A to 6D show the measurement results of the specific resistance (relative value) of wafers in the radial direction at the points shown in Fig. 2 positions marked a to d. Fig. Figure 6A shows the change of the relative resistivity in the radial direction at position a, Fig. Figure 6B shows the change of the relative resistivity in the radial direction at position b, Fig. Figure 6C shows the change of the relative resistivity in the radial direction at position c, and Fig. Figure 6D shows the change in the relative resistivity in the radial direction at position d. In Verification 3, the resistivity for each wafer was measured at an interval of 5 mm in the radial direction. As a result, for example, the relative resistivity for the wafer at position a, i.e., before the addition of the subdopant to the crystal (see Fig. 6A), is generally constant, and the radial gradient of resistivity (RRG) is generally constant at 5% or less. Note that RRG is a value expressed as a percentage and is determined by dividing the difference between the maximum and minimum values ​​in a group of resistivity measurements taken at any point within a single silicon crystal substrate by the minimum value.

[0045] For example, the relative resistivity increases in the center of the wafer cut at position b where the subdopant is added (see Fig. 6B), and the RRG deteriorates (the RRG exceeds 5%). For example, for the wafer cut at position c, i.e., after the addition of the sub-dopant, the relative resistivity at the outer periphery increases (see Fig.6C), and the RRG remains in a state greater than 5%. For example, the relative resistivity in the radial direction does not increase for the wafer cut at position d, i.e., after the sub-dopant is added to the far end side (see Fig. 6D), and the relative resistivity stabilizes and the RRG improves to 5% or less.

[0046] By measuring the resistivity in the radial direction of the processed wafer described above, the position in the crystal length direction where the RRG stabilizes again within 5% after the RRG has deteriorated beyond 5% can be determined. This means that the need to reject the section up to the "stable position" determined in Example 1 is confirmed by verification. Furthermore, the verification confirms that the RRG is within 5% for all wafers cut outside the reject area. Example 2

[0047] A counterdoped crystal of the lightly doped n-type was grown under the same conditions as in Example 1, except that the desired oxygen concentration was 1.20 10 18 / cm 3 fraud.

[0048] Then a single crystal ingot was prepared as in Fig.2, by circumferential grinding to 300 mm in diameter and cutting off the head and end cone sections. Then, the resistivity on the side surface in the crystal length direction was measured using the four-point probe method in the single-crystal ingot state, similar to Example 1 (see Fig. Fig. 2).

[0049] As a result, for the single-crystal ingot of Example 2, the relative resistivity near the positions where a subdopant was added increased sharply at 500 mm and 800 mm crystal length, similar to Example 1, and then the relative resistivity decreased sharply and stabilized at the desired resistivity.

[0050] In addition, in Example 2, the reject area is defined as a range from a position 10 mm from the reference position (the position where the increase in relative resistivity in the crystal length direction increases to 10% or more) back to the head side to the stable position (the position where the change rate of relative resistivity in the crystal length direction becomes 1% or less and starts to depend on segregation).

[0051] Since the single-crystal ingot in Example 2 has a desired oxygen concentration of 1.20·10 18 / cm 3 and a change in the thermal donor of 7.21·10 12 / cm 3 , the reference position (the crystal length position where the subdopant was added) was detected by the resistivity change of 12% or more. Comparison example

[0052] In the comparative example, a counterdoped lightly doped n-type crystal was grown under the same conditions as in Example 1, and a single crystal ingot similar to that of Example 1 was prepared.

[0053] In the comparative example, the manufactured single-crystal ingot was cut into blocks, and the evaluation of resistivity (an evaluation of whether the RRG is within 5%) was performed on test sample wafers cut at the time of manufacturing the blocks (conventional method).

[0054] However, since this method is unclear at which mm position from the end of the ingot the resistivity increases (where the RRG exceeds 5%), the positions where the test sample wafers were cut were 490 mm and 510 mm in the crystal length direction for the 500 mm position where the sub-dopant was added. After processing the respective test sample wafers and evaluating the resistivity, the RRG was found to be 3.4% for the 490 mm case and 7.0% for the 510 mm case, based on the crystal length direction. Table 1 shows the RRG results of the cut samples. Table 1 Cutting position of a hole saw RRG Evaluation 490 mm 3,4% Good 510 mm 7,0% Not good

[0055] The resistivity evaluation was therefore performed using test sample wafers cut to a constant thickness by a wire saw. While the resistivity evaluation for a test sample wafer cut at a position of 511 mm in the crystal length direction and subjected to processing resulted in an RRG of 5.2%, the resistivity evaluation for a test sample wafer cut at a position of 512 mm in the crystal length direction and subjected to processing again resulted in an RRG of 3.9%. The kerf loss in the crystal length due to cutting the ingot and processing was 22 mm. Table 2 Cutting position of a hole saw RRG Evaluation 511 mm 5,2% Not good 512 mm 3,9% Good

[0056] In Examples 1 and 2, as described above, the reject region can be immediately determined by specifying the high-resistivity region from the resistivity distribution on the side surface of the single-crystal ingot. After rejection, wafer processing can be performed using the silicon ingot with an RRG of 5% or less. This significantly improves the wafer yield (possibly by about 30% on average) compared to the (conventional) method of cutting test pattern wafers from the single-crystal ingot and performing resistivity evaluation. Furthermore, the time required for resistivity evaluation could be significantly reduced compared to the conventional method (possibly about 12 hours on average).

[0057] Note that, as another comparative example, there is a method in which a counterdoped, lightly doped n-type crystal is pulled up under conditions similar to those in Example 1, and an ingot is cut at a position where a sub-dopant was added in the direction of the crystal length based on log data. However, a problem arose in that an ingot could not be cut at the position where the sub-dopant was added because the position where the sub-dopant was added was ambiguous due to the circumferential grinding of the single-crystal ingot and the sawing process of the wafers.

[0058] The present invention is not limited to the above embodiments and examples. The above embodiments and examples are illustrative examples, and any product having substantially the same configuration and similar effects as the technical concept described in the claims of the present invention is included within the technical scope of the present invention. Industrial applicability

[0059] As described above, the method for evaluating silicon single crystal ingots is useful for silicon single crystal ingots manufactured by the Czochralski method (CZ method), and is particularly suitable for silicon single crystal ingots in order to keep the resistivity of the wafers within a desired range. List of reference symbols 1 single-crystal ingot (silicon single-crystal ingot) QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] JP-A-2012-129308

[0009] JP-A-2016-060667

[0009]

Claims

[1] A method for evaluating a silicon single crystal ingot that has been counter-doped by adding a sub-dopant during pulling of the silicon single crystal ingot according to the CZ method, the method comprising: a step of measuring the resistivity in a crystal length direction by a four-point probe method on a side surface in a silicon single crystal ingot state; and a step of determining a reject area based on a distribution of resistivity in the crystal length direction. [2] A method for evaluating a silicon single crystal ingot according to claim 1, wherein a reference position is defined as a position where the peak value of the resistivity is higher than the resistivity of a stable region of a head side where the change in the resistivity is stable by more than 10%, and the reject region is defined as a region between positions separated from the reference position toward the head side and the end side by predetermined lengths. [3] The method for evaluating a silicon single crystal ingot according to claim 1, wherein a reference position is defined as a position where the peak value of the resistivity is higher than the resistivity of a stable region of a head side where the change in the resistivity is stable by more than 10%, and further wherein a stable position is defined as a position where the change in the resistivity on an end side from the reference position is stable and the change in the resistivity is not larger than a threshold value and starts to be dependent on segregation, and wherein the reject region is defined as a region from a position of a predetermined length from the reference position back to the head side to the stable position. [4] A method for evaluating a silicon single crystal ingot according to claim 1, wherein a measurement range of the resistivity includes at least a range from the position where sub-dopant is added to the position where the resistivity increases and then stabilizes again to the desired resistivity. [5] A method for evaluating a silicon single crystal ingot according to any one of claims 1 to 4, wherein the specific resistance of a silicon single crystal ingot to be grown is 10 Ω·cm or more. [6] The method for evaluating a silicon single crystal ingot according to claim 2 or 3, wherein the reject range is defined as a range from the reference position 8 mm or more to 12 mm or less toward the head side and 10 mm or more to 20 mm or less toward the end side.

Images