阅读说明：本技术 硅层的评价方法和硅外延晶片的制造方法 (Method for evaluating silicon layer and method for manufacturing silicon epitaxial wafer ) 是由 牧濑佐也加 佐俣秀一 三次伯知 宫崎澄夫 于 2019-01-18 设计创作，主要内容包括：本发明提供一种硅层的评价方法,其包括：在硅层的表面上形成氧化膜；进行使上述形成的氧化膜的表面带有负电荷的带电处理；以及通过范德堡(van der Pauw)法测定上述带电处理后的硅层的电阻率。(The invention provides an evaluation method of a silicon layer, which comprises the following steps: forming an oxide film on a surface of the silicon layer; performing a charging process for negatively charging the surface of the oxide film; and measuring the resistivity of the charged silicon layer by van der Pauw method.)

1. A method for evaluating a silicon layer, comprising:

forming an oxide film on a surface of the silicon layer;

performing a charging process for negatively charging the surface of the oxide film; and

the resistivity of the charged silicon layer was measured by the van der Pauw method.

2. The method for evaluating a silicon layer according to claim 1, wherein the charging treatment is performed by corona discharge treatment.

3. The method of evaluating a silicon layer according to claim 1, wherein the charging treatment is performed by a voltage application treatment.

4. The method for evaluating a silicon layer according to any one of claims 1 to 3, wherein the silicon layer is a silicon epitaxial layer included in an evaluation sample having a pn structure.

5. The method for evaluating a silicon layer according to any one of claims 1 to 4, wherein the silicon layer is a high-resistance silicon epitaxial layer having a resistivity measurement value of 200 Ω cm or more by an extended resistance method.

6. The method for evaluating a silicon layer according to any one of claims 1 to 3, wherein the silicon layer is an active layer included in an SOI wafer.

7. A method for manufacturing a silicon epitaxial wafer, comprising:

manufacturing a candidate silicon epitaxial wafer to be shipped as a product;

further comprising:

evaluating a silicon epitaxial layer for evaluation formed in the same or approximately the same step of epitaxial growth conditions as the silicon epitaxial layer included in the silicon epitaxial wafer that is a candidate for shipment from the factory as described above by the evaluation method; and

if the resistivity obtained for the silicon epitaxial layer for evaluation is within a range that is acceptable for good products, the silicon epitaxial wafer that is a candidate for shipment as a product is subjected to preparation for shipment as a product.

8. A method for manufacturing a silicon epitaxial wafer, comprising:

forming a silicon epitaxial layer for evaluation under a test epitaxial growth condition;

evaluating the silicon epitaxial layer for evaluation by the evaluation method according to any one of claims 1 to 5;

determining the epitaxial growth conditions for which the test epitaxial growth conditions are changed as the epitaxial growth conditions used in the product manufacturing step or determining the test epitaxial growth conditions as the epitaxial growth conditions used in the product manufacturing step, based on the resistivity determined for the silicon epitaxial layer for evaluation; and

and manufacturing a product-factory silicon epitaxial wafer through a product manufacturing step including an epitaxial growth step performed under the determined epitaxial growth conditions.

Technical Field

The present invention relates to a method for evaluating a silicon layer and a method for manufacturing a silicon epitaxial wafer.

Background

As a method for measuring resistivity with respect to a semiconductor wafer, various methods are known (for example, the spreading resistance method described in japanese patent application laid-open No. 2005-223098 (the entire contents of which are hereby specifically incorporated by reference in a published manner)).

Disclosure of Invention

A silicon epitaxial wafer, which is one type of semiconductor wafer, has a silicon epitaxial layer on a single-crystal silicon substrate. For silicon epitaxial wafers, the resistivity of the silicon epitaxial layer affects the performance of the wafer. Therefore, for example, a silicon epitaxial layer is formed on a monitor wafer, and the evaluation result obtained by measuring the resistivity of the silicon epitaxial layer may be used for quality assurance of a product wafer and step management of an epitaxial growth step. In order to ensure such quality and process control with high reliability, it is desirable that the measurement accuracy be high. With respect to the measurement accuracy, the smaller the deviation of the measured value (measurement deviation), the higher the accuracy can be said. As for the measurement variation, among various methods known as a method for measuring resistivity with respect to a silicon epitaxial wafer, it can be said that the measurement variation of the spreading resistance method is small. However, in the spreading resistance method, the present inventors presume that the higher the resistivity of the measurement object, the larger the variation in contact resistance between the probe and the sample, and the larger the measurement variation. Therefore, the present inventors have considered that a new evaluation method with less measurement variation in resistivity measurement of silicon epitaxial layers of various resistivities should be provided. The same can be said for the measurement of resistivity of various silicon layers in the above-described aspect.

One embodiment of the present invention provides a novel evaluation method capable of reducing measurement variations in the measurement of the resistivity of a silicon layer.

One embodiment of the present invention relates to a method for evaluating a silicon layer, including:

forming an oxide film on a surface of the silicon layer;

performing a charging process for negatively charging the surface of the oxide film; and

the resistivity of the silicon layer after the above-mentioned electrification treatment was measured by the van der Pauw method.

In one embodiment, the charging treatment may be performed by corona discharge treatment.

In one embodiment, the charging process may be performed by a voltage application process.

In one embodiment, the silicon layer may be a silicon epitaxial layer included in an evaluation sample having a pn structure.

In one embodiment, the silicon layer may be a high-resistance silicon epitaxial layer having a resistivity measurement value of 200 Ω cm or more by an extension resistance method.

In one embodiment, the Silicon layer may be an active layer included in an SOI (Silicon-On-Insulator) wafer.

Another embodiment of the present invention relates to a method for manufacturing a silicon epitaxial wafer, including:

manufacturing a candidate silicon epitaxial wafer to be shipped as a product;

further comprising:

evaluating, by the evaluation method, a silicon epitaxial layer for evaluation formed in the same or approximately the same epitaxial growth step as a silicon epitaxial layer included in the silicon epitaxial wafer that is a candidate for shipment from the factory as a product; and

if the resistivity obtained for the silicon epitaxial layer for evaluation is within a range that is acceptable for good products, the silicon epitaxial wafer that is a candidate for shipment as a product is subjected to preparation for shipment as a product.

Another embodiment of the present invention relates to a method for manufacturing a silicon epitaxial wafer, including:

forming a silicon epitaxial layer for evaluation under a test epitaxial growth condition;

evaluating the silicon epitaxial layer for evaluation by the evaluation method;

determining an epitaxial growth condition, which is a condition for changing the test epitaxial growth condition, as an epitaxial growth condition used in a product manufacturing step or determining the test epitaxial growth condition as an epitaxial growth condition used in the product manufacturing step, based on the resistivity obtained for the evaluation silicon epitaxial layer; and

the silicon epitaxial wafer for product shipment is manufactured through a product manufacturing step including the epitaxial growth step performed under the above-identified epitaxial growth conditions.

According to one embodiment of the present invention, a method for evaluating a silicon layer can be provided, which can reduce measurement variations in the measurement of the resistivity of the silicon layer.

Detailed Description

[ evaluation method of silicon layer ]

One embodiment of the present invention relates to a method for evaluating a silicon layer (hereinafter also simply referred to as "evaluation method") including: forming an oxide film on a surface of the silicon layer; performing a charging process for negatively charging the surface of the oxide film; and measuring the resistivity of the charged silicon layer by van der Pauw method.

The van der Pauw (van der Pauw) method is a method known as a method for evaluating physical properties such as resistivity, hall coefficient, and the like of a semiconductor sample. The van der Pauw (van der Pauw) method is a method in which electrodes are provided at 4 sites on the surface of a sample, measurement such as voltage measurement is performed at a plurality of sites, and the values obtained in this manner are averaged to obtain a measurement result. Therefore, the present inventors have considered that the measurement accuracy is expected to be improved as compared with a method of performing measurement only at one site (for example, a four-probe method known as a method of measuring resistivity of a semiconductor sample). Further, the van der Pauw (van der Pauw) method uses a film-like sample as a model, and therefore the present inventors considered that it is possible to measure a thin film such as a silicon epitaxial layer of a silicon epitaxial wafer with high accuracy.

For the above reasons, the present inventors have adopted the van der pauw method to measure the resistivity of the silicon layer. Further, the present inventors have repeatedly conducted studies and newly found that, as a pretreatment for resistivity measurement by the van der pauw method, formation of an oxide film and a charging treatment for negatively charging the surface of the oxide film reduce measurement variations in resistivity measurement of a silicon layer, and completed the above-described evaluation method.

The following describes the evaluation method in further detail.

< evaluation object >

The object of evaluation by the above evaluation method is a silicon layer. The silicon layer may be a single crystal silicon layer, or in one embodiment, may be a silicon epitaxial layer. The silicon epitaxial layer can be produced by epitaxially growing a film of p-type or n-type single crystal silicon. For the formation of the silicon epitaxial layer by epitaxial growth, a known technique can be applied. The conductivity type of the silicon epitaxial layer can be controlled by the type of dopant and the resistivity of the silicon epitaxial layer can be controlled by the dopant concentration.

In the evaluation of the silicon epitaxial layer, in order to suppress leakage of a current to a portion located under the silicon epitaxial layer at the time of resistivity measurement, it is preferable to measure the resistivity of the silicon epitaxial layer included in the evaluation sample having a pn structure by the van der Pauw method.

As for the pn structure, an evaluation sample having a pn structure can be produced by forming an n-type silicon epitaxial layer on a p-type semiconductor substrate. Further, an evaluation sample having a pn structure can be produced by forming a p-type silicon epitaxial layer on an n-type semiconductor substrate. In the pn structure, the depletion layer of the pn junction electrically separates the silicon epitaxial layer from the portion (substrate) located below. Note that if the resistivity of the silicon epitaxial layer becomes high, the intrinsic potential becomes small, and the width of the depletion layer becomes wide, so that the potential gradient may become small. In such a case, by applying a reverse voltage of the pn junction to the substrate side to increase the intrinsic potential, leakage of current to a portion (substrate) located below the silicon epitaxial layer can be further suppressed. For the application of the reverse voltage, a well-known technique can be applied.

The silicon epitaxial layer to be evaluated may preferably be a high-resistance silicon epitaxial layer having a resistivity measurement value of 200 Ω cm or more by an extended resistance method. The measurement value obtained by the spreading resistance method in the present invention and the present specification is a value obtained by a known measurement by the spreading resistance method. The silicon epitaxial layer can be measured for a resistivity of 200 Ω cm or more by the spreading resistance method, and measurement variations can be reduced as compared with other resistivity measurement methods. However, as described above, in the spreading resistance method, the present inventors have estimated that the higher the resistivity of the evaluation target, the larger the contact deviation between the probe and the sample, and as a result, the larger the measurement deviation. In contrast, according to the above evaluation method, it is possible to evaluate a high-resistance silicon epitaxial layer having a measured resistivity of 200 Ω cm or more by the spreading resistance method while reducing measurement variations in the measurement of resistivity. The measured resistivity value by the spreading resistance method of the silicon epitaxial layer to be evaluated may be 300 Ω cm or more, may be 400 Ω cm or more, and may be 500 Ω cm or more. Further, the measured value of resistivity of the silicon epitaxial layer to be evaluated by, for example, the spreading resistance method can be 1000 Ω cm or less, but even a silicon epitaxial layer showing a measured value of resistivity exceeding this value can reduce measurement variation in resistivity measurement according to the above-described evaluation method.

In one embodiment, the silicon layer to be evaluated by the above evaluation method may be an active layer of an SOI wafer having a single crystal silicon substrate, an insulating film (for example, an oxide film), and a silicon layer (referred to as an active layer) in this order. Since the SOI wafer has the insulating film, leakage of current to a portion located below the silicon layer can be suppressed during resistivity measurement.

The active layer to be evaluated may preferably be a high-resistance silicon layer having a measured resistivity of 200 Ω cm or more by an extended resistance method. As described above, in the spreading resistance method, the present inventors presume that the higher the resistivity of the evaluation target, the greater the contact deviation between the probe and the sample, and as a result, the greater the measurement deviation. In contrast, according to the above evaluation method, the high-resistance silicon layer having a measured resistivity of 200 Ω cm or more obtained by the spreading resistance method can be evaluated with reduced measurement variation in resistivity measurement. The measured value of the resistivity of the silicon layer to be evaluated by the spreading resistance method may be 300 Ω cm or more, 400 Ω cm or more, or 500 Ω cm or more. In addition, although the measured resistivity value of the silicon layer to be evaluated by, for example, the spreading resistance method can be 1000 Ω cm or less, even if the silicon layer shows a measured resistivity value exceeding this value, the measurement variation in the resistivity measurement can be reduced by the above-described evaluation method.

For the fabrication of SOI wafers, known techniques can be applied. The conductivity type of the active layer may be controlled by the kind of dopant, and the resistivity of the active layer may be controlled by the dopant concentration.

< pretreatment for resistivity measurement >

In the above evaluation method, the formation of an oxide film and the charging treatment were performed as pretreatment before resistivity measurement by the van der Pauw method. These pretreatments will be described in further detail below.

(formation of oxide film)

As a pretreatment for resistivity measurement, an oxide film is first formed on the surface of the silicon layer to be evaluated. The "oxide film" in the present invention and the present specification is a silicon oxide film (e.g., SiO)₂A film). Although a natural oxide film may be formed on the surface of the silicon layer, it is considered that the amount of negative charge retention of the natural oxide film by the charging treatment described later is small, and it is difficult to sufficiently reduce measurement variations. Therefore, in the above evaluation method, an oxide film is formed before the electrification treatment. The thickness of the oxide film on the surface of the silicon layer is preferably larger than the thickness of the natural oxide film, and from this viewpoint, it is preferably 1nm or more. The thickness of the oxide film refers to the total thickness of the natural oxide film and the oxide film formed by the oxide film formation, when the oxide film is formed without performing the natural oxide film removal process on the silicon epitaxial layer in which the natural oxide film is present. Further, the thickness of the oxide film on the surface of the silicon layer may be, for example, 20nm or less.

The oxide film formation may be performed by oxidation treatment using a treatment liquid, that is, wet oxidation, or may be performed by oxidation treatment without using a treatment liquid, that is, dry oxidation, and dry oxidation is preferable from the viewpoint of simplicity of oxidation treatment. The dry oxidation may be performed by a known method such as thermal oxidation or plasma treatment, and is preferably thermal oxidation. The thermal oxidation can be performed by disposing an evaluation sample including a silicon layer to be evaluated in a heated oxidizing atmosphere. The oxidizing atmosphere is an atmosphere containing at least an oxidizing gas, and may be, for example, an atmosphere containing 10 to 100 vol% of oxygen. The oxidizing atmosphere may have an atmosphere temperature of, for example, 700 to 1300 ℃ and a heating time of, for example, 1 to 1000 minutes. However, the thermal oxidation conditions are not limited to the above examples.

(electrification treatment)

In the above evaluation method, a charging treatment for negatively charging the surface of the formed oxide film is performed. In general, in the van der pauw (van der pauw) method, electrodes are provided at four corners of a sample surface having a square shape on a surface on which the electrodes are formed, a voltage is applied between 2 adjacent electrodes, and the applied voltage is measured by the remaining 2 electrodes. Before the charging treatment, the shape of the sample is preferably adjusted to a shape suitable for measurement by the van der Pauw method. Before the charging treatment, an electrode for use in the van der Pauw method (hereinafter referred to as "measurement electrode") is preferably provided. The measurement electrode can be formed by a known method. The measurement electrode is preferably a metal electrode having good ohmic properties with respect to the silicon layer to be evaluated. From this viewpoint, the measurement electrode is preferably a metal (for example, Mg) electrode having a work function of 3.7eV or less when the silicon layer to be evaluated is n-type, and is preferably a metal (for example, Au) electrode having a work function of more than 3.7eV when the silicon layer to be evaluated is p-type.

The charging treatment in the above evaluation method is a treatment for negatively charging the surface of the oxide film on the surface of the silicon layer. Since the oxide film is positively charged, it is considered that, if the oxide film is not negatively charged, an accumulation layer (in the case where the silicon layer is n-type) or an inversion layer (in the case where the silicon layer is p-type) is formed in the region of the silicon layer on the oxide film side, and if the silicon layer is formed to have high resistance, the influence thereof becomes large, and it is estimated that the measurement variation is caused. On the other hand, by negatively charging the surface of the oxide film, it is considered that the formation of the accumulation layer can be suppressed in the case of the n-type, and the formation of the inversion layer can be suppressed in the case of the p-type. Thus, the present inventors have estimated that measurement variations in resistivity measurement by the van der Pauw (van der Pauw) method can be reduced.

One example of the charging treatment is a treatment in which negative charges are accumulated on the surface of the oxide film. Such a treatment may be a corona discharge treatment. The surface of the oxide film can be negatively charged by generating negative ions by corona discharge.

The more the number of times of the corona discharge treatment is, the more negative charges can be accumulated on the surface of the oxide film. From the viewpoint of further reducing measurement variation, it is preferable to accumulate more negative charges. On the other hand, the larger the amount of negative charges accumulated on the surface of the oxide film, the larger the value of resistivity obtained by measurement tends to be. The reason for this is considered to be that, as the amount of negative charges accumulated on the surface of the oxide film increases, an inversion layer (in the case where the silicon layer is n-type) or an accumulation layer (in the case where the silicon layer is p-type) is formed in a region on the oxide film side of the silicon layer. In this case, for example, a preliminary experiment may be performed to determine a correction formula, and a value obtained by correcting the value of the resistivity obtained by the determination formula by the correction formula may be used as the resistivity used for the evaluation of the silicon layer. However, for example, since the allowable range of the resistivity obtained under a certain charging condition is set in advance and the silicon layer can be evaluated by whether or not the resistivity obtained under the charging condition is within the allowable range, the silicon layer can be sufficiently evaluated regardless of the tendency that the value of the resistivity obtained by measurement tends to be large or small.

The number of times of the corona discharge treatment is preferably 1 or more from the viewpoint of reducing measurement variation, and is preferably 5 or less, more preferably 4 or less, and further preferably 3 or less from the viewpoint of preventing occurrence of dielectric breakdown of an oxide film. As described above, from the viewpoint of reducing measurement variations, it is considered that it is preferable to suppress formation of an accumulation layer (in the case where the silicon layer is n-type) or an inversion layer (in the case where the silicon layer is p-type), and to suppress increase in the value of resistivity obtained by measurementIn order to suppress the increase in the measurement value of the resistivity and reduce the measurement variation, the silicon layer is preferably in a flat band state or a state close to the flat band state, more preferably in a flat band state, and from this viewpoint, the number of times of the corona discharge treatment is preferably in the range of 1 to 3 times, more preferably 1 or 2 times, and most preferably 1 time, and from this viewpoint, the total charge amount per unit area obtained by the corona discharge treatment is preferably-1.0 × 10¹²charges/cm²Hereinafter, more preferably-3.0 × 10¹¹～-8.5×10¹¹charges/cm²The range of (1).

In addition, another mode of the charging process is a voltage application process. The voltage application treatment can be performed by, for example, providing an electrode on the surface of the oxide film and applying a voltage between the electrode and the back surface of the wafer (for example, the substrate side in a pn structure). By applying a voltage so as to apply a negative voltage to the oxide film, the oxide film surface can be negatively charged. The absolute value of the applied voltage is preferably one tenth or more of the flat band voltage of the silicon layer from the viewpoint of suppressing the formation of the accumulation layer (in the case where the silicon layer is n-type) or the inversion layer (in the case where the silicon layer is p-type) and reducing the measurement variation. In addition, from the viewpoint of preventing occurrence of dielectric breakdown of the oxide film, the absolute value of the applied voltage is preferably a value lower than the dielectric breakdown voltage of the oxide film. As described above with respect to the corona discharge treatment, the silicon layer is preferably in a flat band state or a state close to the flat band state, and more preferably in a flat band state. The flat band voltage of the silicon layer can be determined by CV (capacitance-voltage) measurement. Therefore, for example, CV measurement may be performed on the silicon layer to be evaluated, or CV measurement may be performed on a silicon layer formed under the same conditions as the silicon layer to be evaluated, and the voltage application condition may be determined based on the obtained flat band voltage.

< resistivity measurement by Van der Pauw method >

In the above evaluation method, after the above charging treatment, the resistivity of the silicon layer was measured by the van der Pauw method. A method of measuring resistivity by the van der Pauw method is known, and in the above evaluation method, resistivity measurement by the van der Pauw method can be performed by a known method. For example, when the measurement by the van der Pauw method is performed on the same silicon layer a plurality of times, the measurement variation can be reduced by performing the oxide film formation and the charging treatment as the pretreatment, as compared with the measurement without performing the oxide film formation and/or the charging treatment. The small measurement variation means high measurement accuracy and high reliability of the measurement result. Therefore, according to the method capable of reducing the measurement variation, even the measurement value obtained in 1 measurement can obtain a highly reliable measurement value.

In the above evaluation method, as an index for evaluating the silicon layer to be evaluated, a value obtained by 1 resistivity measurement may be used, or a representative value (for example, arithmetic mean, minimum value, maximum value, or the like) of measurement values obtained by a plurality of measurements may be used.

[ method for producing silicon epitaxial wafer ]

A method for manufacturing a silicon epitaxial wafer according to an embodiment of the present invention (hereinafter referred to as a "first manufacturing method") includes:

manufacturing a candidate silicon epitaxial wafer to be shipped as a product;

further comprising:

evaluating a silicon epitaxial layer for evaluation formed in the same or nearly the same epitaxial growth step as a silicon epitaxial layer included in the silicon epitaxial wafer that is a candidate for shipment from the factory as a product, by the evaluation method; and

if the resistivity obtained for the silicon epitaxial layer for evaluation is within a range that is acceptable for good products, the silicon epitaxial wafer that is a candidate for shipment as a product is subjected to preparation for shipment as a product.

A method for manufacturing a silicon epitaxial wafer according to another embodiment of the present invention (hereinafter referred to as "second manufacturing method") includes:

forming a silicon epitaxial layer for evaluation under a test epitaxial growth condition;

evaluating the silicon epitaxial layer for evaluation by the evaluation method;

determining an epitaxial growth condition, which is a condition for changing the test epitaxial growth condition, as an epitaxial growth condition used in a product manufacturing step or determining the test epitaxial growth condition as an epitaxial growth condition used in the product manufacturing step, based on the resistivity obtained for the evaluation silicon epitaxial layer; and

the silicon epitaxial wafer for product shipment is manufactured through a product manufacturing step including the epitaxial growth step performed under the above-identified epitaxial growth conditions.

The first manufacturing method may be referred to as a method using the results of the evaluation by the above-described evaluation method in order to ensure the quality of silicon epitaxial wafers shipped as products. The second manufacturing method may be referred to as a method using the result of evaluation by the above-described evaluation method for so-called step management. In both methods, the evaluation using the evaluation method according to one embodiment of the present invention described above was performed.

The first and second manufacturing methods will be described in further detail below.

< first production method >

(production of silicon epitaxial wafer to be shipped as a product)

The production of a candidate silicon epitaxial wafer shipped as a product in the first production method can be performed by a known production process of a silicon epitaxial wafer. Generally, a silicon epitaxial wafer is manufactured by placing a single crystal silicon wafer such as a polished wafer in an epitaxial growth furnace, and epitaxially growing (vapor phase growth) a film of single crystal silicon on the surface of the polished wafer to form a silicon epitaxial layer. The conductivity type of the silicon epitaxial layer can be controlled by the kind of dopant, and the resistivity can be controlled by the concentration of the dopant. By mixing a source gas of a dopant with a gas for vapor phase growth, a silicon epitaxial layer containing a dopant can be formed. In the first manufacturing method, it is determined whether or not the silicon epitaxial layer included in the silicon epitaxial wafer that is a candidate for shipment as a product has a resistivity of a good product level, based on the result of evaluating the resistivity of the silicon epitaxial layer by the above-described evaluation method.

(evaluation of formation of silicon epitaxial layer and evaluation)

The silicon epitaxial layer for evaluation is a silicon epitaxial layer formed by the same or nearly the same epitaxial growth step as a silicon epitaxial layer included in a silicon epitaxial wafer that is a candidate for shipment from a factory as a product. The "same epitaxial growth step" refers to an epitaxial growth step performed under the same epitaxial growth conditions in the same epitaxial growth furnace as the silicon epitaxial layer of the candidate silicon epitaxial wafer shipped as a product. The "approximately same epitaxial growth step" refers to an epitaxial growth step performed under the same epitaxial growth conditions in the same epitaxial growth furnace as the silicon epitaxial layers of the candidate silicon epitaxial wafers shipped as products, except for the time (processing time) for performing epitaxial growth. In the "same epitaxial growth furnace", the silicon epitaxial layer of the silicon epitaxial wafer which is a candidate for shipment as a product may be formed and the silicon epitaxial layer may be evaluated to be formed first, but the furnace is not conservatively controlled between them. The epitaxial growth conditions include the type, flow rate, mixing ratio, and treatment time of the gas for vapor phase growth. Variations that are generally possible due to the device or the like are allowed for "the same epitaxial growth conditions". The silicon single crystal substrate for growth evaluation of the silicon epitaxial layer does not need to be the same silicon single crystal wafer as a candidate silicon epitaxial wafer shipped as a product. As described above, it is preferable to select the substrate so that an evaluation sample having a pn structure can be produced. The evaluation sample obtained in this manner is evaluated by the evaluation method according to one embodiment of the present invention. The details of the evaluation are as described previously.

(determination of whether or not it is good)

In the first manufacturing method, the resistivity obtained for the evaluation silicon epitaxial layer is used to determine whether or not the silicon epitaxial wafer shipped as a product candidate has the resistivity required for the product (determination of whether or not the wafer is good). The resistivity used for determining whether the silicon epitaxial layer is good or not good may be 1 measurement value obtained for 1 evaluation silicon epitaxial layer, or may be a representative value (for example, arithmetic mean, minimum value, maximum value, or the like) of measurement values obtained for 1 evaluation silicon epitaxial layer by a plurality of measurements. In addition, the resistivity of a plurality of silicon epitaxial layers for evaluation may be measured by the above-described evaluation method, and a representative value (for example, arithmetic mean, minimum value, maximum value, or the like) of the measured values may be used to determine whether the silicon epitaxial layers are good or not. The determination of whether the resistivity is good or not is performed based on whether the resistivity is within a range allowed for good product or not. The allowable range of good product is determined according to the required quality of the product wafer. When the resistivity is within the allowable range as a good product, a preparation for shipping as a product is performed on a silicon epitaxial wafer that is a candidate for shipping as a product. Examples of preparations for shipment as products include bundles and the like. In this way, according to the first manufacturing method, a silicon epitaxial wafer having a resistivity required for a product wafer can be stably supplied to the market. On the other hand, when the resistivity is out of the range allowed by good products, preparation for shipment as a target product is not performed on a silicon epitaxial wafer that is a candidate for shipment as a product. In this case, the silicon epitaxial wafer that is a candidate for shipment as a product may be subjected to, for example, preparation for shipment as another level of product, or may be discarded.

< second production method >

In the second manufacturing method, in order to determine the epitaxial growth conditions employed in the product manufacturing steps, the evaluation results by the above-described evaluation method are used. The details are as follows.

(evaluation of formation of silicon epitaxial layer and evaluation)

"test epitaxial growth conditions" refers to candidate epitaxial growth conditions employed in the product fabrication steps. Under the epitaxial growth conditions, a silicon epitaxial layer for evaluation was formed. In the test epitaxial growth conditions, the kind of gas for vapor phase growth, flow rate, mixing ratio, treatment time, the kind of epitaxial growth furnace used, the kind of member used in the epitaxial growth furnace, and the like may be included.

As in the case of the silicon epitaxial layer for evaluation in the first production method, the silicon epitaxial layer for evaluation may be formed on a single crystal silicon substrate, and the substrate is preferably selected so that a sample for evaluation having a pn structure can be produced. The obtained evaluation sample was evaluated by the evaluation method according to one embodiment of the present invention. The details of the evaluation are as described previously.

(determination of epitaxial growth conditions used in the production steps)

In the second manufacturing method, the resistivity obtained for evaluating the silicon epitaxial layer is used to determine the epitaxial growth conditions used in the product manufacturing step. The resistivity used for this determination may be 1 measurement value obtained for 1 silicon epitaxial layer for evaluation, or may be a representative value (for example, arithmetic mean, minimum value, maximum value, or the like) of measurement values obtained for 1 silicon epitaxial layer for evaluation and measured for a plurality of times. In addition, the resistivity of a plurality of silicon epitaxial layers for evaluation can be measured by the above-described evaluation method, and a representative value (for example, arithmetic mean, minimum value, maximum value, or the like) of the measured values can be used. The epitaxial growth conditions used in the product production step may be determined based on whether or not the resistivity obtained for evaluating the silicon epitaxial layer is within a range that is acceptable for good products. As in the first manufacturing method, the allowable range of good products may be determined according to the quality required for the product wafer. In the case where the above-described resistivity is within the range allowed as a good product, test epitaxial growth conditions under which formation of a silicon epitaxial layer is evaluated may be used as epitaxial growth conditions used in the product manufacturing step. By employing the epitaxial growth conditions determined in this manner in the production step and shipping the production silicon epitaxial wafer through the step of forming the silicon epitaxial layer under the epitaxial growth conditions, it is possible to stably supply the silicon epitaxial wafer having the resistivity required for the production wafer to the market.

On the other hand, when the resistivity obtained for the silicon epitaxial layer to be evaluated is out of the acceptable range, the epitaxial growth conditions obtained by changing the test epitaxial growth conditions are determined as the epitaxial growth conditions used in the product manufacturing step. The conditions for applying the modification are preferably conditions in which the influence on the resistivity is taken into consideration. Examples of such conditions include the flow rate of the source gas of the dopant. By employing the epitaxial growth conditions obtained by changing the test epitaxial growth conditions in the production process and shipping the production silicon epitaxial wafer through the step of forming the silicon epitaxial layer under the epitaxial growth conditions, the silicon epitaxial wafer having the resistivity required for the production wafer can be stably supplied to the market. The silicon epitaxial layer for evaluation may be formed again under the conditions obtained by changing the test epitaxial growth conditions, evaluated by the evaluation method according to one embodiment of the present invention, and subjected to the determination 1 or 2 or more times to adopt the conditions or further change in the product manufacturing process.

For the details of the first manufacturing method and the second manufacturing method, known techniques related to the manufacture of silicon epitaxial wafers can be applied.