JP5742739B2 - Screening method of silicon substrate for metal contamination assessment - Google Patents

Description

  The present invention relates to a method for selecting a silicon substrate for evaluating metal contamination used when evaluating metal contamination of a heat treatment furnace in a silicon substrate manufacturing process or a device manufacturing process.

  In the manufacturing process of a semiconductor silicon substrate (wafer) and the manufacturing process of a semiconductor device, if the wafer is contaminated with metal impurities or the like, the performance of the product is adversely affected. Therefore, reduction of metal contamination is a very important issue.

  As a method for evaluating metal contamination during a silicon substrate manufacturing process or a device manufacturing process, measurement of a recombination lifetime by a microwave photoconductive decay method (μ-PCD method) is widely used. In this μ-PCD method, a light pulse having an energy larger than the band gap of a silicon single crystal is first irradiated to generate excess carriers in the wafer. The electrical conductivity of the wafer increases due to the generated excess carriers, but thereafter, the electrical conductivity decreases with the lapse of time because excess carriers disappear due to recombination. By detecting this change as a time change of the reflected microwave power and analyzing it, the recombination lifetime can be obtained. The recombination lifetime is shortened when there are metal impurities or defects that form a level that becomes a recombination center in the forbidden band. From this, metal impurities, crystal defects, and the like in the wafer can be evaluated by measuring the recombination lifetime (for example, Non-Patent Document 1).

  When the sample to be evaluated has a wafer shape, excess carriers generated by the light pulse are not only recombined inside the wafer and disappear, but also diffused to the front and back surfaces of the wafer and disappear due to surface recombination. Therefore, in order to evaluate metal contamination inside the wafer, it is necessary to suppress surface recombination on the front surface and the back surface. As a method for suppressing surface recombination, thermal oxidation treatment (oxide film passivation) and electrolytic solution treatment (sometimes abbreviated as chemical passivation treatment or CP treatment) are generally used. In oxide film passivation, care must be taken not to generate metal contamination or crystal defects in a heat treatment process for forming an oxide film. Therefore, when evaluating metal contamination in a heat treatment furnace other than the oxidation furnace, for example, an epitaxial growth furnace for manufacturing an epitaxial wafer, a chemical passivation treatment is used.

  Known solutions for chemical passivation treatment include iodine alcohol solution (for example, Non-Patent Document 2) and quinhydrone alcohol solution (for example, Patent Document 1). In the case of a quinhydrone alcohol solution, it takes time until the surface passivation effect is stabilized (for example, Non-Patent Document 3). Therefore, iodine alcohol solution is used when it is desired to obtain the evaluation result of metal contamination as soon as possible.

JP 2002-329692 A

JEIDA-53-1997 “Method for measuring recombination lifetime of silicon wafer by reflection microwave photoconductive decay method” T. T. et al. S. Horanyi et al. , Appl. Surf. Sci. 63 (1993) 306. H. Takato et al. , Jpn. J. et al. Appl. Phys. 41 (2002) L870.

  As the performance of semiconductor devices increases, even a small amount of metal contamination has an adverse effect on device performance, and reducing metal contamination is an extremely important issue. In particular, in an image sensor such as a CCD (Charge Coupled Device) or CIS (CMOS Image Sensor), a weak white flaw or dark current becomes a problem as the light receiving sensitivity and resolution are improved, and a very small amount of metal contamination is adversely affected. Is concerned. For this reason, in an epitaxial wafer widely used as a substrate for an image sensor, it is strongly desired to reduce not only metal contamination in the device manufacturing process but also metal contamination in the process of manufacturing the epitaxial wafer.

  In order to reduce metal contamination in a silicon substrate manufacturing process or a device manufacturing process, a method for evaluating a very small amount of metal contamination with high sensitivity and high accuracy is required.

  As described above, the recombination lifetime measurement by the μ-PCD method is widely used as a method for evaluating metal contamination. In order to evaluate metal contamination with high sensitivity by this method, it is desirable to use a silicon substrate having a high initial value of the recombination lifetime at a stage before metal contamination as a silicon substrate for metal contamination evaluation.

  Conventionally, as a method of preparing a silicon substrate for metal contamination evaluation, the resistivity of the silicon substrate is limited, for example, n-type or p-type and the resistivity is 1 Ω · cm or more (Non-patent Document 1). . Furthermore, in order to evaluate metal contamination with high sensitivity and high reliability by the recombination lifetime, it is disclosed to use a silicon wafer having a resistivity in the range of 100 to 10000 Ω · cm (Japanese Patent Laid-Open No. 2010-040793). Issue gazette).

  However, the initial value of the recombination lifetime of the silicon substrate is not determined only by the resistivity, but the grown-in defect (crystal growth introduction defect) formed during crystal growth becomes a carrier recombination center. (Hereinafter, recombination centers formed during crystal growth may be referred to as grown-in recombination centers). Therefore, there is a problem that a silicon substrate with a high initial value of the recombination lifetime cannot always be produced simply by limiting the resistivity range of the silicon substrate. Therefore, as one method for selecting the silicon substrate for metal contamination evaluation, a method of measuring the initial value of the recombination lifetime and using a silicon substrate having a high actual measurement value as the silicon substrate for metal contamination evaluation is conceivable.

  On the other hand, the grown-in recombination center is thermally unstable and may disappear in the heat treatment process for evaluating metal contamination. Therefore, even a substrate having a low initial value of the recombination lifetime due to the grown-in recombination center may be used as a silicon substrate for metal contamination evaluation. However, from the result of actual measurement of the initial value, the silicon substrate for metal contamination evaluation As a result, there was a problem that the silicon substrate was judged as unsuitable. Of course, there is no problem if a silicon crystal having a very low density of grown-in recombination centers is produced. However, in order to produce such a silicon crystal, the production conditions are extremely limited, and the productivity is lowered. The problem of high costs arises.

  The present invention has been made in view of such problems, and provides a method for selecting a silicon substrate for metal contamination evaluation that can evaluate metal contamination of a heat treatment furnace in a silicon substrate manufacturing process or device manufacturing process with high sensitivity. The purpose is to provide.

  The present invention has been made to solve the above problems, and is a method for selecting a silicon substrate for metal contamination evaluation used when evaluating metal contamination in a heat treatment furnace using a measured value of recombination lifetime. A step of preparing a silicon substrate that is a candidate for the metal contamination evaluation silicon substrate from a silicon single crystal ingot, a step of heat-treating the candidate silicon substrate, and a surface of the candidate silicon substrate. A step of performing a surface passivation treatment, a step of measuring the recombination lifetime of the silicon substrate after the heat treatment and the surface passivation treatment by a microwave photoconductive decay method, and a measurement value obtained by the measurement When it is a predetermined reference value or more, the silicon substrate on which the measurement is performed is acceptable as a metal contamination evaluation silicon substrate. And a step of selecting a silicon substrate produced from the same silicon single crystal ingot as the silicon single crystal ingot produced from the silicon substrate judged acceptable by the judgment as the metal contamination evaluation silicon substrate. A method for selecting a silicon substrate for metal contamination evaluation is provided.

  By such a method for selecting a metal contamination evaluation silicon substrate, it is possible to select a metal contamination evaluation silicon substrate based on a recombination lifetime measurement value obtained by removing the influence of a thermally unstable grown-in recombination center.

  In this case, the step of preparing the candidate silicon substrate includes a step of cutting the silicon substrate from the silicon single crystal ingot, and a step of removing damage layers due to the cutting of the front and back surfaces of the silicon substrate. It is preferable.

  If it is a silicon substrate from which damage due to cutting of the front surface and the back surface is removed by such steps, the recombination lifetime in a state where the surface recombination is suppressed can be measured.

  The heat treatment is preferably performed at a heat treatment temperature of 1000 ° C. to 1200 ° C. and a heat treatment time of 1 minute to 60 minutes.

  By performing heat treatment under such heat treatment conditions, the grown-in recombination centers can be effectively eliminated.

  The surface passivation treatment can be performed by chemical passivation treatment or by forming an oxide film on the surface of the silicon substrate.

  Since the chemical passivation treatment has a high passivation effect, it is preferable to perform the surface passivation treatment by the chemical passivation treatment because the influence of surface recombination can be more effectively suppressed. The surface passivation treatment can also be performed by forming an oxide film on the surface of the silicon substrate. When an oxide film is formed on the surface of the silicon substrate by the heat treatment, the oxide film can be used as an oxide film for surface passivation.

  Moreover, when performing a surface passivation process by a chemical passivation process, it is preferable to perform the said chemical passivation process using an iodine alcohol solution.

  Such a chemical passivation treatment has a high passivation effect and stabilizes soon after the treatment, so that the recombination lifetime can be measured quickly.

  The reference value is preferably 3 msec or more.

  By adopting such a reference value, the value of the recombination lifetime of the silicon substrate for metal contamination evaluation in the stage before metal contamination is 3 msec or more, so the deterioration of the recombination lifetime due to metal contamination is increased. Contamination can be evaluated with high sensitivity.

  According to the method for selecting a silicon substrate for evaluating metal contamination according to the present invention, a silicon substrate for evaluating metal contamination can be selected based on a recombination lifetime measurement value from which the influence of a thermally unstable Grown-in recombination center is removed. . As a result, it is possible to select a metal contamination evaluation silicon substrate capable of highly sensitively evaluating the metal contamination of the heat treatment furnace in the silicon substrate manufacturing process and the device manufacturing process.

It is the flowchart which showed the outline of the selection method of the silicon substrate for metal contamination evaluation which concerns on this invention. It is the graph which showed the relationship between the heat processing temperature and recombination lifetime in an experiment example. It is the graph which showed the relationship between the heat processing time and recombination lifetime in an experiment example.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings, but the present invention is not limited thereto.

  FIG. 1 shows an outline of a method for selecting a silicon substrate for metal contamination evaluation according to the present invention.

  First, as shown in FIG. 1A, a silicon substrate that is a candidate for a silicon substrate for metal contamination evaluation is prepared from a silicon single crystal ingot (step a). The step of preparing the candidate silicon substrate is preferably based on a step of cutting the silicon substrate from the silicon single crystal ingot and a step of removing the damaged layer by cutting the front and back surfaces of the silicon substrate. The method for removing the damaged layer is not particularly limited. For example, the damaged layer can be removed by removing the surface layer by about 50 μm or more by chemical etching. In order to improve the flatness of the silicon substrate surface, surface grinding can be performed after the silicon substrate is cut and before chemical etching. Further, it may be a processing process including chemical mechanical mirror polishing as used in the manufacturing process of a normal silicon substrate product. In any case, it is important to remove the damage layer caused by cutting. The diameter of the silicon substrate prepared here is preferably the same as the diameter of a wafer processed in a heat treatment furnace for evaluating metal contamination, and can be, for example, 6 to 12 inches (150 to 300 mm). The thickness of the silicon substrate can be set to 0.5 to 2 mm, for example.

  In the present invention, the method for preparing the silicon single crystal ingot before cutting the silicon substrate is not particularly limited. For example, it can be produced by the Czochralski method (CZ method) or the floating zone method (FZ method). The “silicon single crystal ingot” in the present invention refers to an ingot-shaped single crystal rod made of a silicon single crystal, as long as it can be produced by cutting a wafer-shaped silicon substrate from the ingot. For example, the “silicon single crystal ingot” in the present invention is a concept including a single crystal rod (often referred to as a silicon single crystal block) in a state where a silicon single crystal ingot produced by the CZ method or the FZ method is divided.

  As described above, after preparing a silicon substrate as a candidate for a metal substrate for metal contamination evaluation, as shown in FIG. 1B, the candidate silicon substrate is heat-treated (step b). As the heat treatment furnace used in this heat treatment, various heat treatment furnaces capable of heat treating the semiconductor substrate can be used, and an epitaxial growth furnace can also be used.

  In this heat treatment, the heat treatment temperature is preferably 1000 ° C. or more and 1200 ° C. or less, and the heat treatment time is preferably 1 minute or more and 60 minutes or less. Moreover, it is more preferable to perform this heat treatment in an oxidizing atmosphere. The fact that such heat treatment conditions are preferable is based on the knowledge obtained by experimental examples described later.

  If the heat treatment temperature is 1000 ° C. or higher, the grown-in recombination centers can be effectively eliminated. If heat processing temperature is 1200 degrees C or less, the fall of the recombination lifetime by metal contamination and a lattice defect being newly introduce | transduced by the heat processing can be suppressed. If the heat treatment time is 1 minute or longer, the grown-in recombination centers can be effectively eliminated. There is no problem even if the heat treatment time is long, but if it is too long, it becomes inefficient, and therefore it is preferably 60 minutes or less.

  When the heat treatment atmosphere is an oxidizing atmosphere, an oxide film is formed on the surface of the silicon substrate. By forming the oxide film, interstitial silicon is injected into the silicon substrate, and Grown-in recombination centers caused by vacancies can be effectively eliminated. In addition, by forming an oxide film, metal contamination in the heat treatment can be suppressed, and further, the oxide film can be used as an oxide film for passivation, so that an independent chemical passivation treatment is performed as described later. The recombination lifetime can be measured without performing

  Next, as shown in FIG. 1C, surface passivation treatment is performed on the surface of the candidate silicon substrate (step c). This surface passivation treatment is preferably performed by chemical passivation treatment, but can also be performed by forming an oxide film on the surface of the silicon substrate (oxide film passivation). When performing surface passivation by oxide film passivation, the heat treatment step (step b) and the surface passivation treatment step (step c) can be performed simultaneously.

  Since the chemical passivation has a higher passivation effect than the oxide film passivation and can more effectively suppress the influence of surface recombination, the recombination life can be achieved by employing the chemical passivation treatment in this surface passivation treatment step (step c). Time can be evaluated more strictly. In the case where the heat treatment step (step b) is a heat treatment in which no oxide film is formed, such as a heat treatment using an epitaxial growth furnace, chemical passivation is easier than oxide film passivation in which an oxide film is formed later.

  The chemical passivation treatment is preferably performed using an iodine alcohol solution. Thus, the chemical passivation treatment using an iodine alcohol solution has a high passivation effect and stabilizes quickly after the treatment, so that the recombination lifetime can be measured quickly.

  When an oxide film is formed in the heat treatment step (step b), the recombination lifetime can be measured without performing chemical passivation treatment using the oxide film as a passivation oxide film. However, since the oxide film passivation has a lower passivation effect than the chemical passivation process, if it is desired to strictly evaluate the recombination lifetime, the oxide film formed in the heat treatment process (process b) is removed and then the chemical passivation process is performed. It is preferable to carry out.

  Next, as shown in FIG. 1D, the recombination lifetime of the silicon substrate after the heat treatment and the surface passivation treatment is measured by the microwave photoconductive decay method (μ-PCD method) (step d). ). The measurement conditions in the μ-PCD method may be those generally used. For example, measurement can be performed under the conditions described in Non-Patent Document 1. A commercially available measuring apparatus can be used. Note that, when the surface passivation treatment is performed by the chemical passivation treatment, if a natural oxide film is formed on the surface of the silicon substrate, the natural oxide film is removed with a hydrofluoric acid aqueous solution before the chemical passivation treatment. When the heat treatment furnace used in the heat treatment step (step b) is an epitaxial growth furnace, the recombination lifetime of the silicon substrate after the epitaxial growth can be measured.

  Next, as shown in FIG. 1 (e), when the measurement value obtained by the measurement in step d is equal to or greater than a predetermined reference value, the measured silicon substrate passes the silicon contamination evaluation silicon substrate. (Step e). When the measured value of the recombination lifetime is greater than or equal to the reference value, the silicon substrate for metal contamination evaluation is accepted. By setting this reference value to 3 msec, for example, it is possible to select a silicon substrate for metal contamination evaluation for highly sensitive evaluation of a very small amount of metal contamination that adversely affects the performance of recent high performance devices. As described above, when the reference value is 3 msec, the value of the recombination lifetime of the silicon substrate for metal contamination evaluation in the stage before metal contamination is 3 msec or more, and deterioration of the recombination lifetime due to metal contamination can be increased. .

  Next, as shown in FIG. 1 (f), silicon produced from the same silicon single crystal ingot (same crystal lot) as the silicon single crystal ingot from which the silicon substrate judged acceptable by the judgment in step e was produced. The substrate is sorted as a metal contamination evaluation silicon substrate (step f). The silicon substrate may be produced from the same silicon single crystal ingot as that produced the silicon substrate determined to be acceptable after the steps a to e are completed, but the metal contamination evaluation is performed from the silicon single crystal ingot in the step a. When a silicon substrate that is a candidate for a silicon substrate for manufacturing is manufactured, a plurality of silicon substrates may be manufactured at the same time.

  In this way, the silicon substrate for metal contamination evaluation selected by the method for selecting a silicon substrate for metal contamination evaluation according to the present invention has a high sensitivity to a very small amount of metal contamination that adversely affects the performance of high performance devices in recent years. It is suitable as a silicon substrate for metal contamination evaluation for evaluation.

  Next, in the present invention, the reason why the heat treatment temperature is preferably 1000 ° C. or more and 1200 ° C. or less and the heat treatment time is preferably 1 minute or more and 60 minutes or less is based on the knowledge obtained by the following experiment.

(Experimental example)
By the floating zone method, a silicon single crystal ingot having a conductivity type of p-type, a resistivity of 4500 Ω · cm (crystal a), a conductivity type of n-type, and a resistivity of 43 Ω · cm (crystal b) was grown. Each silicon single crystal ingot has a diameter of 200 mm and a crystal axis orientation of <100>. Then, from these silicon single crystal ingots, a mirror-polished silicon substrate was produced by a standard wafer processing process.

  Next, in order to investigate the value of the recombination lifetime before the heat treatment on the fabricated silicon substrate, the chemical passivation treatment using iodine ethanol solution to remove the natural oxide film with hydrofluoric acid aqueous solution and suppress the surface recombination Then, the recombination lifetime was measured by the μ-PCD method. As a result, the recombination lifetime of crystal a was 1067 μsec, and the recombination lifetime of crystal b was 619 μsec.

  Next, the heat treatment time is fixed to 30 minutes for the silicon substrates manufactured from the same silicon single crystal ingot, and the heat treatment temperatures are 500 ° C., 600 ° C., 700 ° C., 800 ° C., 900 ° C., 1000 ° C., 1100 ° C. The temperature was changed to 1200 ° C. and heat treatment was performed in an oxygen atmosphere. The heat treatment temperature was fixed at 1000 ° C., and the heat treatment time was changed to 10 seconds, 1 minute, 10 minutes, 30 minutes, and 60 minutes, and the heat treatment was performed in an oxygen atmosphere.

  Next, the oxide film formed on the surface of the heat-treated silicon substrate is removed with a hydrofluoric acid aqueous solution and subjected to chemical passivation using an iodine ethanol solution to suppress surface recombination, and then μ-PCD. The recombination lifetime was measured by the method. When the heat treatment condition was 1000 ° C. for 30 minutes, the recombination lifetime was measured using the oxide film formed by the heat treatment as an oxide film for passivation before removing the oxide film.

  The measurement results of the recombination lifetime when the heat treatment time is fixed at 30 minutes and the heat treatment temperature is changed are shown in FIGS. The graph in FIG. 2A corresponds to the crystal a, and the graph in FIG. 2B corresponds to the crystal b.

  As shown in FIG. 2A, in the case of the crystal a, it was found that when the heat treatment temperature was 1000 ° C. or higher, the recombination lifetime value was higher than that before the heat treatment. This result shows that the grown-in recombination centers disappeared by heat treatment at 1000 ° C. or higher. In the case of the crystal b, as shown in FIG. 2B, it was found that when the heat treatment temperature was 900 ° C. or higher, the recombination lifetime value was higher than that before the heat treatment. The reason why the recombination lifetime of the crystal b is higher than the temperature of the crystal a is considered to be that the grown-in recombination center of the crystal b is thermally unstable.

  As a result of measuring the lifetime before removing the oxide film on a silicon substrate subjected to heat treatment at 1000 ° C. for 30 minutes, the result was 6320 μsec for crystal a and 2368 μsec for crystal b. In either case, although the recombination lifetime was slightly lower than the recombination lifetime measured by removing the oxide film with a hydrofluoric acid solution and then performing chemical passivation, the recombination lifetime was higher than before the heat treatment. I found out.

  The measurement results of the recombination lifetime when the heat treatment temperature is fixed at 1000 ° C. and the heat treatment time is changed are shown in FIGS. The graph in FIG. 3A corresponds to the crystal a, and the graph in FIG. 3B corresponds to the crystal b.

  As shown in FIG. 3, it was found that for any crystal, when the heat treatment time was 1 minute or longer, the recombination lifetime value was higher than that before the heat treatment. This result shows that the Grown-in recombination center can be sufficiently eliminated if the heat treatment time is 1 minute or longer.

  From these results, when the heat treatment temperature is at least 1000 ° C. and the heat treatment time is 1 minute or longer, the grown-in recombination centers can be more effectively eliminated, and the value of the recombination lifetime is increased. I understood it.

  EXAMPLES Hereinafter, although an Example and a comparative example are given and this invention is demonstrated further more concretely, these do not limit this invention.

Example 1
A silicon single crystal ingot having a conductivity type of p-type and a resistivity of 4500 Ω · cm was grown by the floating zone method. The diameter is 200 mm, and the crystal axis orientation is <100>. And from the silicon single crystal ingot, a silicon substrate having a mirror polished finish was produced by a standard wafer processing process.

  Next, in order to investigate the value of the recombination lifetime before the heat treatment on the fabricated silicon substrate, the chemical passivation treatment using iodine ethanol solution to remove the natural oxide film with hydrofluoric acid aqueous solution and suppress the surface recombination Then, the recombination lifetime was measured by the μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 1378 μsec.

  Next, a silicon substrate manufactured from the same silicon single crystal ingot was subjected to heat treatment at 1000 ° C. for 10 minutes in an oxygen atmosphere. After that, the oxide film formed on the surface of the heat-treated silicon substrate is removed with a hydrofluoric acid aqueous solution and subjected to chemical passivation treatment using an iodine ethanol solution to suppress surface recombination, and then the μ-PCD method. The recombination lifetime was measured by As a result, the in-plane average value of the recombination lifetime was 6727 μsec. Here, the standard value of the recombination lifetime after the heat treatment was set to 3000 μsec, and the silicon substrate for metal contamination evaluation was accepted.

  Next, in order to evaluate the metal contamination of the epitaxial growth furnace, a silicon substrate manufactured from the same silicon single crystal ingot was put into the epitaxial growth furnace, and an undoped epitaxial layer having a thickness of about 10 μm was grown. Then, chemical passivation treatment was performed using iodine ethanol solution, and recombination lifetime was measured by μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 4022 μsec.

(Comparative Example 1)
A silicon single crystal ingot having an n-type conductivity and a resistivity of 43 Ω · cm was grown by the floating zone method. The diameter is 200 mm, and the crystal axis orientation is <100>. And from the silicon single crystal ingot, a silicon substrate having a mirror polished finish was produced by a standard wafer processing process.

  Next, in order to investigate the value of the recombination lifetime before the heat treatment on the fabricated silicon substrate, the chemical passivation treatment using iodine ethanol solution to remove the natural oxide film with hydrofluoric acid aqueous solution and suppress the surface recombination Then, the recombination lifetime was measured by the μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 962 μsec.

  Next, a silicon substrate manufactured from the same silicon single crystal ingot was subjected to heat treatment at 1000 ° C. for 10 minutes in an oxygen atmosphere. After that, the oxide film formed on the surface of the heat-treated silicon substrate is removed with a hydrofluoric acid aqueous solution and subjected to chemical passivation treatment using an iodine ethanol solution to suppress surface recombination, and then the μ-PCD method. The recombination lifetime was measured by As a result, the in-plane average value of the recombination lifetime was 2729 μsec. Here, although the reference value of the recombination lifetime after the heat treatment was set to 3000 μsec and the silicon substrate for metal contamination evaluation was rejected, the following experiment was performed in order to compare the sensitivity to metal contamination with Example 1.

  In order to evaluate the metal contamination of the epitaxial growth furnace, a silicon substrate produced from the same silicon single crystal ingot was placed in the same epitaxial growth furnace as in Example 1, and an epitaxial layer having a thickness of about 10 μm was grown without doping. Thereafter, the recombination lifetime was measured in the same manner as in Example 1. As a result, the in-plane average value of the recombination lifetime was 1917 μsec.

  Compared with Example 1, it can be seen that the degree of decrease in recombination lifetime due to epitaxial growth (difference between after heat treatment and after epitaxial growth) is small.

(Example 2)
By the Czochralski method, a silicon single crystal ingot having an n-type conductivity, a resistivity of 63 Ω · cm, and an oxygen concentration of 10.7 ppma was grown. The diameter is 200 mm, and the crystal axis orientation is <100>. And from the silicon single crystal ingot, a silicon substrate having a mirror polished finish was produced by a standard wafer processing process.

  Next, in order to investigate the value of the recombination lifetime before the heat treatment on the fabricated silicon substrate, the chemical passivation treatment using iodine ethanol solution to remove the natural oxide film with hydrofluoric acid aqueous solution and suppress the surface recombination Then, the recombination lifetime was measured by the μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 3818 μsec.

  Next, a silicon substrate manufactured from the same silicon single crystal ingot was subjected to heat treatment in a hydrogen atmosphere at 1130 ° C. for 3 minutes. Thereafter, the natural oxide film was removed with an aqueous hydrofluoric acid solution, and after chemical passivation treatment using an iodine ethanol solution was performed to suppress surface recombination, the recombination lifetime was measured by the μ-PCD method. As a result, the in-plane average value of the recombination lifetime was 5275 μsec. Here, the standard value of the recombination lifetime after the heat treatment was set to 3000 μsec, and the silicon substrate for metal contamination evaluation was accepted.

  Next, in order to evaluate the metal contamination of the epitaxial growth furnace, a silicon substrate manufactured from the same silicon single crystal ingot was put into the epitaxial growth furnace, and an undoped epitaxial layer having a thickness of about 10 μm was grown. Then, chemical passivation treatment was performed using iodine ethanol solution, and recombination lifetime was measured by μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 4112 μsec.

(Example 3)
By the Czochralski method, a silicon single crystal ingot having an n-type conductivity, a resistivity of 50 Ω · cm, and an oxygen concentration of 5.9 ppma was grown. The diameter is 200 mm, and the crystal axis orientation is <100>. And from the silicon single crystal ingot, a silicon substrate having a mirror polished finish was produced by a standard wafer processing process.

  Next, in order to investigate the value of the recombination lifetime before the heat treatment on the fabricated silicon substrate, the chemical passivation treatment using iodine ethanol solution to remove the natural oxide film with hydrofluoric acid aqueous solution and suppress the surface recombination Then, the recombination lifetime was measured by the μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 4049 μsec.

  Next, a silicon substrate manufactured from the same silicon single crystal ingot was subjected to heat treatment in a hydrogen atmosphere at 1130 ° C. for 3 minutes. Thereafter, the natural oxide film was removed with an aqueous hydrofluoric acid solution, and after chemical passivation treatment using an iodine ethanol solution was performed to suppress surface recombination, the recombination lifetime was measured by the μ-PCD method. As a result, the in-plane average value of the recombination lifetime was 4834 μsec. Here, the standard value of the recombination lifetime after the heat treatment was set to 3000 μsec, and the silicon substrate for metal contamination evaluation was accepted.

  Next, in order to evaluate the metal contamination of the epitaxial growth furnace, a silicon substrate produced from the same silicon single crystal ingot is placed in the same epitaxial growth furnace as in Example 2, and an undoped epitaxial layer having a thickness of about 10 μm is grown. I let you. Then, chemical passivation treatment was performed using iodine ethanol solution, and recombination lifetime was measured by μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 3882 μsec.

(Comparative Example 2)
By the Czochralski method, a silicon single crystal ingot having an n-type conductivity, a resistivity of 5 Ω · cm, and an oxygen concentration of 12.6 ppma was grown. The diameter is 200 mm, and the crystal axis orientation is <100>. And from the silicon single crystal ingot, a silicon substrate having a mirror polished finish was produced by a standard wafer processing process.

  Next, in order to investigate the value of the recombination lifetime before the heat treatment on the fabricated silicon substrate, the chemical passivation treatment using iodine ethanol solution to remove the natural oxide film with hydrofluoric acid aqueous solution and suppress the surface recombination Then, the recombination lifetime was measured by the μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 2641 μsec.

  Next, a silicon substrate manufactured from the same silicon single crystal ingot was subjected to heat treatment in a hydrogen atmosphere at 1130 ° C. for 3 minutes. Thereafter, the natural oxide film was removed with an aqueous hydrofluoric acid solution, and after chemical passivation treatment using an iodine ethanol solution was performed to suppress surface recombination, the recombination lifetime was measured by the μ-PCD method. As a result, the in-plane average value of the recombination lifetime was 2748 μsec. Here, the reference value of the recombination lifetime after the heat treatment was set to 3000 μsec, and the silicon substrate for metal contamination evaluation was rejected. However, in order to compare the sensitivity to metal contamination with Example 2 and Example 3, the following experiment was performed. Went.

  In order to evaluate the metal contamination of the epitaxial growth furnace, a silicon substrate manufactured from the same silicon single crystal ingot is placed in the same epitaxial growth furnace as in the second and third embodiments, and an undoped epitaxial layer having a thickness of about 10 μm is formed. Grown up. Then, chemical passivation treatment was performed using iodine ethanol solution, and recombination lifetime was measured by μ-PCD method. The measurement was performed by mapping at intervals of 2 mm over the entire wafer surface. As a result, the in-plane average value of the recombination lifetime was 2269 μsec.

  Compared with Example 2 and Example 3, it can be seen that the degree of decrease in recombination lifetime due to epitaxial growth (difference between after heat treatment and after epitaxial growth) is small.

  From the results of the above examples and comparative examples, it was found that according to the present invention, a silicon substrate for metal contamination evaluation that can evaluate metal contamination of the heat treatment furnace with high sensitivity can be selected.

  The present invention is not limited to the above embodiment. The above-described embodiment is an exemplification, and the present invention has substantially the same configuration as the technical idea described in the claims of the present invention, and any device that exhibits the same function and effect is the present invention. It is included in the technical scope of the invention.

Claims (5)

A method for selecting a silicon substrate for metal contamination evaluation used when evaluating metal contamination in a heat treatment furnace using a measured value of recombination lifetime,
Preparing and preparing a silicon substrate that is a candidate for the silicon substrate for metal contamination evaluation from a silicon single crystal ingot;
A step of performing heat treatment for eliminating the grown-in recombination centers on the candidate silicon substrate with a heat treatment temperature of 1000 ° C. or higher and 1200 ° C. or lower and a heat treatment time of 1 minute or longer and 60 minutes or shorter ;
Performing a surface passivation treatment on the surface of the candidate silicon substrate;
Measuring the recombination lifetime of the silicon substrate after the heat treatment and surface passivation treatment by a microwave photoconductive decay method;
When the measurement value obtained by the measurement is equal to or greater than a predetermined reference value, the step of determining that the silicon substrate on which the measurement is performed is acceptable as a metal substrate for metal contamination evaluation;
Selecting a silicon substrate produced from the same silicon single crystal ingot as the silicon single crystal ingot from which the silicon substrate judged acceptable by the judgment is used as the metal contamination evaluation silicon substrate. Screening method of silicon substrate for metal contamination evaluation. The step of preparing the candidate silicon substrate includes:
Cutting the silicon substrate from the silicon single crystal ingot;
2. The method for selecting a silicon substrate for metal contamination evaluation according to claim 1, further comprising: removing a damage layer due to the cutting of the front surface and the back surface of the silicon substrate. 3. The selection of a silicon substrate for metal contamination evaluation according to claim 1 or 2 , wherein the surface passivation treatment is performed by chemical passivation treatment or by forming an oxide film on the surface of the silicon substrate. Method. The method for selecting a silicon substrate for metal contamination evaluation according to claim 3 , wherein the chemical passivation treatment is performed using an iodine alcohol solution. The method for selecting a silicon substrate for metal contamination evaluation according to any one of claims 1 to 4 , wherein the reference value is set to 3 msec or more.

Images