JP2003335599A - Process for identifying defect distribution in silicon single crystal ingot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily identify a boundary between regions [Pv] and [V] within an ingot wherein a high concentration of oxygen may be dissolved. <P>SOLUTION: The ingots drawn up from a silicon melt at various drawing speeds are sliced in axial directions to prepare first and second samples containing regions [V]-[I]. Oxygen concentrations in the samples are measured, and when the concentrations are ≥1.2×10<SP>18</SP>atoms/cm<SP>3</SP>, the first sample is heat-treated at 800°C for 4 hr in a nitrogen or oxidative atmosphere and subsequently at 1,000°C for 16 hr. Recombination lifetime in the entire first sample is measured to determine a boundary between regions [Pi] and [I]. Meanwhile, the second sample is heat-treated at 1,100-1,200°C for 1-4 hr in an oxidative atmosphere, selectively etched and observed under a light microscope to identify an oxidation induced stacking fault (OISF) region to determine the boundary between regions [V] and [Pv]. The regions [V], [Pv], [Pi] and [I] are defined in the specification, respectively. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of identifying a defect distribution of a silicon single crystal ingot (hereinafter referred to as an ingot) pulled by the Czochralski method (hereinafter referred to as a CZ method). More specifically, the concentration of oxygen dissolved in the ingot is 1.2 × 10 ¹⁸ atoms / c.
The present invention relates to a method for identifying a Grown-in defect-free region in an ingot having m ³ or more or 9.0 × 10 ¹⁷ atoms / cm ³ or less.

[0002]

2. Description of the Related Art Crystal originated particles (Crystal Originated Particl) are one of the factors that decrease the yield of devices due to the recent miniaturization of semiconductor integrated circuits.
e, hereinafter referred to as COP. ) And oxidation-induced stacking faults (Oxi
dation Induced Stacking Fault, hereafter referred to as OISF. The existence of microdefects of oxygen precipitates as nuclei) or the presence of interstitial dislocations (hereinafter referred to as L / D).

COP is a pit caused by crystals which appears on the surface of a wafer when SC-1 cleaning is performed on a mirror-polished silicon wafer with a mixed solution of ammonia and hydrogen peroxide. When this wafer is measured with a particle counter, this pit shows particles (Light Point Defect,
LPD). COP is an electrical characteristic, for example, a time-dependent dielectric breakdown characteristic of an oxide film.
ric Breakdown, TDDB), oxide film breakdown voltage characteristics (Time Ze
This may cause deterioration of ro Dielectric Breakdown, TZDB, etc. Further, if the COP exists on the wafer surface, a step may be generated in the device wiring process, which may cause disconnection. Then, the element isolation portion also causes a leak or the like, which lowers the product yield.

OISF is considered to have a core of minute oxygen precipitation formed during crystal growth, and is a stacking fault that is manifested in a thermal oxidation process or the like when manufacturing a semiconductor device. This OISF causes defects such as increasing the leak current of the device. L / D is also called a dislocation cluster, or is called a dislocation pit because an etching pit having an orientation is generated when a silicon wafer having this defect is immersed in a selective etching solution containing hydrofluoric acid as a main component. This L / D also causes deterioration of electrical characteristics such as leak characteristics and isolation characteristics.

From the above, from the silicon wafer used for manufacturing the semiconductor integrated circuit, COP, OI
There is a need to reduce SF and L / D.

A defect-free ingot having no COP, OISF and L / D and a silicon wafer sliced from this ingot are disclosed in Japanese Patent Laid-Open No. 11-1393. This defect-free ingot is composed of a perfect region [P], where [P] is a perfect region in which neither a vacancy-type point defect aggregate nor an interstitial silicon-type point defect aggregate is detected in the ingot. It is an ingot. The perfect region [P] has a region [V] in which a vacancy type point defect is dominant in the ingot and supersaturated vacancies are aggregated, and an interstitial silicon type point defect is dominant in a supersaturated lattice. The interstitial silicon is present between the silicon and the region [I] having defects.

Further, in Japanese Unexamined Patent Publication No. 2001-102385, a perfect region [P] having no defects in which point defects are aggregated is a region [Pv] where vacancy type point defects are predominant, and an interstitial silicon type point defect. Is classified into the region [Pi] where it is dominant. The region [Pv] is a region adjacent to the region [V] and having a vacancy type point defect concentration lower than the minimum vacancy type point defect concentration capable of forming OISF nuclei. The region [Pi] is a region adjacent to the region [I] and having an interstitial silicon type point defect concentration lower than the lowest interstitial silicon type point defect concentration capable of forming interstitial dislocations.

In the ingot composed of the perfect area [P], the pulling speed of the ingot is V (mm / min),
When the temperature gradient in the vertical direction of the ingot in the vicinity of the solid-liquid interface between the silicon melt and the silicon ingot is G (° C./mm), ring-shaped O is generated when the thermal oxidation treatment is performed.
ISF (P band) disappears at the center of the wafer, and L
V / G (mm ² /
Min./℃)

Conventionally, the following method has been adopted to measure the secondary defects caused by heat treatment in the ingot, that is, the distribution of aggregated defects in the axial direction and the radial direction of the ingot. First, an ingot is sliced in the axial direction to prepare a sample. Next, after mirror-etching this sample, it is heat-treated at 800 ° C. for 4 hours in a nitrogen or oxidizing atmosphere, and subsequently, heat-treated at 1000 ° C. for 16 hours. This heat-treated sample is treated with copper decoration (copp
erdecoration), secco-etching,
It is measured by methods such as X-ray topography analysis and lifetime measurement.
Generally, the density of oxygen precipitates formed in a sample by heat treatment is almost proportional to the oxygen concentration. The concentration of oxygen dissolved in the ingot is less than ^{1.2 × 10 18 atoms / cm 3} , 9.0 × 10 17 If it exceeds atoms / cm ^3, oxygen precipitates densely within the ingot by the heat treatment Therefore, according to the above method, the region [V], P band, region [P
v], region [Pi], B band and region [I] can be defined.

[0010]

However, the concentration of oxygen dissolved in the ingot is 1.2 × 10 ¹⁸ atoms / cm ³
If it is higher than the above, the density of oxygen precipitates appearing by the heat treatment is too high, and the area [V] is increased by the conventional method.
It was difficult to clearly measure the P band corresponding to the boundary between the region and the region [Pv]. Further, when the concentration of oxygen dissolved in the ingot is as low as 9.0 × 10 ¹⁷ atoms / cm ³ or less, the density of oxygen precipitates appearing by the heat treatment is not sufficiently high, and the area of the conventional method is P band and region [Pi] corresponding to the boundary between [V] and region [Pv]
It was difficult to clearly measure the B band corresponding to the boundary between the region and the region [I].

An object of the present invention is to easily detect the boundary between the region [Pv] and the region [V] in the ingot, even if the concentration of oxygen dissolved in the ingot is high. It is to provide an identification method. Another object of the present invention is to simplify the boundary between the region [V] and the region [Pv] and the boundary between the region [Pi] and the region [I] in the ingot, even if the concentration of oxygen dissolved in the ingot is low. Another object of the present invention is to provide a method for identifying the defect distribution of a silicon single crystal ingot.

[0012]

The invention according to claim 1 is
(a) A first slice containing a region [V], a region [Pv], a region [Pi] and a region [I], which is obtained by slicing a silicon single crystal ingot pulled from a silicon melt at different pulling speeds in an axial direction. And a step of producing the second sample, and (b) the first
And the step of measuring the oxygen concentration of the second sample, and (c) the oxygen concentration of the first and second samples is 1.2 × 10 respectively.
¹⁸ atoms / cm ³ (formerly ASTM, same below) 1st
The sample is subjected to a first heat treatment at 800 ° C. for 4 hours and then a second heat treatment at 1000 ° C. for 16 hours in a nitrogen or oxidizing atmosphere, and (d) the recombination lifetime of the whole heat-treated first sample is measured. Process and (e) above
(d) a step of defining a boundary between the region [Pi] and the region [I] in the first sample from the measurement result of the step, and (f)
The second sample is placed in an oxidizing atmosphere at 1100 to 1200 ° C.
And (g) the step of selectively etching the second sample subjected to the third heat treatment, and (h) the oxidation-induced stacking fault by observing the second sample subjected to the selective etching with an optical microscope. A step of identifying a region (OISF)
(i) A method of identifying a defect distribution of a silicon single crystal ingot, which includes the step of defining a boundary between the region [V] and the region [Pv] in the second sample, based on the observation result of the step (h). However, the region [V] is a region having a defect in which vacancy type point defects are predominant and excessive vacancies are aggregated, and the region [P]
v] is a region where vacancy-type point defects are predominant and does not have a defect in which vacancies are aggregated, and region [Pi] is predominant in interstitial silicon-type point defects and does not have a defect in which interstitial silicon is agglomerated. The region and the region [I] are regions in which interstitial silicon type point defects are dominant and defects in which interstitial silicon is aggregated are included.

According to the first aspect of the invention, the concentration of oxygen dissolved in the ingot is 1.2 × by the above method.
Even at a high concentration of 10 ¹⁸ atoms / cm ³ or more, the boundary between the region [Pv] and the region [V] in the ingot can be easily identified from the OISF distribution.

According to the second aspect of the present invention, (a) the silicon single crystal ingot pulled from the silicon melt at different pulling speeds is sliced in the axial direction, and the region [V], region [Pv], region [ Pi] and a region [I] containing first and second samples, (b) measuring the oxygen concentration of the first and second samples, and (c) oxygen of the first and second samples. Each concentration is 9.0 × 10 ¹⁷ atoms / cm
³ (former ASTM, the same applies to the following) In the following cases, the first sample is subjected to a third heat treatment in an oxidizing atmosphere at 1100 to 1200 ° C. for 1 to 4 hours, and (d) the third heat treated first sample is selectively etched. And (e) a step of observing the selectively etched first sample with an optical microscope to identify an oxidation-induced stacking fault region, and (f) an observation result of the above (e) step. [V] and area [P
v], the step of (g) selectively etching the second sample, and (h) the step of observing the selectively etched second sample with an optical microscope to identify the interstitial dislocation region. i) A method of identifying the defect distribution of a silicon single crystal ingot, which includes the step of defining the boundary between the region [Pi] and the region [I] in the second sample, based on the measurement result of the step (h). However, area [V], area [P
v], area [Pi] and area [I] have the same meanings as described in claim 1.

According to the third aspect of the invention, (a) the silicon single crystal ingot pulled from the silicon melt at different pulling speeds is sliced in the axial direction, and the region [V], region [Pv], region [ Pi] and a region [I] containing first and second samples, (b) measuring the oxygen concentration of the first and second samples, and (c) oxygen of the first and second samples. Each concentration is 9.0 × 10 ¹⁷ atoms / cm
³ (formerly ASTM, same as below) In the following cases, the first sample is subjected to a first heat treatment at 800 ° C. for 4 hours in a nitrogen or oxidizing atmosphere, and subsequently a second heat treatment is performed at 1000 ° C. for 16 hours, and (d) is heat treated. The step of measuring the recombination lifetime in the entire first sample, and (e) the measurement result of the step (d), the boundary between the area [Pi] and the area [I] and the area [V] in the first sample And (f) subjecting the second sample to a fourth heat treatment at 700 to 800 ° C. for 1 to 20 hours in a nitrogen or oxidizing atmosphere, and subsequently at 1000 ° C. for 1 to 20 hours. Fifth
The heat treatment step, (g) the step of measuring the recombination lifetime in the whole heat-treated second sample, and (h) the measurement result of the step (g) above, showing that the region [Pi] and the area of the second sample are A method of identifying a defect distribution of a silicon single crystal ingot, which includes the step of defining the boundary of [I] and the boundary of the area [V] and the area [Pv]. However, the area [V], the area [Pv], the area [Pi], and the area [I] have the same meanings as described in claim 1.

In the invention according to claim 2 or 3, by performing the above method, even if the concentration of oxygen dissolved in the ingot is as low as 9.0 × 10 ¹⁷ atoms / cm ³ or less, a region in the ingot is formed. The boundary between [V] and the area [Pv] and the boundary between the area [Pi] and the area [I] can be easily identified.

[0017]

BEST MODE FOR CARRYING OUT THE INVENTION (1) Ingot that is a target of the measuring method of the present invention The ingot that is a target of the first identifying method of the present invention is a region [V], a region [Pv], a region [Pi], and The V / G ratio is controlled to be raised so as to include the region [I]. The oxygen concentration of this ingot is 1.2 × 10 ¹⁸ atoms / cm ³
(Formerly ASTM, same below) or sliced sample of this ingot is heat treated at 800 ° C. for 4 hours under nitrogen or oxidizing atmosphere and then 1000
It is an ingot in which the boundary between the region [V] and the region [Pv] in this sample cannot be distinguished when the recombination lifetime is measured after heat treatment at 16 ° C. for 16 hours.

The ingot which is the target of the second identification method of the present invention is pulled up by controlling the V / G ratio so as to include the region [V], the region [Pv], the region [Pi] and the region [I]. To be The oxygen concentration of this ingot is 9.0 × 10.
Is it below ¹⁷ atoms / cm ³ (formerly ASTM, same below),
Alternatively, when a sample obtained by slicing this ingot was heat-treated at 800 ° C. for 4 hours in a nitrogen or oxidizing atmosphere, and then heat-treated at 1000 ° C. for 16 hours, the recombination lifetime was measured, the region [ It is an ingot in which the boundary between Pi] and area [I] cannot be identified.

Such an ingot is manufactured by the CZ method or MCZ.
Method, it is pulled from the silicon melt in the hot zone furnace based on Voronkov's theory. Generally, when a silicon single crystal ingot is pulled from a silicon melt in a hot zone furnace by the CZ method or the MCZ method, point defects and agglomerates of the point defects are agglomerated as defects in the silicon single crystal. : Three-dimensional defect) occurs. There are two general types of point defects: vacancy type point defects and interstitial silicon type point defects. A vacancy type point defect is a defect in which one silicon atom is separated from one of normal positions in the silicon crystal lattice. Such holes are aggregated to form a hole-type defect. On the other hand, lattice silicon aggregates to form interstitial silicon type defects.

Point defects are generally introduced at the interface between the silicon melt (molten silicon) and the ingot (solid silicon). However, by continuously pulling up the ingot, the portion that was the contact surface is pulled up and cooled. During cooling, vacancies or interstitial silicon undergo diffusion and pair annihilation reactions. Excessive point defects upon cooling to about 1100 ° C. are vacancy agglomerate aggregates.
s) or interstitial silicon-type defect aggregates (interstitial)
agglomerates). In other words, it is a three-dimensional structure in which excessive point defects form agglomerates.

Aggregates of vacancy-type defects include LSTD (Laser Scattering Tomograph Defects) in addition to COP described above.
Or, it contains defects called FPD (Flow Pattern Defects), and the agglomerates of interstitial silicon type defects are L / D described above.
Including defects called. FPD means Secco etching (HF: K ₂ Cr ₂ O ₇ (0.15 mol) for 30 minutes on a silicon wafer produced by slicing an ingot.
/ l) = 2: 1 mixed solution) is a source of traces that exhibit a unique flow pattern that appears when L
The STD is a source that has a refractive index different from that of silicon and generates scattered light when infrared rays are irradiated into the silicon single crystal.

According to Boronkov's theory, in order to grow a high-purity ingot with a small number of defects, the pulling rate of the ingot is V (mm / min), and the temperature gradient near the solid-liquid interface between the silicon melt and the silicon ingot is G. (C / mm), V / G (mm ² / min · ° C.) is controlled. According to this theory, as shown in FIG. 9, V / G is taken as a horizontal axis, and the vacancy type defect concentration and the interstitial silicon type defect concentration are taken as the same vertical axis, and the relationship between V / G and point defect concentration is shown. It is represented graphically to explain that the boundary between the vacancy region and the interstitial silicon region is determined by V / G. More specifically, when the V / G ratio is above the critical point (V / G) _c , an ingot with an increased vacancy concentration is formed, while when the V / G ratio is below the critical point (V / G) _c , interstitial silicon is formed. An ingot with increased concentration is formed. In FIG. 9, [I] indicates a region ((V / G) ₁ or less) where interstitial silicon is dominant and interstitial silicon is agglomerated, and [V]
Indicates a region ((V / G) ₂ or more) in which vacancies are predominant and agglomeration of vacancies is present, and [P] indicates agglomerates of vacancies and interstitial silicon defects. The non-existent perfect region ((V / G) ₁ to (V / G) ₂ ) is shown. OISF is provided at the boundary with the area [V] adjacent to the area [P].
There is a P band region ((V / G) ₂ to (V / G) ₃ ) that forms nuclei. In the P band region, there are fine plate-like precipitates,
OISF (stacking fault) is formed by heat treatment in an oxidizing atmosphere. Further, a B band region exists at the boundary of the region [I] adjacent to the region [P]. In this B band region, fine dislocation clusters are present at a high density.

This perfect area [P] is further classified into an area [Pi] and an area [Pv]. [Pi] is V / G
The ratio is a region from (V / G) ₁ to the critical point, and is a region where interstitial silicon is predominant and does not have aggregated defects. [Pv] is the above (V / G) ₂ from the critical point of V / G ratio.
Is a region up to and where voids are predominant and does not have agglomerated defects. That is, [Pi] is a region adjacent to the region [I] and having an interstitial silicon concentration lower than the lowest interstitial silicon concentration capable of forming interstitial dislocations,
[Pv] is a region adjacent to the region [V] and having a vacancy concentration lower than the minimum vacancy concentration capable of forming OISF nuclei. Therefore, since the ingot which is the target of the first or second identification method of the present invention includes the region [V], the region [Pv], the region [Pi], and the region [I], the V / G ratio is It will be controlled and pulled so as to include the respective regions from (V / G) ₁ or less to the critical point and (V / G) ₃ or more. (2) First Identification Method of the Present Invention Next, the first identification method of the present invention will be described based on FIGS. 1 and 4. First, as shown in FIG. 1, first and second measurement samples are prepared from a test ingot. That is, the ingot is pulled up from the silicon melt stored in the quartz crucible of the pulling device based on the CZ method or the MCZ method. At this time, the pulling speed V (mm / min) of the ingot is changed from the high speed (top side) so that the above-mentioned region [V], region [Pv], region [Pi] and region [I] are included in the axial direction of the ingot. Change the speed from low speed (bottom side) or low speed (bottom side) to high speed (top side) and pull up. Then, as shown in FIG. 4, the ingot obtained in the test is sliced in the axial direction and mirror-etched to prepare a measurement sample having a mirror-finished surface having a thickness of 500 to 2000 μm.

Next, returning to FIG. 1, the concentration of oxygen dissolved in the first and second samples is determined by FT-IR (Fourier transform).
Infrared absorption spectroscopy) method. The oxygen concentration of the first and second samples is 1.2 × 10 ¹⁸
In the case of atoms / cm ³ or more, the first sample is first heat-treated at 800 ° C. for 4 hours in a nitrogen or oxidizing atmosphere, and subsequently, second heat-treatment is performed at 1000 ° C. for 16 hours. When the concentration of oxygen dissolved in the first and second samples is out of the above range, the following identification method is performed. Oxygen concentration is 9.0 x 1
When it is 0 ¹⁷ atoms / cm ³ or less, the second identification method or the third identification method described later is used. Oxygen concentration is 1.2 × 10
When it is within the range of less than ¹⁸ atoms / cm ^{3 and more} than 9.0 × 10 ¹⁷ atoms / cm ³ , as shown in FIG. 7, after performing the first and second heat treatments which are conventionally performed, , By performing recombination lifetime measurement, as shown in FIG.
The boundary between the area [V] and the area [Pv] and the boundary between the area [Pi] and the area [I] can be clearly discriminated from each other.

For the sample having an oxygen concentration of 1.2 × 10 ¹⁸ atoms / cm ³ or more, even if the recombination lifetime of the sample is measured after the first and second heat treatments, the recombination lifetime of the sample is From the measurement result, the boundary between the region [V] and the region [Pv] cannot be clearly identified. This is because in the sample having an oxygen concentration of 1.2 × 10 ¹⁸ atoms / cm ³ or more, oxygen precipitation nuclei are easily formed, so that the boundary between the region [V] and the region [Pv] becomes hard to be recognized.

Therefore, in the identification method of the present invention, the first and second heat treatments are performed on the first sample, and the recombination lifetime is measured on the first sample.
As shown in (a), the boundary between the area [Pi] and the area [I] is defined. Then, the second sample was subjected to 1 in an oxidizing atmosphere.
A third heat treatment is performed at 100 to 1200 ° C. for 1 to 4 hours. OISF becomes apparent by performing the third heat treatment. The second sample, which has been subjected to the third heat treatment, is subjected to selective etching without stirring, and by observing this selective etched portion with an optical microscope, as shown in FIG. The boundary between [V] and the region [Pv] can be easily identified by an optical microscope.
Examples of the selective etching include seco etching and non-chrome etching.

The area [Pi] and the area [I] of FIG.
Of the inflection points indicated by the X mark on the boundary, the most in the area [V]
Pulling speed V from the near inflection point position₂Is defined in FIG.
The cross mark at the boundary position between the area [V] and the area [Pv] in (b)
Of the inflection points shown, the inflection point position closest to the area [I]
Pulling speed V from installation₁Stipulate. This pulling speed V₁~
V ₂If the ingot is pulled at a pulling speed within the range of
No pore aggregates and interstitial silicon aggregates are present.
-Fact area [P], that is, Grown-in
Can make a got.

(3) Second Identification Method of the Present Invention As shown in FIG. 2, first, first and second measurement samples are prepared from a test ingot. That is, the ingot is pulled up from the silicon melt stored in the quartz crucible of the pulling device based on the CZ method or the MCZ method. At this time, the pulling speed V (mm / min) of the ingot is changed from the high speed (top side) so that the above-mentioned region [V], region [Pv], region [Pi] and region [I] are included in the axial direction of the ingot. Change the speed from low speed (bottom side) or low speed (bottom side) to high speed (top side) and pull up. Then, as shown in FIG. 4, the obtained test ingot is sliced in the axial direction and mirror-etched to obtain 500-2.
A measurement sample with a mirror surface is made with a thickness of 000 μm.

Next, returning to FIG. 2, the concentration of oxygen dissolved in the first and second samples is determined by FT-IR (Fourier transform).
Infrared absorption spectroscopy) method. When the oxygen concentration is 9.0 × 10 ¹⁷ atoms / cm ³ or less, the first sample is 1100 to 1 in an oxidizing atmosphere.
A third heat treatment is performed at 200 ° C. for 1 to 4 hours. First and second
When the concentration of oxygen dissolved in the sample is out of the above range, the following identification method is performed. Oxygen concentration is 1.2 × 10 ¹⁸ at
When it is oms / cm ³ or more, the first identification method is performed. Oxygen concentration is less than 1.2 × 10 ¹⁸ atoms / cm ³ and 9.0 ×
When it is within the range of more than 10 ¹⁷ atoms / cm ³ , the conventional identification method described above is performed. This oxygen concentration is 9.0 × 10 ¹⁷ at
For the samples having an oms / cm ³ or less, even if the recombination lifetime of the sample is measured after the first and second heat treatments, the region [Pi] and the region [Pi] I] cannot be clearly distinguished from the boundary. This has an oxygen concentration of 9.0 × 10 ¹⁷ atoms / cm ³
This is because it is difficult to form oxygen precipitate nuclei in the following samples only by performing the first and second heat treatments, and thus the boundary between the region [Pi] and the region [I] becomes difficult to understand.

By applying this third heat treatment, the OIS
F becomes actual. The first sample, which has been subjected to the third heat treatment, is subjected to selective etching, and by observing the selectively etched portion with an optical microscope, as shown in FIG. 6A, a region [V] is generated due to the manifested OISF distribution. The boundary between the area and the area [Pv] can be easily identified by an optical microscope.

Next, by selectively etching the second sample and observing the selectively etched portion with an optical microscope, the B band region showing L / D can be identified as shown in FIG. 6B. .

Of the inflection points indicated by x at the boundary between the area [V] and the area [Pv] in FIG. 6A, the pulling speed V ₁ is defined from the inflection point position closest to the area [I]. Then, Fig. 6
Among the inflection points indicated by crosses on the boundary between the region [Pi] and the region [I] in (b), the pulling speed V ₂ is defined from the inflection point position closest to the region [V]. This pulling speed V _{1 to} V ₂
When the ingot is pulled up at a pulling speed within the range, a perfect region [P] in which no agglomerates of vacancies and agglomerates of interstitial silicon are present, that is, a Grown-in defect-free region ingot can be produced.

(4) Third Identification Method of the Present Invention This third identification method is a method used at the same oxygen concentration as the second identification method. As shown in FIG.
First, first and second measurement samples are prepared from a test ingot. That is, the ingot is pulled up from the silicon melt stored in the quartz crucible of the pulling device based on the CZ method or the MCZ method. At this time, the pulling speed V (mm / min) of the ingot is changed from the high speed (top side) so that the above-mentioned region [V], region [Pv], region [Pi] and region [I] are included in the axial direction of the ingot. Change the speed from low speed (bottom side) or low speed (bottom side) to high speed (top side) and pull up. Then, as shown in FIG. 4, the obtained test ingot is axially sliced and mirror-etched to prepare a measurement sample having a mirror-finished surface having a thickness of 500 to 2000 μm.

Next, returning to FIG. 3, the concentration of oxygen dissolved in the first and second samples is determined by FT-IR (Fourier transform).
Infrared absorption spectroscopy) method. When the oxygen concentration is 9.0 × 10 ¹⁷ atoms / cm ³ or less, the first sample has a concentration of 80 under nitrogen or an oxidizing atmosphere.
First heat treatment at 0 ° C for 4 hours, then at 1000 ° C for 1 hour
The second heat treatment is performed for 6 hours. By performing recombination lifetime measurement on the first sample, the area [V] and the area [Pv]
And the boundary between the area [Pi] and the area [I]. When the boundary between the region [Pi] and the region [I] cannot be sufficiently discriminated, the second sample is placed under nitrogen or an oxidizing atmosphere,
The fourth heat treatment is performed at 700 to 800 ° C. for 1 to 20 hours, and then the fifth heat treatment is performed at 1000 ° C. for 1 to 20 hours. When the oxygen concentration is low and the oxygen precipitation nuclei cannot grow at 800 ° C., the temperature is lowered to 700 ° C.
By performing the fourth and fifth heat treatments in which the heat treatment conditions are expanded to ˜800 ° C., the boundary between the region [Pi] and the region [I] can be easily distinguished. By performing the recombination lifetime measurement again on the second sample that has undergone the fifth heat treatment, the boundary between the region [Pi] and the region [I] can be identified because the heat treatment condition has been extended to a low temperature.

[0035]

As described above, according to the method of identifying the point defect distribution of the silicon single crystal ingot of the present invention,
Even if the concentration of oxygen dissolved in the ingot is high, the boundary between the region [Pv] and the region [V] in the ingot can be easily identified. Further, even if the concentration is low, the boundary between the region [V] and the region [Pv] and the boundary between the region [Pi] and the region [I] in the ingot can be easily identified.
As a result, a silicon single crystal ingot having the perfect region [P] can be easily manufactured, and a silicon wafer having the perfect region [P] can be easily obtained from this ingot.

[Brief description of drawings]

FIG. 1 is a flowchart relating to a measurement sample in the first identification method of the present invention.

FIG. 2 is a flowchart regarding a measurement sample in the second identification method of the present invention.

FIG. 3 is a flowchart regarding a measurement sample in the third identification method of the present invention.

FIG. 4 is a diagram showing a situation in which a measurement sample is produced from a test ingot.

FIG. 5 (a) is a diagram showing a recombination lifetime of the entire first sample after the second heat treatment in the first identification method. (b) The figure which shows the recombination lifetime of the whole 2nd sample after performing the 3rd heat processing in a 1st identification method.

FIG. 6 (a) is a diagram showing a recombination lifetime of the entire first sample after the third heat treatment in the second identification method. (b) L / D of the entire second sample in the second identification method
The figure which shows distribution.

FIG. 7 is a flowchart regarding a measurement sample in a conventional identification method.

FIG. 8 is a diagram showing a recombination lifetime of the entire sample when a conventional identification method is applied.

FIG. 9 is a graph showing V / G and point defect concentration based on the Voronkov theory, with V / G as a horizontal axis and the vacancy type point defect concentration and the interstitial silicon type point defect concentration as the same vertical axis. The figure which shows a relationship.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Kazuhiro Harada             1-2-1 Shibaura, Minato-ku, Tokyo Sumitomo Mitsubishi             Inside Silicon Co., Ltd. F-term (reference) 4G077 AA02 BA04 CF10 GA01 GA06                       HA12

Claims (3)

[Claims] 1. (a) A silicon single crystal ingot pulled up from a silicon melt at different pulling speeds is axially sliced to obtain a region [V], a region [Pv], a region [Pi] and a region [I]. ] Of the first and second samples are included, (b) the step of measuring the oxygen concentration of the first and second samples, and (c) the oxygen concentration of the first and second samples is 1 respectively. 2 × 10 ¹⁸ atoms / cm ³ (former ASTM, the same applies hereinafter), the first sample was first heat-treated at 800 ° C. for 4 hours in a nitrogen or oxidizing atmosphere, and then 1
A second heat treatment for 16 hours at 000 ° C .; (d) a step of measuring the recombination lifetime of the whole heat-treated first sample; and (e) a measurement result of the step (d). Defining a boundary between the region [Pi] and the region [I] in the sample, and (f) subjecting the second sample to a third heat treatment at 1100 to 1200 ° C. for 1 to 4 hours in an oxidizing atmosphere, ( g) The third heat-treated second
A step of selectively etching the sample; (h) a step of observing the selectively etched second sample with an optical microscope to identify an oxidation-induced stacking fault region (OISF); and (i) an observation result of the step (h). And a step of defining a boundary between the region [V] and the region [Pv] in the second sample, the method of identifying a defect distribution of a silicon single crystal ingot. However, the region [V] is a region having defects in which vacancy type point defects are predominant and excessive vacancies are aggregated, and the region [Pv] is in which vacancy type point defects are predominant and vacancies are agglomerated. In the region [Pi] which does not have interstitial silicon type point defects, the interstitial silicon type point defects are predominant, and in the region [I] which does not have defects in which interstitial silicon is agglomerated, the interstitial silicon type point defects are predominant This is a region having a defect in which silicon is aggregated. 2. (a) A silicon single crystal ingot pulled up from a silicon melt at different pulling speeds is axially sliced to obtain a region [V], a region [Pv], a region [Pi] and a region [I. ], And (b) measuring the oxygen concentration of the first and second samples, and (c) the oxygen concentration of the first and second samples is 9 respectively. 0.0 × 10 ¹⁷ atoms / cm ³ (former ASTM, the same applies hereinafter) or less, subjecting the first sample to a third heat treatment at 1100 to 1200 ° C. for 1 to 4 hours in an oxidizing atmosphere; 3. Selectively etching the heat-treated first sample; (e) observing the selectively etched first sample with an optical microscope to identify an oxidation-induced stacking fault region (OISF); From the observation result of step e), Area [V]
And (g) selectively etching the second sample, and (h) observing the selectively etched second sample with an optical microscope to find an interstitial dislocation region (L). / D), and (i) the above
From the measurement result of the step (h), in the second sample,
And a step of defining a boundary between the region [Pi] and the region [I].
However, the area [V], the area [Pv], the area [Pi], and the area [I] have the same meanings as described in claim 1. 3. (a) A silicon single crystal ingot pulled from a silicon melt at different pulling speeds is sliced in the axial direction, and the region [V], the region [Pv], the region [Pi] and the region [I] are sliced. ], And (b) measuring the oxygen concentration of the first and second samples, and (c) the oxygen concentration of the first and second samples is 9 respectively. When it is below 0.0 × 10 ¹⁷ atoms / cm ³ (former ASTM, the same applies hereinafter), the first sample is first heat-treated at 800 ° C. for 4 hours in a nitrogen or oxidizing atmosphere, and then 1
A second heat treatment for 16 hours at 000 ° C .; (d) a step of measuring the recombination lifetime of the whole heat-treated first sample; and (e) a measurement result of the step (d). Defining a boundary between the region [Pi] and the region [I] and a boundary between the region [V] and the region [Pv] in the sample, and (f) the second sample in a nitrogen or oxidizing atmosphere at 700- 4th heat treatment at 800 ° C for 1 to 20 hours, then 5th at 1000 ° C for 1 to 20 hours
Heat-treating, (g) measuring the recombination lifetime of the whole heat-treated second sample,
(h) a step of defining a boundary between the region [Pi] and the region [I] and a boundary between the region [V] and the region [Pv] in the second sample from the measurement result of the step (g). Identification method of defect distribution of silicon single crystal ingot. However, the area [V], the area [Pv], the area [Pi], and the area [I] have the same meanings as described in claim 1.