JP2002047094A - Silicon single crystal pulling-up method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an ingot having no point defect agglomerates and exhibiting intrinsic gettering(IG) effect when formed into a wafer. SOLUTION: When a region of less than the lowest interstitial silicon concentration capable of forming an interstitial-type large dislocation, which is adjacent to a region [I] having interstitial silicon point defects dominantly therein and also belongs to a perfect region [P] having no point defect agglomerate, is defined as [PI], and further, a region which is adjacent to a region [V] having vacancy point defects dominantly therein, belongs to the above region [P] and is less than a vacancy concentration capable of forming COP(crystal originated particles) or FPD(flow pattern defects), is defined as [PV], a silicon single crystal ingot which comprises both regions [PI] and [PV] and of which an oxygen concentration is 1.4×1018 atoms/cm3 or more (by the former ASTM) is heated and kept at a temperature of 600-500 deg.C for 2-50 hr at the point of time when the temperature of the silicon single crystal ingot is lowered to a prescribed level within the rang of 1,000-600 deg.C due to the pulling-up to obtain the objective ingot.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for producing a silicon single crystal free of point defect aggregates by the Czochralski method (hereinafter referred to as CZ method). More specifically, intrinsic gettering (hereinafter I
G) A method for pulling a silicon single crystal having a source.

[0002]

2. Description of the Related Art In recent years, in the process of manufacturing a semiconductor integrated circuit, an oxidation-induced stacking fault (hereinafter referred to as OSF) is a cause of lowering the yield.
That. ) Nuclei of oxygen precipitates and microcrystalline particles (Crystal Originated Particl
e, hereinafter referred to as COP. ) Or interstitial dislocations (Int
erstitial-type Large Dislocation, hereinafter referred to as LD. ). OSF introduces minute defects serving as nuclei during crystal growth, becomes apparent in a thermal oxidation step or the like when manufacturing a semiconductor device, and causes defects such as an increase in leak current of the manufactured device. Also COP
Are pits caused by crystals that appear on the wafer surface when the mirror-polished silicon wafer is washed with a mixed solution of ammonia and hydrogen peroxide. When this wafer is measured with a particle counter, these pits are also detected as light scattering defects together with the original particles. This C
OP is an electrical characteristic, for example, a time-dependent dielectric breakdown characteristic (Time Dependent dielectric Breakdown, TDD) of an oxide film.
B), oxide film breakdown voltage characteristics (Time Zero Dielectric Breakdo
wn, TZDB) and the like. Also COP
Is present on the wafer surface, a step is generated in a device wiring process, which may cause disconnection. This also causes a leak and the like in the element isolation portion, and lowers the product yield. Furthermore, LD is also called a dislocation cluster,
Alternatively, when a silicon wafer having this defect is immersed in a selective etching solution containing hydrofluoric acid as a main component, a pit is generated, and thus the silicon wafer is also called a dislocation pit. This LD also causes deterioration of electrical characteristics such as leak characteristics and isolation characteristics.

[0003] From the above, OSF, CO, etc. can be obtained from a silicon wafer used for manufacturing a semiconductor integrated circuit.
There is a need to reduce P and LD. A method for manufacturing a defect-free silicon ingot having no OSF, COP and LD is disclosed in Japanese Patent Application Laid-Open No. H11-1393. This defect-free silicon ingot is composed of perfect regions [P] in which no aggregates of vacancy type point defects and no aggregates of interstitial silicon type point defects exist in the ingot. The perfect region [P] is a region [I] where interstitial silicon type point defects are predominantly present,
It intervenes between the region [V] where the vacancy type point defect is predominantly present in the silicon single crystal ingot. In the silicon wafer composed of the perfect region [P], the pulling speed of the ingot is V (mm / min), and the temperature gradient in the vertical direction of the ingot near the interface between the silicon melt and the ingot is G (° C./mm). At this time, it is made by determining the value of V / G (mm ² / min · ° C.) so that the OSF generated in a ring shape during the thermal oxidation treatment disappears at the center of the wafer.

[0004]

On the other hand, some semiconductor device manufacturers require a silicon wafer that does not have an OSF, a COP, and an LD, and that has a capability of gettering metal contamination generated in a device process. .
If the wafer does not have sufficient gettering ability, contamination with metal in the device process causes junction leakage, device operation failure due to trap levels due to metal impurities, and the like, thereby lowering product yield. Although the silicon wafer cut from the ingot composed of the perfect region [P] does not have the OSF, the COP, and the LD, in the heat treatment in the device process, oxygen precipitation does not always occur uniformly in the wafer surface,
As a result, the IG effect may not be sufficiently obtained.

[0005] An object of the present invention is to provide an IG effect when formed into a wafer even for an ingot consisting of both the region [P _V ] and the region [P _I ] or an ingot consisting only of the region [P _I ]. Another object of the present invention is to provide a method of pulling a silicon single crystal free of point defect aggregates.

[0006]

The invention according to claim 1 is
The region where interstitial silicon type point defects predominantly exist in a silicon single crystal ingot is [I], the region where vacancy type point defects predominantly exist is [V], and the interstitial silicon type point defects are This is an improvement in the method of pulling a silicon single crystal ingot consisting of the perfect region [P] when the perfect region where no aggregate of the above and the aggregate of vacancy type point defects are present is [P]. The characteristic configuration is such that when a region adjacent to the above-mentioned region [I] and belonging to the above-mentioned perfect region [P] and having a lower interstitial silicon concentration that can form an interstitial dislocation is less than [P _I ], [P _V ] and the region [P _I ], and the oxygen concentration is 1.4 × 10 ¹⁸
atoms / cm ³ (old ASTM) or more, and the temperature of the silicon single crystal ingot is 1000 to 1000
When the temperature falls to a predetermined temperature within a temperature range of 600 ° C., the silicon single crystal ingot is heated to raise the temperature to 600 to
Maintaining at 500 ° C. for 2 to 50 hours. Claim 2
The characteristic feature of the invention according to the invention is that the region [P _V ] and the region [P _I ] are both included, and the oxygen concentration is 1.1 to 1.1.
When the temperature of the silicon single crystal ingot is equal to or higher than 2 × 10 ¹⁸ atoms / cm ³ (former ASTM) and the temperature of the silicon single crystal ingot falls to a predetermined temperature within a temperature range of 1000 to 600 ° C., the silicon single crystal ingot is heated. The temperature is maintained at 600 to 500 ° C. for 20 to 50 hours.

A characteristic feature of the invention according to claim 3 is that it comprises only the region [P _I ] and has an oxygen concentration of 1.4 × 1.
0 ¹⁸ atoms / cm ³ (old ASTM) or more
Pulling raises the temperature of silicon single crystal ingot to 100
When the temperature falls to a predetermined temperature within a temperature range of 0 to 600 ° C., the silicon single crystal ingot is heated to raise the temperature to 60 ° C.
It is to maintain at 0 ° C for 20 to 50 hours or at 500 ° C for 6 to 50 hours. A characteristic feature of the invention according to claim 4 is that it has only the region [P _I ] and has an oxygen concentration of 1.0.
1-1.2 × 10 ¹⁸ atoms / cm ³ (former ASTM)
As described above, when the temperature of the silicon single crystal ingot drops to a predetermined temperature within the temperature range of 1000 to 600 ° C. by pulling, the silicon single crystal ingot is heated and maintained at 500 ° C. for 50 hours. is there.

In the inventions according to the first to fourth aspects, when the ingot is composed of both the region [P _V ] and the region [P _I ], or when the ingot is composed of only the region [P _I ], the oxygen concentration of the ingot becomes a predetermined value. When the ingot being pulled is heated under the above conditions in the above concentration range, oxygen precipitation nuclei having a predetermined density or more appear in the state of the ingot. When a silicon wafer is cut out from the heat-treated ingot and the wafer is heat-treated in a device manufacturing process of a semiconductor device maker, the oxygen precipitate nuclei are converted to oxygen precipitates (Bulk MicroDefect,
Hereinafter, it is called BMD. ), And even if the wafer is composed of both the region [P _V ] and the region [P _I ] or a wafer composed only of the region [P _I ], the entire surface of the wafer has an IG effect.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION The silicon single crystal ingot of the present invention is manufactured by pulling a silicon melt in a hot zone furnace from a silicon melt in a predetermined velocity profile based on the Voronkov theory by the CZ method. Generally, when a silicon single crystal ingot is pulled up from a silicon melt in a hot zone furnace by the CZ method, point defects and agglomerates: Defects). Point defects have two general forms: vacancy type point defects and interstitial silicon type point defects. A vacancy-type point defect is one in which one silicon atom has separated from one of the normal positions in the silicon crystal lattice. Such holes become hole type point defects. On the other hand, if an atom is found at a position (interstitial site) other than the lattice point of the silicon crystal, this becomes an interstitial silicon point defect.

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 starts to cool down with pulling up. During cooling, vacancy-type point defects or interstitial silicon-type point defects merge with each other by diffusion to form vacancy agglomerates or interstitial agglomerates. Is formed. In other words, the aggregate is a three-dimensional structure generated due to the merging of point defects. Aggregates of vacancy-type point defects are LSTDs (Laser Scattering Tomograms) in addition to the COPs described above.
An agglomerate of interstitial silicon-type point defects includes a defect called LD, which includes a defect called aph defects or FPD (Flow Pattern Defects). FPD stands for Secco etching (Sico etching, silicon wafer) prepared by slicing an ingot without stirring for 30 minutes.
(K ₂ Cr ₂ O ₇ : 50% HF: pure water = 44 g: 2000 cc: 1000 cc)
LSTD is a source of traces exhibiting a unique flow pattern that appears when etching is performed using a mixture of
This is a source that has a refractive index different from that of silicon and generates scattered light when a silicon single crystal is irradiated with infrared rays.

[0011] Boronkov's theory states that the number of defects is small and high.
Raising the ingot to grow a pure ingot
V (mm / min), the ingot and silicon melt
The temperature gradient in the ingot near the interface is G (° C / mm)
V / G (mm^Two/ Min / ℃)
It is. In this theory, as shown in FIG.
On the horizontal axis, vacancy-type point defect concentration and interstitial silicon-type point
V / G and point defect density are plotted on the same vertical axis for defect density.
The relationship between vacancy region and interstitial silicon
Explain that the boundaries of the area are determined by V / G
I have. More specifically, when the V / G ratio is above the critical point,
While an ingot having a predominant point defect density is formed, V /
When the G ratio is below the critical point, the interstitial silicon-type point defect concentration is excellent.
A vigorous ingot is formed. In FIG. 1A, [I]
Is mainly dominated by interstitial silicon type point defects.
Region where point defects of recon type exist ((V / G)₁Below)
In [V], vacancy-type point defects in the ingot are dominant.
In the region where the aggregate of vacancy type point defects exists ((V /
G)_Two[P] is an aggregate of vacancy type point defects and
Aggregates free of interstitial silicon-type point defects
Area ((V / G) ₁~ (V / G)_Two). Area [P]
The region [V] adjacent to the region [V]
SF] ((V / G)_Two~ (V / G)_Three) Exists.

The perfect area [P] is further classified into an area [P _I ] and an area [P _V ]. [P _I ] is V / G
The ratio is from (V / G) ₁ to the critical point,
[P _V ] is a region where the V / G ratio is from the critical point to the above (V / G) ₂ . That is, [P _I ] 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 an interstitial dislocation, and [P _V] ] Is a region adjacent to the region [V] and having a vacancy-type point defect concentration lower than the lowest vacancy-type point defect concentration capable of forming an OSF.

A predetermined pulling speed in this embodiment
The profile is such that the ingot is
Pulling up due to temperature gradient when pulled out of concrete
Aggregation of interstitial silicon-type point defects when the ratio of polishing speed (V / G)
Prevent body outbreak (V / G) ₁The above is the void type point
The vacancy-type point defect at the center of the ingot
Restrict to the dominant region (V / G)_TwoBelow
It is decided to be held.

The pulling speed profile is determined by simulating the reference ingot in the axial direction experimentally or by combining these techniques, based on the above-mentioned Boronkov theory by simulation. That is, this determination is made by slicing the ingot sliced in the axial direction in the transverse direction after the simulation, confirming it in a wafer state, and repeating the simulation. For the simulation, a plurality of kinds of pulling speeds are determined within a predetermined range, and a plurality of reference ingots are grown. That is, as shown in FIG. 1E, sectional views of the ingot when the pulling speed is gradually reduced from 1.2 mm / min to 0.4 mm / min to continuously reduce V / G are shown in FIGS. 1B, 1C and 1C. Each is shown in FIG. 1D.
The horizontal axis in each figure is drawn corresponding to the horizontal axis (V / G) in FIG. 1A. FIG. 1B is a conceptual diagram by an X-ray topograph after heat-treating the ingot in an N ₂ atmosphere at 1000 ° C. for 40 hours. In this figure, the region [V], the region [OSF], the region [P _V ], the region [P _I ], and the region [I] appear as the pulling speed decreases. FIG. 1C
Is three ingots immediately after being pulled (as-grown)
It is a defect distribution figure of the crystal at the time of seco etching for 0 minutes. In this figure, COP, F
PD appears, and LD appears in a region corresponding to the above [I]. FIG. 1D further shows that the ingot was placed under a wet O ₂ atmosphere.
FIG. 4 is a crystal defect distribution diagram when heat-treated at 1100 ° C. for 1 hour and then subjected to secco etching for 2 minutes. In this figure, OS
F appears. The V / G in this embodiment is set so that the ingot consists of both the region [P _V ] and the region [P _I ] or only the region [P _I ].

Incidentally, the aggregates of point defects such as COP and LD may have different values of detection sensitivity and detection lower limit depending on the detection method. Therefore, in the present specification, "there is no aggregate of point defects" means the product of the observation area and the etching allowance by an optical microscope after subjecting a mirror-finished silicon single crystal to non-stirring seco etching. When observed as an inspection volume, one agglomerate of flow pattern (vacancy type defect) and dislocation cluster (interstitial silicon type point defect) has one defect per 1 × 10 ⁻³ cm ³ of inspection volume. The lower limit of detection (1 × 10 ³ / cm ³ )
Means that the number of point defect aggregates is equal to or less than the lower limit of detection.

FIGS. 2 to 4 show a pulling apparatus for pulling a silicon single crystal ingot at the above-mentioned predetermined V / G. As shown, a silicon single crystal pulling apparatus 10
Has a chamber 11. In this chamber 11, a quartz crucible 13 for storing a silicon melt 12 is provided,
The outer surface of the quartz crucible 13 is covered with a graphite susceptor 14. The lower surface of the quartz crucible 13 is the graphite susceptor 1
The lower end of the support shaft 16 is fixed to a crucible driving means 17 via the support shaft 4 (FIG. 2). The crucible driving means 17 has a first rotation motor (not shown) for rotating the quartz crucible 13 and a lifting motor for raising and lowering the quartz crucible 13, and these motors can rotate the quartz crucible 13 in a predetermined direction. At the same time, it can be moved up and down. The outer peripheral surface of the quartz crucible 13 is surrounded by a first heater 18 at a predetermined interval from the quartz crucible 13, and the first heater 18 is surrounded by a heat retaining cylinder 19. The first heater 18 heats and melts the high-purity polycrystalline silicon charged into the quartz crucible 13 to form a silicon melt.

A cylindrical casing 21 is connected to the upper end of the chamber 11. The casing 21 is provided with a pulling means 22. The pulling means 22 includes a pulling head (not shown) rotatably provided at the upper end of the casing 21 in a horizontal state, a second rotation motor (not shown) for rotating the head, and a quartz crucible 13 from the head.
Wire cable 22a hanging down toward the center of rotation
And a pulling motor (not shown) provided in the head for winding or feeding the wire cable 22a. A silicon melt 12 is provided at the lower end of the wire cable 22a.
A seed crystal 23 for pulling up the silicon single crystal ingot 24 by immersion in the substrate is attached.

The ingot 24 is located between the outer peripheral surface of the silicon single crystal ingot 24 and the inner peripheral surface of the quartz crucible 13.
2 is provided with a heat shielding member 26 surrounding the outer peripheral surface of FIG.
~ FIG. 4). The heat shielding member 26 has a cylindrical tubular portion 26a that blocks radiant heat from the heater 18, and a flange portion 26b that supports the tubular portion 26a at its upper edge. By mounting the flange portion 26b on the heat retaining cylinder 19,
The heat shielding member 26 is fixed in the chamber 11 so that the lower edge of the cylindrical portion 26a is located above the surface of the silicon melt 12 by a predetermined distance.

On the lower edge of the cylindrical portion 26a, a cone-shaped first heat radiation suppressing member 3 having a diameter decreasing upward.
1, a connecting member 33 is hung down from a lower edge of the cylindrical portion 26a, and a second cone-shaped heat radiation suppressing member 32 formed to have a smaller diameter toward the upper side is attached to a lower end of the connecting member 33. (FIGS. 2 to 4). The second heat radiation suppressing member 32 is provided below the first heat radiation suppressing member 31 at a predetermined interval. Further, in the chamber 11 above the heat shielding member 26, a cooling cylinder 34 surrounds the ingot 24 being pulled up so as to surround the ingot 24.
1 to be provided. This cooling cylinder 34
A second cylindrical heater 35 is attached to the upper inner surface of the first heater. A cooling passage 34 a through which a cooling fluid passes is formed inside the cooling cylinder 34 below the second heater 35 (in the wall of the cooling cylinder 34), and a lower portion of the cooling cylinder 34 (a portion protruding into the chamber 11). Is formed with a slit 34b extending in the vertical direction (FIGS. 2 and 3). The cooling passage 34a is formed to meander inside the cooling cylinder 34 (inside the wall) so as not to be exposed from the inner peripheral edge of the slit 34b (FIG. 3).
The cooling water is passed through 4a. The slit 34b is formed for visually recognizing the ingot 24 being pulled up from outside the chamber 11.

The first and second heat radiation suppressing members 31, 32
, Ie, the first and second heat radiation suppressing members 31 and 32
The inclination angle θ of the first and second heat radiation suppressing members 31, 32 with respect to the horizontal plane including the lower edge of the lower edge is 80 degrees or less, preferably 20 to
Each is set within the range of 60 degrees (FIG. 4). The heat shielding member 26 and the first and second heat radiation suppressing members 31 and 32 are made of Mo (molybdenum), W (tungsten), C (carbon), or graphite whose surface is coated with SiC.

Further, the chamber 11 is connected to gas supply / discharge means 36 for supplying an inert gas to the silicon single crystal ingot 24 side of the chamber 11 and discharging the inert gas from the inner peripheral surface side of the crucible of the chamber 11. (FIG. 2). One end of the gas supply / discharge means 36 is connected to the peripheral wall of the casing 21 and the other end is a tank (not shown) for storing the inert gas.
And a supply pipe 36a connected to the
And a discharge pipe 36b whose other end is connected to a vacuum pump (not shown). Supply pipe 36a
And these pipes 36a, 36
First and second flow control valves 36c and 36d are provided to adjust the flow rate of the inert gas flowing through b.

The output shaft (not shown) of the pulling motor is provided with a rotary encoder (not shown), and the crucible driving means 17 has a weight sensor (for detecting the weight of the silicon melt 12 in the quartz crucible 13). (Not shown), and a linear encoder (not shown) for detecting the elevation position of the support shaft 16. Each detection output of the rotary encoder, weight sensor and linear encoder is a controller (not shown)
The control output of the controller is connected to the pulling motor of the pulling means 22 and the lifting motor of the crucible driving means 17, respectively. Further, the controller is provided with a memory (not shown). The memory has a wire cable 2 for detecting output of the rotary encoder.
2a, ie, a silicon single crystal ingot 24
Of the silicon melt 1 in the quartz crucible 13 with respect to the detection output of the weight sensor.
The second liquid level is stored as the second map. The controller uses the quartz crucible 1 based on the detection output of the weight sensor.
The motor for raising and lowering the crucible driving means 17 is controlled so that the liquid level of the silicon melt 12 in 3 is always kept at a constant level.

The ingot 24 pulled up from the silicon melt 12 by the pulling device shown in FIG.
It has an oxygen concentration of 10 ¹⁸ atoms / cm ³ (former ASTM) or higher. In the vicinity of the solid-liquid interface between the ingot and the silicon melt, the temperature of the second heat radiation suppressing member 32 rises due to the radiant heat from the high-temperature silicon melt 12, or the radiant heat from the silicon melt 12 (dashed line in FIG. 4). Indicated by arrows)
Alternatively, the heat radiation from the ingot 24 is reflected by the second heat radiation suppressing member 32, so that the rapid heat radiation from the ingot 24 is suppressed (portion a in FIG. 4). As a result, a sharp drop in the temperature of the outer peripheral portion of the ingot 24 can be prevented, so that the crystal temperature gradient in the pulling direction in the ingot 24 becomes substantially uniform from the center to the outer peripheral surface. Therefore, since the occurrence of thermal stress in the ingot 24 can be suppressed, slip generation and dislocation are improved, and the distribution of vacancy type point defects or interstitial silicon type point defects in the radial plane is also uniform. Become.

In an open portion between the first and second heat radiation suppressing members 31 and 32 of the ingot 24, radiant heat from the heater 18 (indicated by solid arrows in FIG. 4) is applied to the ingot 24.
Irradiation is performed on the outer peripheral surface of the ingot 24 to keep the temperature of the outer peripheral portion of the ingot 24 (portion b in FIG. 3), and in the portion surrounded by the first heat radiation suppressing member 31, as shown by a dashed line arrow in FIG. The temperature of the first heat radiation suppressing member 31 rises due to the radiant heat from the silicon melt 12 of the ingot 2 or
The first heat radiation suppressing member 31 reflects the heat radiation from the heat sink 4, thereby suppressing the rapid heat radiation from the ingot 24 (portion c in FIG. 3). As a result, the uniformity of the radial distribution of the crystal temperature gradient in the pulling direction near the solid-liquid interface is improved,
The reaction time of the diffusion and the annihilation of the point defect in the ingot 24 on the slope can be made uniform, and the concentration of the point defect can be optimally controlled.

In a portion above the first heat radiation suppressing member 31, the radiant heat from the first heater 18 and the silicon melt 12 is dissipated by the first heat radiation suppressing member 31 and the cylindrical portion 26 of the heat shielding member 26.
a, the cooling water passes through the cooling passage 34a in the cooling cylinder 34 above the first heat radiation suppressing member 31, so that the ingot 24 is rapidly cooled (FIG. 4d). part). As a result, it is possible to increase the temperature gradient value in the pulling direction in the ingot 24 in this portion and a portion thereunder.

Further, when the ingot 24 cooled by the cooling cylinder 34 is pulled up and reaches the position of the second heater 35, it is heated by the second heater 34. The lowermost position of the second heater 35 is determined by the ingot 2 pulled up.
This is the position where the temperature of 4 drops to a predetermined temperature. This predetermined temperature is selected from a temperature range of 1000 to 600 ° C. If the temperature is lower than 600 ° C., oxygen precipitation nuclei are difficult to stably develop at a high temperature, and if the temperature exceeds 1000 ° C., OSF nuclei are generated. Preferably it is 900-600 degreeC.

Next, the heating condition of the second heater 35 depends on whether the pulled ingot consists of both the above-mentioned region [P _V ] and the above-mentioned region [P _I ] or only the region [P _I ]. And the oxygen concentration contained in the ingot. (a) The ingot is composed of both the region [P _V ] and the region [P _I ] shown in FIG. 6 and has an oxygen concentration of 1.4 × 10
^{In the case of 18} atoms / cm ³ or more (old ASTM) or more, the second
Heat to maintain time. (b) The ingot is composed of both the region [P _V ] and the region [P _I ] shown in FIG.
In the case of 2 × 10 ¹⁸ atoms / cm ³ (former ASTM), the second heater 35 sets the temperature at 600 to 500 ° C. to 20 × 10 ¹⁸ atoms / cm ^3.
Heat to maintain for ~ 50 hours. (c) The ingot consists of only the region [P _I ] shown in FIG. 6 and its oxygen concentration is 1.4 × 10 ¹⁸ atoms / cm.
³ (old ASTM) or higher, the second heater 35
20 to 50 hours at 600 ° C or 6 to 5 at 500 ° C
Heat to maintain 0 hours. (d) The ingot consists of only the region [P _I ] shown in FIG. 6 and its oxygen concentration is 1.1 to 1.2 × 10 ¹⁸ atoms.
In the case of s / cm ³ (old ASTM), the second heater 35 heats at 500 ° C. for 50 hours.

As the oxygen concentration of the ingot is lower and the region [P _I ] is larger, oxygen precipitation nuclei tend to be less likely to be generated. In the case where oxygen precipitate nuclei are hardly generated, in order to sufficiently develop the oxygen precipitate nuclei, the temperature of the ingot is adjusted to 500 ° C. to 600 ° C. by adjusting the power of the second heater.
C. and the heating time is about 50 hours. If the heating temperature and the heating maintenance time of the second heater are each less than the lower limit, the expression of oxygen precipitation nuclei is not sufficient, and the expression density of the oxygen precipitation nuclei does not change even if each exceeds the upper limit,
Determined as above.

FIG. 5 shows the relationship between the ingot temperature and the distance from the silicon melt surface when an ingot consisting only of the region [P _V ] is pulled up. When the second heater 35 is not operated, the temperature of the ingot decreases relatively quickly with the pull-up length, as shown by the broken line in the plot of the white circle in FIG. 5, whereas as shown by the plot of the black circle in FIG. When the second heater 35 according to this embodiment is operated, the temperature of the ingot, which has dropped to a predetermined temperature within the temperature range of 1000 to 600 ° C. by pulling, is reduced. Second
The temperature distribution can be changed by the power of the heater.
For example, when the oxygen concentration of the ingot is about 1.5 × 10 ^{18 at}
When the crystal is maintained at a temperature profile of 500 ° C. for 6 hours when the temperature profile is 500 ms / cm ³ (old ASTM), as indicated by reference symbol L in FIG. Oxygen precipitation nuclei having a predetermined density or more are developed in the state of the ingot. Symbol H in FIG. 5 is a temperature profile in which the temperature of the ingot is maintained at a position of 1050 ° C. That is, the ingot maintenance temperature can be changed according to the power of the second heater. When a silicon wafer is cut out from the heat-treated ingot and the wafer is heat-treated in a device manufacturing process of a semiconductor device maker, the oxygen precipitate nuclei grow into BMD. Whether the ingot consists of both the region [P _V ] and the region [P _I ], or consists only of the region [P _I ], the wafer cut from these ingots has an IG effect over the entire surface of the wafer. Become like The BMD density is measured by selectively etching the wafer surface with a Wright etchant, and then observing the surface at a depth of 250 μm from the wafer surface with an optical microscope. The volume density is 1 × 10 ⁸ / cm ³ or more, preferably 1 × 1
0 is ^eight / cm ³ ~1 × 10 ¹¹ atoms / cm ^3.

[0030]

Next, examples of the present invention will be described together with comparative examples. <Examples 1 to 8> Boron (B) -doped p having a diameter of 6 inches was obtained by using a silicon single crystal pulling apparatus shown in FIG.
The mold silicon ingot was pulled up. This ingot had a straight body diameter of 6 inches, a length of 600 mm, a crystal orientation of (100), a resistivity of 1 to 15 Ωcm, and an oxygen concentration of 1.4 to 1.5 × 10 ¹⁸ atoms / cm ³ (old). AST
M). For ingots, the V / G at the time of pulling is set at 0.
With decreasing continuously from 24 mm ² / min ° C. until 0.18 mm ² / minute ° C., while heating by the second heater grown under conditions of eight as shown in Table 1, was grown by two in each condition. The heating of the ingot by the second heater was performed under the eight conditions shown in Table 1 when the temperature of the ingot dropped to 900 ° C. by pulling. One ingot of the two grown ingots cuts the center of the ingot in the pulling direction as shown in FIG. 6, and examined the position of each area.
A silicon wafer was cut out from another portion from the portion shown in the line of FIG. 6 to obtain a sample. FIG. 6 corresponds to FIG. 1B. In this example, the wafer serving as the sample includes both the region [P _V ] and the region [P _I ].

<Comparative Examples 1 to 14> Pulled up as in Example 1.
As shown in Table 1 while heating the ingot thus obtained with the second heater
They were grown under 14 conditions. Oxygen concentration of this ingot
Is 1.4 to 1.5 × 10 ¹⁸atoms / cm^Three(Old AS
TM) and the line in FIG. 6 at the same position as in Example 1.
A sample was cut out from the portion shown in FIG. This sample
Wafers are placed in the area [P_V] And area [P_I] From both sides
Become.

Comparative Example 15 The oxygen concentration was 1.4 to 1.5.
× 10 ¹⁸ atoms / cm ³ (former ASTM)
A silicon wafer was cut out from the portion indicated by the line in FIG. 6 which was the same position as in Example 1 except that the second heater was not operated, and used as a sample. The wafer serving as the sample includes both the region [P _V ] and the region [P _I ].

<Comparative Evaluation 1> Examples 1 to 8 and Comparative Example 1
~ 15 wafers were each placed in a wet oxygen atmosphere for 120
After heating at 0 ° C. for 60 minutes to perform the OSF revealing heat treatment, seco etching was performed for 2 minutes. As a result, for all wafers, the entire O
SF free. Examples 1 to 8 and Comparative Examples 1 to
After heat-treating the 15 wafers under a nitrogen atmosphere at 800 ° C. for 4 hours, and subsequently heat-treating at 1000 ° C. for 16 hours,
After the wafer surface is etched by 100 μm with a mixed acid, the wafer surface is further selectively etched with a Wright etchant, and the center of the wafer and the radius R (3 inches) of the wafer are determined by observation with an optical microscope.
Each BM near 3, that is, about 2 inches from the center of the wafer
D was measured. Then, the volume density was converted from the etching depth of the light etching to the volume density. Table 1 shows the results.

[0034]

[Table 1]

As is clear from Table 1, the oxygen concentration of the wafer is relatively high at 1.4 to 1.5 × 10 ¹⁸ atoms / cm ³ and includes the region [P _V ]. Example 1 heated at 600 to 500 ° C. for 2 to 50 hours
1 × 1 which is considered to have an IG effect for wafers of ~ 8
0 had ^eight / cm ³ or more BMD volume density. On the other hand, in Comparative Examples 1 to 14 in which the second heater was heated at a high temperature of 700 ° C. or higher and Comparative Example 15 in which the second heater was disabled, a BMD volume density having an IG effect was obtained at the center of the wafer. With 2 / 3R of the wafer, a BMD volume density of only 10 ⁷ / cm ³ was obtained, and the IG effect over the entire surface of the wafer could not be expected.

<Examples 9 to 12> When the oxygen concentration is 1.1 to
An ingot was grown in the same manner as in Example 1 except that the concentration was 1.2 × 10 ¹⁸ atoms / cm ³ (old ASTM). Second
The ingot was heated by the heater under the four conditions shown in Table 2 when the temperature of the ingot was lowered to 900 ° C. by pulling. FIG. 6 showing the same position as in the first embodiment.
Cut out the silicon wafer from the part shown in the line,
A sample was used. The wafer serving as this sample has an area [P _V ]
And the area [P _I ].

<Comparative Examples 16 to 33>
Shown in Table 2 while heating the raised ingot with the second heater
They were raised under 18 different conditions. The oxygen concentration of this ingot
The degree is 1.1 to 1.2 × 10 ¹⁸atoms / cm^Three(Old A
STM) and the line in FIG.
A sample was cut out from the portion shown in FIG. This sample
The ruta wafer has an area [P_V] And area [P_I]
Become.

<Comparative Example 34> The oxygen concentration was 1.1 to 1.2.
× 10 ¹⁸ atoms / cm ³ (former ASTM)
A silicon wafer was cut out from the portion indicated by the line in FIG. 6 which was the same position as in Example 1 except that the second heater was not operated, and used as a sample. The wafer serving as the sample includes both the region [P _V ] and the region [P _I ].

<Comparative Evaluation 2> The wafers of Examples 9 to 12 and Comparative Examples 16 to 34 were subjected to OSF
After performing the revealing heat treatment, seco etching was performed for 2 minutes. As a result, all the wafers were OSF-free over a depth of 20 μm from the surface. Comparative evaluation 1 was performed on the wafers of Examples 9 to 12 and Comparative Examples 16 to 34.
In the same manner as described above, the wafer center and the wafer radius R (3
) Of each BMD was determined. Table 2 shows the results.

[0040]

[Table 2]

As is clear from Table 2, the wafer contained the region [P _V ], but the oxygen concentration was 1.1 to 1.0.
Since it is relatively low at 2 × 10 ¹⁸ atoms / cm ³ , only the wafers of Examples 9 to 12 heated at 600 to 500 ° C. for 20 to 50 hours with the second heater are considered to have an IG effect only 1 ×. It did not have a BMD volume density of 10 ⁸ / cm ³ or more. And the second heater at 600-500 ° C
Comparative Example 16 to 6 hours or heated at a high temperature of 700 ° C. or more
33 and Comparative Example 34 in which the second heater was deactivated, a BMD volume density of 4 × 10 ⁷ / cm ³ or less was obtained at で R of the wafer, and an IG effect over the entire surface of the wafer could be expected. Did not.

<Examples 13-17> The oxygen concentration was 1.4-.
An ingot was grown in the same manner as in Example 1 except that the density was 1.5 × 10 ¹⁸ atoms / cm ³ (old ASTM). Second
The ingot was heated by the heater under the five conditions shown in Table 3 when the temperature of the ingot was lowered to 900 ° C. by pulling. A silicon wafer was cut out from the portion indicated by the line in FIG. 6 to obtain a sample. The wafer serving as this sample consists only of the region [P _I ].

<Comparative Examples 35 to 51> An ingot pulled in the same manner as in Example 13 was grown under the 17 conditions shown in Table 3 while being heated by the second heater. The oxygen concentration of this ingot was 1.4 to 1.5 × 10 ¹⁸ atoms / cm ³ (former ASTM), and it was cut out from the portion shown in the line of FIG. The wafer serving as the sample consists only of the region [P _I ].

<Comparative Example 52> The oxygen concentration was 1.4 to 1.5.
× 10 ¹⁸ atoms / cm ³ (former ASTM)
A silicon wafer was cut out from the portion indicated by the line in FIG. 6 which was the same position as in Example 13 except that the second heater was not operated, and used as a sample. The wafer serving as this sample consists only of the region [P _I ].

<Comparative Evaluation 3> The wafers of Examples 13 to 17 and Comparative Examples 35 to 52 were subjected to OS
After performing the heat treatment for revealing F, seco etching was performed for 2 minutes. As a result, 20 μm from the surface of all wafers
Was OSF-free over the entire depth. Further, the wafers of Examples 13 to 17 and Comparative Examples 35 to 52 were processed in the same manner as in Comparative evaluation 1 to determine the center of the wafer and the radius R of the wafer.
Each BMD volume density in the vicinity of ／ of (3 inches) was determined. Table 3 shows the results.

[0046]

[Table 3]

As is clear from Table 3, the wafer had only the region [P _I ], but the oxygen concentration was 1.4.
Because of the relatively high value of about 1.5 × 10 ¹⁸ atoms / cm ³ , the IG of the wafers of Examples 13 to 17 heated by the second heater at 600 ° C. for 20 to 50 hours or at 500 ° C. for 6 to 50 hours. 1 × 10 ⁸ which is said to be effective /
It had a BMD volume density of at least cm ³ . Further, the second heater is used at 500 ° C. for 2 hours, at 600 ° C. for 2 to 6 hours or 700
In Comparative Examples 35 to 51 heated at a high temperature of not less than 0 ° C. and Comparative Example 52 in which the second heater was deactivated, only a BMD volume density of 4 × 10 ⁷ / cm ³ or less was obtained at 2 / 3R of the wafer. The IG effect over the entire surface could not be expected.

Example 18 Oxygen concentration is 1.1 to 1.2
An ingot was grown in the same manner as in Example 1 except that the density was × 10 ¹⁸ atoms / cm ³ (old ASTM). The ingot was heated by the second heater under the conditions shown in Table 4 when the temperature of the ingot was lowered to 900 ° C. by pulling. A silicon wafer was cut out from the portion indicated by the line in FIG. 6 to obtain a sample. The wafer serving as this sample consists only of the region [P _I ].

<Comparative Examples 53 to 73> The ingot pulled up in the same manner as in Example 18 was grown under the 21 conditions shown in Table 4 while being heated by the second heater. The ingot had an oxygen concentration of 1.1 to 1.2 × 10 ¹⁸ atoms / cm ³ (former ASTM), and was cut out from the portion shown in the line of FIG. The wafer serving as the sample consists only of the region [P _I ].

<Comparative Example 74> The oxygen concentration was 1.1 to 1.2.
× 10 ¹⁸ atoms / cm ³ (former ASTM)
A silicon wafer was cut out from the portion indicated by the line in FIG. 6 which was the same position as in Example 16 except that the second heater was not operated, and used as a sample. The wafer serving as this sample consists only of the region [P _I ].

<Comparative Evaluation 4> Example 18 and Comparative Example 53
After subjecting the wafers No. to No. 74 to the OSF revealing heat treatment in the same manner as in Comparative evaluation 1, the secco etching was performed for 2 minutes.
As a result, all the wafers were OSF-free over a depth of 20 μm from the surface. Example 18
In the same manner as in Comparative Evaluation 1, the wafers of Comparative Examples 53 to 74 and the center of the wafer and the radius R of the wafer (3 inches)
Of each of the BMDs in the vicinity of 2/3 was determined. Table 4 shows the results.

[0052]

[Table 4]

As is evident from Table 4, the oxygen concentration of the wafer is relatively low at 1.1 to 1.2 × 10 ¹⁸ atoms / cm ^3, and the oxygen concentration of the wafer may consist of only the region [P _I ]. Example 1 heated by a two heater at 500 ° C. for 50 hours
1 × 10 ⁸ which is said to have an IG effect only on wafers of ⁸
Pcs / cm ³ or more. The second heater is used at 500 ° C. for 2 to 20 hours and at 600 ° C. for 2 to 20 hours.
Comparative Example 53 heated for 50 hours or at a high temperature of 700 ° C. or more
-73 and Comparative Example 74 in which the second heater was deactivated, a BMD volume density of 5 × 10 ⁷ / cm ³ or less was obtained at ／ R of the wafer, and it was expected that the IG effect over the entire surface of the wafer was expected. could not.

[0054]

As described above, according to the present invention, even if the ingot is composed of both the region [P _V ] and the region [P _I ], the ingot is composed of only the region [P _I ]. Also has an oxygen concentration of 1.1 × 10 ¹⁸ atoms / cm
^{3 When} the temperature of the silicon single crystal ingot of (old ASTM) or higher is lowered to a predetermined temperature within a temperature range of 1000 to 600 ° C. by pulling, the ingot is heated and maintained at a predetermined temperature for a predetermined time. In addition to the absence of point defect aggregates, oxygen precipitate nuclei having a predetermined density or more appear in the state of the ingot. If a silicon wafer is cut out from the heat-treated ingot and the wafer is heat-treated in a device manufacturing process of a semiconductor device maker, the above-mentioned oxygen precipitate nuclei grow to a BMD volume density of 1 × 10 ⁸ / cm ³ or more. An IG effect can be obtained.

[Brief description of the drawings]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 A: Based on Voronkov's theory, when the V / G ratio is above the critical point, a vacancy-rich ingot is formed and the V / G
The figure which shows that an interstitial silicon-rich ingot is formed when a ratio is below a critical point. B: Conceptual diagram by X-ray topograph after heat treatment of the ingot in an N ₂ atmosphere at 1000 ° C. for 40 hours. C Defect distribution diagram of the crystal when the ingot immediately after pulling (as-grown state) is seco-etched. D is a defect distribution diagram of crystals when the ingot is heat-treated in a humid O ₂ atmosphere and then seco-etched. The figure which shows the change situation of the pulling speed of E ingot.

FIG. 2 is a cross-sectional configuration diagram showing a silicon single crystal pulling apparatus according to the embodiment and the first embodiment of the present invention.

FIG. 3 is a perspective view of a main part including a cooling cylinder of the lifting device.

FIG. 4 is a sectional configuration diagram showing an isothermal surface of a silicon single crystal ingot being pulled by the pulling device.

FIG. 5 is a diagram showing a temperature change state of an ingot according to a distance from a silicon melt surface according to activation and non-activation of a second heater.

FIG. 6 is a view corresponding to FIG. 1B.

[Explanation of symbols]

 REFERENCE SIGNS LIST 10 Pulling device 11 Chamber 12 Silicon melt 13 Quartz crucible 18 First heater 24 Silicon single crystal ingot 26 Heat shielding member 31 First heat dissipation suppressing member 32 Second heat dissipation suppressing member 34 Cooling cylinder 34a Cooling passage 35 Second heater

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Yuji Nakata 1-297 Kitabukuro-cho, Omiya-shi, Saitama Mitsubishi Materials Silicon Research Center F-term (reference) 4G077 AA02 AB01 BA04 CF10 EG20 FE11 FE18 HA12 5F053 AA12 AA50 BB58 DD01 FF04 GG01 HH04 RR03

Claims (4)

[Claims] A region where interstitial silicon type point defects predominantly exist in a silicon single crystal ingot is [I], a region where vacancy type point defects predominantly exist is [V], In a method of pulling a silicon single crystal ingot composed of the perfect region [P], wherein a perfect region where no aggregate of inter-silicon type point defects and no aggregate of vacancy type point defects exist is defined as [P], A region adjacent to [I] and belonging to the perfect region [P] and having a minimum interstitial silicon concentration at which interstitial dislocations can be formed is referred to as [P _I ], and a region adjacent to the region [V] and corresponding to the perfect region When a region belonging to [P] and having a vacancy concentration not higher than that capable of forming a COP or FPD is defined as [P _V ], the silicon single crystal ingot includes both the region [P _V ] and the region [P _I ]. And an oxygen concentration of 1.4 × 1
0 ¹⁸ atoms / cm ³ (old ASTM) or higher, and the temperature of the silicon single crystal ingot is 1 by pulling.
A method for pulling a silicon single crystal, characterized in that the silicon single crystal ingot is heated at a time point when the temperature falls to a predetermined temperature within a temperature range of 000 to 600 ° C. and the temperature is maintained at 600 to 500 ° C. for 2 to 50 hours. . 2. A region where an interstitial silicon type point defect predominantly exists in a silicon single crystal ingot is [I], a region where a vacancy type point defect predominantly exists is [V], In a method of pulling a silicon single crystal ingot composed of the perfect region [P], wherein a perfect region where no aggregate of inter-silicon type point defects and no aggregate of vacancy type point defects exist is defined as [P], A region adjacent to [I] and belonging to the perfect region [P] and having a minimum interstitial silicon concentration at which interstitial dislocations can be formed is referred to as [P _I ], and a region adjacent to the region [V] and corresponding to the perfect region When a region belonging to [P] and having a vacancy concentration not higher than that capable of forming a COP or FPD is defined as [P _V ], the silicon single crystal ingot includes both the region [P _V ] and the region [P _I ]. And the oxygen concentration is 1.1 to
1.2 × 10 ¹⁸ atoms / cm ³ (old ASTM), wherein the temperature of the silicon single crystal ingot is 1 by pulling.
A method for pulling a silicon single crystal, characterized in that the silicon single crystal ingot is heated at a point of time when the temperature drops to a predetermined temperature within a temperature range of 000 to 600 ° C., and the temperature is maintained at 600 to 500 ° C. for 20 to 50 hours. . 3. A region in which interstitial silicon type point defects are predominantly present in a silicon single crystal ingot is [I], and a region in which vacancy type point defects are predominantly is [V]. In a method of pulling a silicon single crystal ingot composed of the perfect region [P], wherein a perfect region where no aggregate of inter-silicon type point defects and no aggregate of vacancy type point defects exist is defined as [P], When a region adjacent to [I] and belonging to the perfect region [P] and having a minimum interstitial silicon concentration capable of forming an interstitial dislocation is defined as [P _I ], the silicon single crystal ingot is formed in the region [P _I ] and an oxygen concentration of 1.4 × 10 ¹⁸ atoms / c
m ³ (former ASTM) or higher, and the temperature of the silicon single crystal ingot is 1 by pulling.
When the silicon single crystal ingot is heated to a predetermined temperature within a temperature range of 000 to 600 ° C., the temperature is raised to 600 ° C. for 20 to 50 hours or 500 ° C. to 6 to 50 hours.
A method for pulling a silicon single crystal, characterized by maintaining the time. 4. A region where interstitial silicon type point defects are predominantly present in a silicon single crystal ingot is [I], a region where vacancy type point defects are predominantly present is [V], In a method of pulling a silicon single crystal ingot composed of the perfect region [P], wherein a perfect region where no aggregate of inter-silicon type point defects and no aggregate of vacancy type point defects exist is defined as [P], When a region adjacent to [I] and belonging to the perfect region [P] and having a minimum interstitial silicon concentration capable of forming an interstitial dislocation is defined as [P _I ], the silicon single crystal ingot is formed in the region [P _I ] and an oxygen concentration of 1.1 to 1.2 × 10 ¹⁸ atom
ms / cm ³ (former ASTM) or more, and the temperature of the silicon single crystal ingot is 1 by pulling.
A method for pulling a silicon single crystal, wherein the silicon single crystal ingot is heated at a time when the temperature has dropped to a predetermined temperature within a temperature range of 000 to 600 ° C., and the temperature is maintained at 500 ° C. for 50 hours.