JPH07267775A - Single crystal pulling up method - Google Patents

Abstract

PURPOSE:To obtain a single crystal having a high number of oxygen atoms by covering the free surface of a molten layer with a cap brought into contact therewith in a process of forming a solid layer on the lower layer of a crucible. CONSTITUTION:The free surface of the molten layer is covered with the cap brought into contact therewith in the process of forming the solid layer on the lower layer in the crucible in the method for pulling up the single crystal from the molten layer while melting the solid layer by dissolving all the raw materials for the crystal in the crucible, then solidifying the raw materials for the crystal at the lower layer in the state of allowing the molten layer to remain on the upper layer in the crucible thereby forming the solid layer. The occupying area of the cap is specified to >=40% of the free surface, by which the single crystal having the number of oxygen of >=12X10<17>-atoms/cc is obtd. Fig. illustrates the process of forming the solid layer 6; (a) is the longitudinal cross section of the crucible l and (b) is a plane view. The cap 12 is supported by a hanging rod 12a fixed to the bottom end of a pulling up shaft. The cap 12 is made of SiO2 and is formed to a disk of a thickness about 10mm. While the shape of the cap is not particularly limited, the cap is so formed as to cover 50 to 70% of the surface of the molten layer 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to melting a solid layer of a raw material for a single crystal in an upper layer and a solid layer of the raw material for a single crystal in a lower layer while melting the solid layer in the crucible. The present invention relates to a single crystal pulling method by a molten layer method for pulling a single crystal from the molten layer.

[0002]

2. Description of the Related Art Various methods for growing a single crystal have hitherto been proposed, one of which is the Czochralski method (hereinafter referred to as the CZ method). FIG. 6 is a cross-sectional view schematically showing a conventional crystal growth apparatus using the CZ method.
Indicates a crucible. The crucible 1 is composed of a bottomed cylindrical quartz inner layer container 1a and a bottomed cylindrical graphite outer layer holding container 1b fitted to the outside of the inner layer container 1a. And is supported by a rotatable support shaft 2. A resistance heating type heater 3 is arranged outside the crucible 1, and a heat retaining cylinder 4 is arranged concentrically outside the heater 3.

In the crucible 1, there is a melt of the crystal raw material which is melted by the heater 3. See also crucible 1
A pulling shaft 7 made of a pulling rod, a wire, or the like that can be rotated in the same direction as the support shaft 2 in the same direction or in the opposite direction and can be moved up and down is suspended from the central axis of the pulling shaft 7. The seed crystal 8 attached to the melted liquid is brought into contact with the surface of the melt, and the pulling shaft 7 is rotated upward in accordance with the crystal growth to pull up the melt, thereby solidifying the melt and growing the single crystal 9.

By the way, in such a CZ method, impurities are added (doping) into the melt in order to adjust the electric resistivity and electric conductivity type of the single crystal 9 before pulling the single crystal. There is a problem that the concentration in the single crystal 9 of No. 1 changes along the pulling direction, a so-called segregation phenomenon occurs, and the single crystal 9 having uniform electric characteristics in the pulling direction cannot be obtained.

The melt layer method is one of the methods for growing a single crystal while suppressing the segregation of impurities as described above. FIG. 7 is a sectional view schematically showing a crystal growth apparatus by the melt layer method. The single crystal raw material in the crucible 1 having the same structure as shown in FIG. 6 is once melted by using the heater 3, and then solidified from the bottom side of the crucible 1 to form the upper layer for the single crystal. The molten layer 5 of the raw material and the solid layer 6 obtained by solidifying the molten liquid are formed in the lower layer. While pulling the single crystal 9 from the molten layer 5, the heater 3 is gradually lowered to gradually melt the solid layer 6 in association with this pulling, and the molten layer 5 and the fixed layer 6 are coexisted with each other. The single crystal 9 is pulled up from the layer 5. The other structure is substantially the same as that shown in FIG. 6, and the corresponding parts are designated by the same reference numerals.

The melt layer method is divided into a melt layer thickness constant method and a melt layer thickness changing method. In the constant melting layer thickness method, when pulling the single crystal 9, the heater 3 is moved from the upper side to the lower side to make the volume of the melting layer 5 in the crucible 1 constant regardless of the amount of the pulled single crystal 9. This is a method of melting the solid layer 6 so as to keep it.
Japanese Examined Patent Publication No. 62-880 and Sho 60-32474.
It is disclosed in Japanese Patent Publication No. In this method, the solid layer 6 having a low impurity concentration is newly melted with the growth of the single crystal 9 to reduce the impurity concentration in the molten layer 5, and the impurity is added to the molten layer 5 in order to adjust the impurity concentration. Addition is performed to keep the impurity concentration in the molten layer 5 substantially constant and suppress segregation of impurities in the single crystal 9.

On the other hand, the molten layer thickness changing method intentionally uses the molten layer 5
Is a method for suppressing the segregation of impurities in the single crystal 9 by keeping the impurity concentration of the molten layer 5 constant without adding impurities during the pulling of the single crystal by changing the liquid amount of
JP-A-61-250961, JP-A-61-250
It is disclosed in Japanese Patent Laid-Open No. 962 and Japanese Patent Laid-Open No. 61-215285. The fused layer method as described above (hereinafter DL
The single crystal pulling method based on the CZ method) has an advantage that the electric resistivity is uniform because the impurity concentration in the pulled single crystal 9 in the pulling direction is constant.

By the way, a major factor that influences the characteristics of a single crystal is the oxygen concentration distribution in the single crystal, in addition to the above-mentioned deviation of impurities. For example, the number of oxygen atoms in a silicon (Si) single crystal is about 8 to 12 × 10 ¹⁷ atoms / cc in the case of a silicon single crystal obtained by the DLCZ method, which is in a supersaturated state. The oxygen atoms of are not solid-solved in the Si single crystal and form oxygen precipitates as they are cooled. In the growth process of this oxygen precipitate, a phenomenon for alleviating strain such as release of Si atoms, release of vacancies and generation of dislocation loop called punch-out is caused.
In addition, Si atoms existing in the interstitial lattice move around oxygen precipitates during high temperature heat treatment, and grow dislocation loops accompanied by stacking faults.

On the other hand, when such a crystal defect exists deep from the wafer surface, that is, deeper than the active region of the device, the dislocation generated in the wafer is fixed and the strength of the wafer is increased. During the heat treatment process of the wafer, oxygen precipitates (SiOx) are grown and strain is generated around them, and impurities that enter the surface area of the wafer in the device process are captured (hereinafter referred to as intrinsic gettering), It also has the advantage of preventing contamination of the active area of the.

FIG. 8 is an explanatory view showing a mode of incorporating oxygen atoms into the molten layer 5 in the CZ method and the DLCZ method.
The right half of the figure is for the CZ method, and the left half is for DLC.
Each case is shown by the Z method.

In the CZ method, the molten liquid is in contact with the inner layer container 1a of the crucible 1 over a wide area except for its free surface.
Due to the melting of the inner layer container 1a from here, a large amount of oxygen atoms are supplied to the melt. On the other hand, in the DLCZ, since the solid layer 6 exists in the lower layer, the molten layer 5 and the crucible 1
The contact area with the inner layer container 1a is extremely small and the inner layer container 1a
The amount of melted oxygen is small, and the amount of oxygen atoms supplied is greatly reduced. By the way, in a silicon wafer used as a semiconductor device, an oxygen concentration higher than a predetermined concentration is required in order to suppress warpage by taking advantage of the oxygen atom described above and to obtain a gettering effect of heavy metals in a device process. In such a silicon substrate, oxygen atoms of 12 × 10 ¹⁷ atoms / cc or more are required.

[0012]

However, as described above, DL
Segregation can be prevented by the CZ method, which is extremely convenient for obtaining a single crystal 9 having a low oxygen concentration, but it is extremely difficult to obtain a single crystal 9 having a high oxygen concentration.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a single crystal pulling method capable of obtaining a single crystal having a high oxygen concentration by using the DLCZ method. is there.

[0014]

The method for pulling a single crystal according to the present invention is to melt all the crystal raw material in the crucible and then solidify the crystal raw material in the lower layer while leaving the molten layer in the upper layer in the crucible. To form a solid layer, while dissolving the solid layer,
In the single crystal pulling method for pulling a single crystal from the melted layer, in the process of forming a solid layer in the lower layer in the crucible, the free surface of the molten layer is covered with a lid in contact with it, and the lid is occupied. The area is 40% or more of the free surface.

[0015]

In the present invention, since the free surface of the molten layer is covered with the lid in contact with the free surface of the molten layer,
The evaporation amount of oxygen atoms from the free surface is suppressed, the oxygen atom concentration in the melt is increased, and the oxygen atom concentration in the solid layer is also increased, and the solid layer melts during the pulling of the single crystal. Oxygen atoms will be supplied to the liquid. Further, by making the area occupied by the lid 40% or more of the free surface of the molten layer, it becomes possible to obtain a single crystal having the number of oxygen atoms of 12 × 10 ¹⁷ atoms / cc or more.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A method for pulling a single crystal according to the present invention will be specifically described below with reference to the drawings. FIG. 1 is a cross-sectional view schematically showing a single crystal growth apparatus used in the method for pulling a single crystal according to the present invention, in which 1 denotes a crucible. The crucible 1 includes a bottomed cylindrical quartz inner layer container 1a and a bottomed cylindrical graphite outer layer holding container 1b fitted to the outside of the inner layer container 1a. A support shaft 2 is attached to a substantially central portion of the bottom of the crucible 1, and the support shaft 2 supports the crucible 1 so that the crucible 1 can rotate and move up and down. Further, a resistance heating type heater 3 is arranged on the outer circumference of the crucible 1, and a heat retaining cylinder 4 is concentrically arranged on the outer circumference of the heater 3.

A chamber 1 (not shown) is provided above the crucible 1.
The pulling shaft 7 is hung so that it can rotate and ascend and descend through a small, substantially cylindrical pull chamber 10a that is connected to the upper part of 0, and a seed crystal 8 is attached to the lower end of the pulling shaft 7. ing. On the other hand, in the crucible 1, a solid layer 1 is formed in the lower part and a molten layer 5 is formed in the upper part, and after the lower end of the seed crystal 8 is immersed in the molten layer 5 in the crucible 1, By rotating and raising the single crystal 9, the single crystal 9 is grown from the lower end of the seed crystal 8.

FIG. 2 is an explanatory view showing the processing steps of the silicon single crystal pulling method by the DLCZ method as described above.
First, polycrystalline silicon 11 which is a raw material for a silicon single crystal is charged into a crucible 1 as shown in FIG.
As shown in (b), the crucible 1 is rotated around the support shaft and moved up and down to melt all the polycrystalline silicon 11 by the heater 3.

After melting all of the polycrystalline silicon 11 which is the raw material for single crystal, the temperature on the bottom side of the crucible 1 is lowered to form the solid layer 6 as shown in FIG. 2 (c). ), The solid layer 6 is melted and the single crystal 9 is melted from the molten layer 5.
The solid layer 6 is formed in the manner shown in FIG. 3A and 3B show a process of forming the solid layer 6, FIG. 3A is a vertical sectional view of the crucible 1, and FIG. 3B is a plan view. As is clear from FIG. 3, the lid 12 is arranged in contact with the free surface of the molten layer 5 so as to cover the free surface of the molten layer 5. The lid 12 is supported on the lower end of the pulling shaft 7 by a suspension rod 12a fixed to the upper surface of the lid 12.

The lid 12 is made of SiO ₂ and has a thickness of 10 m.
It is formed into a disk shape of about m, and the molten layer 5 in the crucible 1
Is formed so as to cover 50 to 100%, preferably about 50 to 70% of the surface. The shape of the lid 12 itself is not particularly limited, and may be rectangular, polygonal, elliptical, or the like. The material of the lid 12 is also SiO
The material is not limited to ₂ as long as it is substantially the same as the material of the molten layer 5.

As a result, when the solid layer 6 is formed in the crucible 1, the evaporation of oxygen atoms from the free surface of the molten layer 5 is suppressed, so that the concentration of oxygen atoms in the molten layer 5 increases, and Correspondingly, the number of oxygen atoms in the solid layer 6 also increases, and as a result, in the process of pulling up the single crystal 9 by the DLCZ method, the molten layer 5
Although the contact area between the inner layer container 1a of the crucible 1 and the crucible 1 is reduced, oxygen atoms are supplied by the dissolution of the solid layer 6 having a large number of oxygen atoms, and the concentration of oxygen atoms in the melt is increased, so that the single crystal 8 It is possible to increase the number of oxygen atoms contained therein.

FIG. 4 is an explanatory view showing the relationship between the occupation rate of the lid 12 on the free surface of the molten layer 5 of the crucible 1 and the number of oxygen atoms in the single crystal 9, with the horizontal axis representing the pulling direction of the single crystal ( 3 is a graph showing the position (mm) of each part in the top (0 is set) and the number of oxygen atoms (× 10 ¹⁷ atoms / cc) on the vertical axis. In the graph, □ indicates the area occupied by the lid is 0% (lid 12
(In the case of not using), the + mark shows the result when the lid occupies 25% of the area, the ⋄ mark shows the case where the lid occupies 50%, and the Δ mark shows the result when the lid occupies 75%.

As is clear from this graph, the number of oxygen atoms in the substrate is 12 × 10 ¹⁷ atoms / cc or more in the case of ⋄, that is, when the area occupied by the lid 12 is 50% or more and it is used for a normal MOS transistor. It can be seen that a single crystal of is obtained. Therefore, the occupied area of the lid 12 is 40% or more 10
It is understood that it is preferable to set it to 0%, preferably 50% or more.

FIG. 5 is a distribution diagram of the electric resistivity of the single crystal obtained when the lid 12 is not provided and when the occupying area of the lid 12 is 25%, 50%, and 75%, respectively. The length in the direction (mm) is shown, and the vertical axis shows the electrical resistivity (Ω-cm). The plot symbols in the graph correspond to those in FIG. The thick line plotted by ● is the average value of the whole.

As is apparent from this graph, the length contained in the target range of electrical resistivity of 8.4 to 12 (Ω-cm) is 0 to 5 for the silicon single crystal obtained by the conventional method.
While it is 50 mm, in the case of a silicon single crystal obtained by the method of the present invention, it is in the range of 50 to 10,000 mm,
It can be seen that the target electrical resistivity is obtained over a length that is approximately twice as long.

[0026]

As described above, in the method of the present invention, by placing the lid on the free surface of the molten layer in the process of forming the solid layer, the evaporation of oxygen is prevented and the oxygen concentration in the molten layer is increased. It is possible to increase the oxygen content in the solid layer by solidifying the solid phase, and during the crystal pulling, oxygen atoms are supplied to the melt due to the melting of the inner wall of the crucible and the dissolution of the solid layer. It has an excellent effect of obtaining a single crystal.

[Brief description of drawings]

FIG. 1 is a schematic cross-sectional view of an apparatus for carrying out a single crystal pulling method according to the present invention.

FIG. 2 is an explanatory view showing a single crystal pulling process by the single crystal pulling method according to the present invention.

FIG. 3 is an explanatory diagram of a solid layer forming process.

FIG. 4 is a graph showing the relationship between the occupied area of the lid and the oxygen atom concentration in the single crystal.

FIG. 5 is an electrical resistivity distribution diagram of a single crystal obtained by the method of the present invention and a single crystal obtained by a conventional method.

FIG. 6 is an explanatory diagram showing a single crystal pulling state by a conventional CZ method.

FIG. 7 is an explanatory diagram showing a single crystal pulling state by a conventional LDCZ method.

FIG. 8 is an explanatory diagram showing a supply path of oxygen atoms to a molten layer in the CZ method and the DLCZ method.

[Explanation of symbols]

 1 crucible 1a inner layer container 1b outer layer holding container 2 support shaft 7 pulling shaft 8 seed crystal 9 single crystal 10 chamber 10a pull chamber

Claims (2)

[Claims] 1. After melting all the raw materials for crystallization in the crucible,
In the state of leaving a molten layer in the upper layer in the crucible to form a solid layer by solidifying the crystallization raw material in the lower layer, while melting this solid layer, in the single crystal pulling method for pulling a single crystal from the molten layer, A method of pulling a single crystal, characterized in that, in the process of forming a solid layer as a lower layer in the crucible, the free surface of the molten layer is covered with a lid in contact with the free surface. 2. The method for pulling a single crystal according to claim 1, wherein an occupied area of the lid covering the free surface of the molten layer is 40% or more of the free surface.