JP2010018446A - Method for producing single crystal and single crystal pulling apparatus - Google Patents

Abstract

[Object] To obtain a single crystal having a high dislocation-free rate while suppressing variation in resistivity with respect to crystal length, without increasing the consumption of inert gas, and without increasing the operation time. A method for producing a single crystal is provided.
An inert gas G flowing from above to below is supplied into a radiation shield 6 disposed above the crucible 3 so as to surround a single crystal pulling region, and the single crystal C has a predetermined length. The inert gas G introduced into the crucible 3 from below the radiation shield 6 until it grows to the height of the space by the rectifying plate 7 provided in the space between the radiation shield 6 and the silicon melt surface M1. When the single crystal C exceeds the predetermined length, the rectifying plate 7 is brought into contact with the radiation shield 6, and the crucible 3 is formed from below the radiation shield 6. The step of flowing the inert gas G led in along the one flow path formed between the rectifying plate 7 and the silicon melt surface M1 is executed.
[Selection] Figure 1

Description

  The present invention relates to a method for producing a single crystal and a single crystal pulling apparatus, which are pulled up while growing a single crystal by the Czochralski method (hereinafter referred to as “CZ method”).

  The CZ method is widely used for the growth of silicon single crystals. In this method, the seed crystal is brought into contact with the surface of the silicon melt contained in the crucible, the crucible is rotated, and the seed crystal is pulled upward while rotating in the opposite direction. In this way, a single crystal is formed.

  As shown in FIG. 6, in the pulling method using the conventional CZ method, first, raw silicon is loaded into a quartz glass crucible 51 and heated by a heater 52 to obtain a silicon melt M. Thereafter, the seed crystal P attached to the pulling wire 50 is brought into contact with the silicon melt M to pull up the silicon crystal C.

  In general, prior to the start of pulling, after the temperature of the silicon melt M is stabilized, as shown in FIG. 7, necking is performed in which the seed crystal P is brought into contact with the silicon melt M to dissolve the tip of the seed crystal P. Necking is an indispensable process for removing dislocations generated in a silicon single crystal due to a thermal shock generated by bringing the seed crystal P into contact with the silicon melt M. The neck portion P1 is formed by this necking. The neck portion P1 is generally required to have a diameter of 3 to 4 mm and a length of 30 to 40 mm or more.

In addition, as a process after the start of pulling, after necking is completed, a crown process for expanding the crystal to the diameter of the straight body part, a straight body process for growing a single crystal as a product, and a single crystal diameter after the straight body process are gradually reduced. The tail process is performed.
Note that an inert gas G is supplied into the furnace body in which the crucible 51 is housed from the upper side to the lower side for the purpose of removing foreign substances in the furnace body atmosphere and adjusting oxygen evaporation.
In addition, the rectifying action of the inert gas G is also performed, and the single crystal C is lifted above and in the vicinity of the quartz glass crucible 51 in order to block the extra radiant heat from the heater 52 and the like to the growing single crystal C. A radiation shield 53 is provided so as to surround the region.

By the way, in a single crystal growth process using such a single crystal pulling apparatus, a dopant (additive) such as boron, phosphorus, antimony, or arsenic is added to the silicon melt to give a predetermined resistivity to the single crystal. Processing is performed.
Among the dopants, for example, antimony and arsenic, which have high vapor pressure, are introduced after melting the raw material silicon, but continue to evaporate from the melt surface during single crystal growth.

  In Patent Document 1, it is noted that the evaporation of antimony, which is particularly easy to evaporate when pulled under reduced pressure, is greatly affected by the flow rate and flow direction of the inert gas sprayed on the silicon melt surface. A method for producing a single crystal is disclosed in which the concentration of antimony at the outer periphery of the single crystal is suppressed by controlling the gas flow rate and the flow direction.

FIG. 8 shows a cross-sectional view of the main part of the single crystal pulling apparatus disclosed in Patent Document 1. As shown in the figure, a frustoconical gas rectifying plate 62 is provided above the silicon melt M in the crucible 60 and below the radiation shield 61. The gas sprayed onto the surface M1 is kept away from the single crystallization progress part C1.
With such a configuration, the vicinity of the single crystallization proceeding portion C1 immediately below the gas rectifying plate 62 is a portion where the gas flow is stagnant. For this reason, the evaporation of antimony from the silicon melt M increases as it approaches the crucible side, so that the antimony concentration in the vicinity of the outer peripheral surface of the single crystal C does not decrease.
Japanese Patent Publication No. 3-7637

  By the way, in the pulling of the single crystal from the silicon melt into which the dopant is introduced as described above, when the crystal growth proceeds and the amount of the melt in the crucible decreases, the dopant in the silicon melt is caused by segregation. It is known that there is a tendency that the concentration of the dopant increases and the dopant concentration of the single crystal decreases (becomes a low resistance crystal).

However, when the single crystal becomes low resistance in the latter stage of growth and the dopant concentration in the silicon melt rises excessively, compositional supercooling occurs at the growth interface, and the single crystal to be grown tends to dislocation. was there.
In addition, the resistivity of the single crystal is increased in the initial stage of growth, and the resistivity of the single crystal is decreased in the later stage of the growth.

  In the single crystal pulling apparatus disclosed in Patent Document 1, the gas rectifying plate 62 suppresses a decrease in the dopant (antimony) concentration around the single crystal C during the pulling process, but the silicon melt that changes during pulling is changed. It could not cope with an excessive increase in the dopant concentration in the liquid. That is, in the latter stage of single crystal growth, compositional supercooling occurred at the growth interface, and the crystal might be dislocated.

In order to reduce the dopant concentration in the silicon melt, the gas flow rate in the furnace is increased to accelerate the evaporation of the dopant, or the single crystal pulling rate is decreased to improve the single crystal dopant concentration. A method is conceivable.
However, the former has a problem that the consumption of the inert gas is greatly increased, and the latter has a long operation time.

  The present invention has been made under the circumstances as described above. In the method for producing a single crystal in which the single crystal is pulled from the crucible by the Czochralski method, variation in resistivity with respect to the crystal length is suppressed, and no dislocation is generated. To provide a single crystal manufacturing method and a single crystal pulling apparatus that can obtain a single crystal with a high conversion rate, do not increase the consumption of inert gas, and do not increase the operation time. Objective.

In order to solve the above problems, a method for producing a single crystal according to the present invention is a method for producing a single crystal by pulling a single crystal from a silicon melt melted in a crucible by a Czochralski method. An inert gas flowing from the upper side to the lower side is supplied into the radiation shield arranged above the crucible so as to surround the crystal pulling region, and the radiation shield is used until the single crystal grows to a predetermined length. The inert gas led into the crucible from below is formed by dividing the space in the height direction by two baffle plates provided in the space between the radiation shield and the silicon melt surface. A step of flowing along the flow path, and when the single crystal exceeds the predetermined length, the rectifying plate is brought into contact with the radiation shield, and an inert gas led into the crucible from below the radiation shield ,in front Characterized in that run and flowing along one of the flow passage formed between the rectifier plate and the silicon melt surface.
The predetermined length of the single crystal when the flow path of the inert gas led out from the lower side of the radiation shield into the crucible is changed is preferably at least 400 mm.
The distance between the lower end of the radiation shield and the silicon melt surface is preferably at least 25 mm.

In this way, the flow of the inert gas is slowed by using two flow paths for the inert gas flowing on the silicon melt until the single crystal has a predetermined length (initial stage of growth). When the single crystal exceeds a predetermined length (later growth stage), the flow of the inert gas is made high-speed by using the flow of the inert gas as one flow path.
By such control, the evaporation of the dopant from the silicon melt is reduced at the initial stage of growth, and the increase in resistivity at the initial stage of growth is suppressed. For this reason, variation in resistivity with respect to the crystal length can be reduced.

Moreover, evaporation of the dopant from the silicon melt is promoted in the later stage of growth, and an excessive increase in the dopant concentration of the silicon melt is suppressed. Furthermore, since the flow rate of the inert gas between the single crystal and the silicon melt increases in the later stage of growth, the temperature gradient increases, thereby suppressing compositional supercooling just below the solid-liquid interface and causing dislocations. It can be difficult.
In addition, since the flow rate of the inert gas supplied into the furnace is constant throughout the growth of the single crystal, the increase in the amount of inert gas consumption is suppressed, and there is no need to reduce the pulling speed. Time can be avoided.

In order to solve the above-mentioned problems, a single crystal pulling apparatus according to the present invention is a single crystal pulling apparatus for pulling a single crystal from a silicon melt in a crucible by the Czochralski method. A radiation shield having an upper portion and a lower portion formed so as to surround the periphery of the single crystal, an inert gas supply means for supplying an inert gas from the upper side to the lower side of the radiation shield, and the radiation shield And a flow straightening plate elevating means for moving the flow straightening plate up and down relative to the radiation shield.
With such an apparatus, the method for producing the single crystal can be carried out, and the effects can be obtained.

  According to the present invention, in the method for producing a single crystal in which the single crystal is pulled from the crucible by the Czochralski method, variation in resistivity with respect to the crystal length is suppressed, and a single crystal having a high dislocation rate is obtained and inert. A single crystal manufacturing method and a single crystal pulling apparatus can be obtained without increasing gas consumption and without increasing the operation time.

  Embodiments of a method for producing a single crystal and a single crystal pulling apparatus according to the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the overall configuration of a single crystal pulling apparatus 1 according to the present invention. The single crystal pulling apparatus 1 includes a furnace body 2 formed by superposing a pull chamber 2b on a main chamber 2a, a crucible 3 provided in the furnace body 2, and a semiconductor raw material (raw material) loaded in the crucible 3 A heater 4 for melting silicon (M) and a pulling mechanism 5 for pulling up a single crystal C to be grown are provided.

The crucible 3 has a double structure, and is composed of a quartz glass crucible 3a on the inside and a graphite crucible 3b on the outside. The pulling mechanism 5 includes a winding mechanism 5a driven by a motor and a pulling wire 5b wound up by the winding mechanism 5a, and a seed crystal P is attached to the tip of the wire 5b.
In the main chamber 2a, upper and lower openings are formed above and in the vicinity of the crucible 3 so as to surround the pulling region of the single crystal C, and extra radiant heat from the heater 4 and the like is formed on the growing single crystal C. A radiation shield 6 is provided so as not to give the noise.

  Further, an inert gas supply port 13 (inert gas supply means) is provided above the pull chamber 2b, and an inert gas discharge port 14 is provided on the bottom surface of the main chamber 2a. That is, the inert gas G supplied from the inert gas supply port 13 flows downward from above, passes through the single crystal C in the quartz glass crucible 3a, and is then discharged from the inert gas discharge port 14. It is made like that.

An annular rectifying plate 7 is suspended by a wire 12a in the space between the radiation shield 6 and the surface of the silicon melt M. The rectifying plate 7 can be moved up and down by winding or unwinding the wire 12a by a motor driven winding mechanism 12 (rectifying plate elevating means).
The rectifying plate 7 is made of, for example, carbon, the lower surface is substantially parallel to the melting surface M1, and the upper surface is shaped along the shield shape.

By this vertical movement control of the rectifying plate 7, one of the following two states is formed in the space above the silicon melt surface M1 during the single crystal pulling process.
In the first state, as shown in FIG. 2, the rectifying plate 7 is separated from the radiation shield 6, and between the rectifying plate 7 and the radiation shield 6, and between the rectifying plate 7 and the silicon melt surface M1. Inert gas G flow paths are formed between the two.
In that case, as shown by the arrow in FIG. 2, the flow of the inert gas G, the inert gas G flowing down in the radiation shield 6 is led into the crucible 3 from below the radiation shield 6, It is made to flow through two flow paths divided in the vertical direction (referred to as a two-system configuration).

Further, as shown in FIG. 3, the second state is a state in which the rectifying plate 7 is in contact with the lower end portion of the radiation shield 6, and the inert gas G is only between the silicon melt surface M <b> 1 and the rectifying plate 7. A flow path is formed.
In this case, as shown in FIG. 3, the inert gas G flowing down the radiation shield 6 passes through one flow path formed between the rectifying plate 7 and the silicon melt surface M1, and the crucible. 3 is configured to flow outside (referred to as a one-system configuration).

As shown in FIG. 1, the single crystal pulling apparatus 1 includes a heater control unit 9 that controls the amount of power supplied to the heater 4 that controls the temperature of the silicon melt M, a motor 10 that rotates the crucible 3, and a motor. And a motor control unit 10a for controlling the number of rotations of ten. Also, an elevating device 11 for controlling the height of the crucible 3, an elevating device control unit 11 a for controlling the elevating device 11, and a wire reel rotating device control unit 16 for controlling the pulling speed and the number of rotations of the grown crystal are provided. Yes. Furthermore, a rectifying plate raising / lowering control unit 15 that controls the height position of the rectifying plate 7 via the winding mechanism 12 of the wire 12a is provided.
Each of these control units 9, 10 a, 11 a, 12, 15, 16 is connected to an arithmetic control device 8 b of the computer 8.

In the single crystal pulling apparatus 1 configured as described above, the raw material silicon M is first loaded into the crucible 3, and the crystal growth process is started as follows based on the program stored in the storage device 8a of the computer 8. Is done.
First, the heater control unit 9 is operated by a command from the arithmetic control device 8b to heat the heater 4, and the raw material silicon M in the crucible 3 is melted.
Further, a dopant such as antimony or arsenic is added and added to the silicon melt M in order to make the single crystal have a desired resistivity.

  Furthermore, the motor control unit 10a, the lifting / lowering device control unit 11a, and the wire reel rotating device control unit 16 are operated by the command of the arithmetic control device 8b, the crucible 3 is rotated, and the winding mechanism 5a is operated to operate the wire. 5b is taken down. Then, the seed crystal P attached to the wire 5b is brought into contact with the silicon melt M, and necking for melting the tip of the seed crystal P is performed to form the neck portion P1.

Thereafter, the pulling conditions are adjusted by parameters of the power supplied to the heater 4 and the single crystal pulling speed according to the command of the arithmetic and control unit 8b, and single crystal pulling such as crown process, straight body process, tail part process, etc. The steps are performed in order.
Further, the inert gas G is supplied from the upper side to the lower side in the furnace at a constant flow rate (for example, 75 L / min) throughout the single crystal growth process. Here, the flow rate of the inert gas G is set according to the resistivity to be given to the single crystal C to be grown.

In the initial stage of pulling up the single crystal C until the straight body portion of the single crystal C grows to a predetermined length (for example, the length of the straight body portion is between 0 mm and 400 mm), it is shown in FIG. Thus, the height position of the rectifying plate 7 is controlled by the winding mechanism 12 so that the flow path of the inert gas G in the space above the silicon melt surface M1 has a two-system configuration.
That is, when the flow path of the inert gas G is branched into two in the height direction, the gas flow rate becomes low, and the evaporation of the dopant from the silicon melt M is not promoted. Thereby, the resistivity rise of the single crystal C in the initial stage of growth is suppressed.
Note that the gap dimension between the radiation shield 6 and the silicon melt surface M1 is, for example, at least 25 mm. That is, an increase in the flow rate due to a narrow gap size of the gas flow path is avoided.

Then, when the growing single crystal C reaches a predetermined height in the late growth stage (for example, the length of the straight body portion is between 400 mm and 750 mm), as shown in FIG. The height position of the rectifying plate 7 is controlled by the winding mechanism 12 so that the flow path of the inert gas G in the space has one system configuration (the state is in contact with the radiation shield 6).
That is, when the flow path of the inert gas G becomes one (from two) in the height direction, the gas flow rate increases, and evaporation of the dopant from the silicon melt M is promoted. Thereby, the excessive raise of the dopant density | concentration in the silicon melt M in the late stage of a growth is suppressed.
The gap dimension between the lower end of the radiation shield 6 and the silicon melt surface M1 is, for example, 30 mm.

  Further, in the latter stage of single crystal growth, an excessive increase in the dopant concentration in the silicon melt M in the later stage of growth is suppressed, and the flow rate of the inert gas G in the space above the silicon melt surface M1 is suppressed as described above. It becomes faster, so that the temperature gradient between the single crystal C and the silicon melt M becomes larger. Therefore, compositional supercooling directly under the solid-liquid interface is suppressed, and a state in which dislocation is difficult to occur is achieved.

As described above, according to the embodiment of the present invention, the flow of the inert gas G flowing through the space above the silicon melt surface M1 is divided into two flows until the single crystal C has a predetermined length (initial growth stage). By using the path, the flow of the inert gas G is made at a low speed. When the single crystal C exceeds a predetermined length (later growth stage), the flow of the inert gas G is made high-speed by using the flow of the inert gas G as one flow path.
By such control, evaporation of the dopant from the silicon melt M is reduced at the initial stage of growth, and an increase in resistivity at the initial stage of growth is suppressed. For this reason, variation in resistivity with respect to the crystal length can be reduced.

In addition, evaporation of the dopant from the silicon melt M is promoted in the later stage of growth, and an excessive increase in the dopant concentration of the silicon melt M is suppressed. Furthermore, since the flow rate of the inert gas G between the single crystal C and the silicon melt M in the later stage of growth is increased, the temperature gradient is increased, thereby suppressing compositional supercooling immediately below the solid-liquid interface, A state in which dislocation is difficult to occur can be achieved.
Moreover, since the flow rate of the inert gas G supplied into the furnace is a constant amount during the growth of the single crystal, the increase in the amount of inert gas consumption is suppressed, there is no need to reduce the pulling speed, and the operation time is long. Time can be avoided.

In the initial stage of single crystal C growth, as shown in FIG. 4, the radiant heat (indicated by a one-dot chain line) from the heater 4 and the crucible 3 passes linearly through the gap between the radiation shield 6 and the rectifying plate 7. It is preferable to form a gas flow path in the gap so as not to block (that is, to block radiant heat to the single crystal C).
With such a configuration, since the radiant heat from the heater 4 and the crucible 3 to the single crystal C is blocked, the single crystal is also in the initial stage of growth when the flow rate of the inert gas G in the space above the silicon melt surface M1 is reduced. The temperature gradient between C and the silicon melt M is increased, compositional supercooling immediately below the solid-liquid interface is suppressed, and dislocations are hardly formed.

  Next, the method for producing a single crystal and the single crystal pulling apparatus according to the present invention will be further described based on examples. In this example, the effect was verified by actually performing an experiment using the single crystal pulling apparatus having the configuration described in the above embodiment.

In this experiment, a silicon single crystal having a diameter of 200 mm was grown several times (10 times), and the dislocation-free rate and resistivity of the pulled single crystal were verified.
Arsenic was used as a dopant for achieving a desired resistivity. The flow rate of gas supplied into the furnace was 75 L / min, and the furnace pressure was fixed at 150 Torr.
In addition, when the length of the straight body of the single crystal is 0 mm or more and less than 400 mm (in the initial stage of growth), the gas flow path has two systems as shown in FIG. 2, and the length of the straight body is 400 mm or more and less than 750 mm. During the period (late growth period), as shown in FIG.
The distance between the silicon melt and the radiation shield was set to 25 mm when the gas flow path had a two-line configuration, and set to 30 mm when the gas flow path had a single-line configuration.

As a comparative example, in a conventional configuration that does not use a rectifying plate, when the length of the straight body of the single crystal is 0 mm or more and less than 400 mm (initial stage of growth), the flow rate of the inert gas is 75 L / min. The inert gas flow rate was set to 100 L / min while the length was 400 mm or more and less than 750 mm (late growth period). The other conditions were the same as in the example, and the dislocation-free rate and resistivity were verified.
As a result of Examples and Comparative Examples, Table 1 shows the dislocation-free rate of the single crystal pulled up, and the graph of FIG. 5 shows the resistivity (Ω · cm) with respect to the crystal length (mm).

As a result of the experiment, as shown in Table 1, in the example, a high dislocation-free rate was obtained in all of the 10 runs. However, in the comparative example, there was a variation between when the dislocation-free rate was high and when it was low.
Further, as shown in the graph of FIG. 5, in the example, since the resistivity at the initial stage of growth was suppressed to a low level, the variation in resistivity with respect to the crystal length could be reduced. However, in the case of the comparative example, the variation in resistivity with respect to the crystal length greatly spread.

  From the experimental results of the above examples, by using the method for producing a single crystal and the single crystal pulling apparatus of the present invention, it is possible to obtain a single crystal having a high dislocation-free rate while suppressing variations in resistivity with respect to the crystal length. It was confirmed that the consumption of the inert gas was not increased and the operation time was not prolonged.

  The present invention relates to a single crystal manufacturing method and a single crystal pulling apparatus for pulling a single crystal by the Czochralski method, and is suitably used in the semiconductor manufacturing industry and the like.

FIG. 1 is a block diagram schematically showing the configuration of a single crystal pulling apparatus according to the present invention. FIG. 2 is a cross-sectional view of a radiation shield and a current plate included in the single crystal pulling apparatus of FIG. FIG. 3 is a cross-sectional view of a radiation shield and a current plate included in the single crystal pulling apparatus of FIG. FIG. 4 is a cross-sectional view showing another embodiment of the radiation shield and the current plate that the single crystal pulling apparatus of FIG. 1 has. FIG. 5 is a graph showing the results of the example. FIG. 6 is a diagram for explaining a conventional pulling method using the CZ method in a single crystal pulling apparatus having a radiation shield. FIG. 7 is a diagram for explaining formation of a neck portion in a pulling method using a conventional CZ method. FIG. 8 is a cross-sectional view of the main part of a conventional single crystal pulling apparatus for controlling the dopant concentration.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Single crystal pulling apparatus 2 Furnace body 2a Main chamber 2b Pull chamber 3 Crucible 3a Quartz glass crucible 3b Graphite crucible 4 Heater 5 Pulling mechanism 6 Radiation shield 7 Rectifier plate 8 Computer 8a Storage device 8b Arithmetic storage device 12 Winding mechanism (rectifier plate Elevating means)
13 Inert gas supply port (inert gas supply means)
14 Inert gas outlet C Single crystal G Inert gas M Raw material silicon, silicon melt P Seed crystal P1 Neck

Claims (4)

A method for producing a single crystal by pulling a single crystal from a silicon melt melted in a crucible by a Czochralski method,
Into the radiation shield disposed above the crucible so as to surround the single crystal pulling region, supplying an inert gas flowing from above to below,
Until the single crystal grows to a predetermined length, the inert gas led into the crucible from below the radiation shield is caused by a rectifying plate provided in a space between the radiation shield and the silicon melt surface. Flowing along two flow paths formed by dividing the space in the height direction;
When the single crystal exceeds the predetermined length, the rectifying plate is brought into contact with the radiation shield, and the inert gas led into the crucible from below the radiating shield is supplied to the rectifying plate and the silicon melt surface. And a step of flowing along one flow path formed between the two and a single crystal manufacturing method.   The predetermined length of the single crystal when the flow path of the inert gas led into the crucible from below the radiation shield is changed is at least 400 mm. A method for producing a single crystal.   The method for producing a single crystal according to claim 1 or 2, wherein a distance between the lower end of the radiation shield and the silicon melt surface is at least 25 mm. In a single crystal pulling apparatus that pulls a single crystal from the silicon melt in the crucible by the Czochralski method,
A radiation shield provided with upper and lower openings formed so as to surround the periphery of the single crystal above the crucible;
An inert gas supply means for supplying an inert gas from the upper side to the lower side of the radiation shield;
A rectifying plate provided between the radiation shield and the silicon melt surface;
A single crystal pulling apparatus comprising a rectifying plate lifting / lowering means for moving the rectifying plate up and down relative to the radiation shield.

Images