JP6834831B2 - Method for manufacturing silicon single crystal - Google Patents

Description

The present invention relates to a method for producing a silicon single crystal.

The single crystal pulling device used in the Czochralski method is provided so as to surround the silicon single crystal above the crucible containing the silicon melt, and includes a heat shield that blocks heat radiated upward from the heater. ing. The heat shield is easily deteriorated because it is exposed to SiO gas that evaporates from the silicon melt. Therefore, studies have been made to suppress such deterioration (see, for example, Patent Document 1).
Patent Document 1 discloses that deterioration of the heat shield due to SiO gas is suppressed by using graphite coated with SiC on the outer shell of the heat shield.

Japanese Patent No. 3634867

However, in the configuration as in Patent Document 1, if the heat shield is used for a long period of time, the SiC coat may be worn and graphite may be exposed. In this case, since graphite is porous, particles (foreign substances) floating in the chamber collect in holes on the surface of the heat shield and fall into the silicon melt, resulting in twin crystals on the silicon single crystal. Defects and dislocations may occur.

An object of the present invention is to provide a method for producing a silicon single crystal capable of producing a high quality silicon single crystal.

The present inventor has obtained the following findings as a result of repeated diligent research.
A growing step of growing a silicon single crystal while introducing an inert gas into the main chamber, a separating step of separating the silicon single crystal from the silicon melt, and a cooling while pulling up the silicon single crystal separated from the silicon melt. It is common practice to perform a cooling step to produce a silicon single crystal. The inert gas flowing toward the silicon melt flows horizontally in the space between the lower end of the heat shield and the surface of the silicon melt (hereinafter, may be referred to as “first gas flow path”) and heats. It flows upward through the space outside the shield (hereinafter, may be referred to as the "second gas flow path"), and below the space outside the pit and the outside of the heater (hereinafter, the "third gas flow path"). It flows in the direction and is discharged. It is discharged to the outside of the main chamber.
Among such manufacturing steps, in the growing step, the height of the first gas flow path (hereinafter, may be referred to as “gap”) is adjusted in order to suppress the occurrence of defects in the silicon single crystal. (Hereinafter, it may be referred to as "gap adjustment"), but the gap adjustment is not performed in the process after the disconnection process.

Focusing on the flow of the inert gas in the step after the disconnection step, a vortex flow may occur in the first gas flow path. The vortex flow is a phenomenon in which a part of the inert gas flowing through the first gas flow path does not flow into the second gas flow path but swirls in the first gas flow path. When a vortex flow is generated, the SiO gas evaporated from the silicon melt accumulates in the first gas flow path, and even if the heat shield made of graphite is coated with SiC, the reaction of the following formula (1) Causes the SiC coat to wear. As a result, the porous graphite is exposed, and particles in the main chamber are accumulated in the exposed graphite holes.
SiC (s) + SiO (g) → 2Si (s) + CO (g)… (1)

Further, even when the heat shield is not coated with SiC, if a vortex flow is generated in the first gas flow path after the disconnection step, the heat shield is formed by the reaction of the following formula (2). The graphite wears and the holes become large, and particles tend to collect in the holes.
2C (s) + SiO (g) → SiC (s) + CO (g)… (2)

When using a heat shield with SiC coating, after the SiC coat is worn, when using a heat shield without SiC coating, silicon after the holes on the surface of the heat shield become large. When producing a single crystal, particles accumulated in holes on the surface of the heat shield are less likely to be discharged to the outside of the main chamber during the growth of the silicon single crystal. It is considered that the accumulated particles fall into the silicon melt and cause twin crystal defects and dislocations in the silicon single crystal.

On the other hand, if the inert gas passes through the second and third gas flow paths and is discharged to the outside of the main chamber without generating a vortex flow in the first gas flow path after the disconnection step, the silicon melt The SiO gas evaporated from the gas is also discharged along the flow of the inert gas without accumulating in the first gas flow path, the reactions (1) and (2) above do not occur, and the wear of the SiC coat and graphite can be suppressed. .. As a result, it is possible to suppress the accumulation of particles in the holes on the surface of the heat shield.
As a result of the study for suppressing the generation of the vortex flow in the first gas flow path in the process after the disconnection step, the present inventor has found that the generation of the vortex flow can be suppressed by appropriately adjusting the gap. The heading has completed the present invention.

That is, the method for producing a silicon single crystal of the present invention is a method for producing a silicon single crystal by the Chokralsky method, in which the crucible, the crucible driving unit for raising and lowering the crucible, and the crucible are heated to form a silicon melt. A heater for generating the crucible, a heat shield provided so as to surround a silicon single crystal above the crucible and formed of graphite, a main chamber accommodating the crucible, the heater and the heat shield, and the main By using a single crystal pulling device provided with an introduction part for introducing an inert gas into the chamber, the seed crystal is brought into contact with the silicon melt and then pulled while introducing the inert gas into the main chamber. The separation step includes a growing step of growing a silicon single crystal, a separating step of separating the silicon single crystal from the silicon melt, and a cooling step of pulling up and cooling the silicon single crystal separated from the silicon melt. In a later step, the distance between the lower end of the heat shield and the surface of the silicon melt is adjusted to 10 mm or more and 150 mm or less by moving the crucible up and down.

According to the present invention, the generation of the vortex flow of the inert gas in the first gas flow path is generated by adjusting the gap so that the gap of the first gas flow path in the step after the disconnection step is 150 mm or less. It can be suppressed and the inert gas can be discharged to the outside of the main chamber via the second and third gas flow paths. As a result, the reaction of the above formula (2) is suppressed, the wear of the surface of the heat shield, particularly the lower end is suppressed, and particles are less likely to accumulate on the surface. Further, by setting the gap of the first gas flow path to 10 mm or more, the gas flow in the first gas flow path is stabilized, so that the waviness of the surface of the silicon melt is suppressed, and the silicon melt and the heat shield are formed. The contact between the silicon melt and the heat shield can be suppressed even if the surface of the silicon melt may undulate for some reason (for example, an earthquake).
Therefore, it is possible to produce a high-quality silicon single crystal free of twin defects and dislocations.

In the method for producing a silicon single crystal of the present invention, the single crystal pulling device including a pull chamber connected to the upper part of the main chamber via a gate valve is used, and the cooling step is performed so that the silicon melt is solidified. The manufacturing method includes a low-temperature cooling step of cooling the silicon single crystal while adjusting the power of the heater, and the manufacturing method closes the gate valve when the silicon single crystal cooled in the low-temperature cooling step is housed in the pull chamber. It is preferable that the closing step is provided and the opening step of opening the main chamber after the closing step is provided, and in the step after the disconnecting step, the interval is adjusted to 40 mm or more by moving the pit up and down. ..

After the disconnection step, when the silicon melt is solidified by adjusting the heater in the low temperature cooling step, that is, when the temperature of the silicon melt in the pit is lowered by lowering the power of the heater and the silicon melt is solidified, the silicon melt is solidified. The object may expand and the surface may rise and come into contact with the heat shield.
According to the present invention, by setting the height of the first gas flow path in the gap adjustment to 40 mm or more, it is possible to suppress contact between the solidified material and the heat shield, and it is possible to suppress damage to the heat shield.
It is preferable that the height of the first gas flow path is 40 mm or more at least at the time when the surface of the silicon melt starts to solidify in the low temperature cooling step and after that.

In the method for producing a silicon single crystal of the present invention, the single crystal pulling device including a pull chamber connected to the upper part of the main chamber via a gate valve is used, and the cooling step is performed so that the silicon melt is solidified. The manufacturing method includes a low-temperature cooling step of cooling the silicon single crystal while adjusting the power of the heater, and the manufacturing method closes the gate valve when the silicon single crystal cooled in the low-temperature cooling step is housed in the pull chamber. A closing step and an opening step of opening the main chamber after the closing step are provided, and in the step after the disconnecting step, the silicon melt surface is solidified by adjusting the power of the heater. It is preferable to adjust the distance to 10 mm or more and 150 mm or less by moving the 坩 堝 up and down.

When the surface of the silicon melt solidifies, the SiO gas does not evaporate from the silicon melt, so that the reactions of the above formulas (1) and (2) do not occur even if a vortex flow is generated.
Therefore, wear of the surface of the heat shield can be suppressed.

In the method for producing a silicon single crystal of the present invention, the single crystal pulling device including a pull chamber connected to the upper part of the main chamber via a gate valve is used, and the cooling step is performed so that the silicon melt does not solidify. The manufacturing method includes a high-temperature cooling step of cooling the silicon single crystal while adjusting the power of the heater, and the manufacturing method closes the gate valve when the silicon single crystal cooled in the high-temperature cooling step is housed in the pull chamber. A recharging step of charging the crucible with a silicon raw material after the closing step and a step of growing the next silicon single crystal after the recharging step are provided for the closing step and the growing of the next silicon single crystal. In the step after the disconnection step, it is preferable to adjust the interval to 10 mm or more and 150 mm or less by moving the crucible up and down at least until the gate valve is closed.

According to the present invention, the generation of vortex flow can be suppressed in a long time from the disconnection step to the closing step, and the effect of suppressing wear on the surface of the heat shield can be enhanced.

In the method for producing a silicon single crystal of the present invention, it is preferable to use the heat shield having a surface coated with SiC.

According to the present invention, by using a heat shield in which SiC coating is applied and graphite is not exposed, it is possible to suppress the accumulation of particles on the surface of the heat shield, resulting in twin defects and dislocations of silicon single crystals. Can be suppressed.

The schematic diagram of the single crystal pulling apparatus which concerns on 1st and 2nd Embodiment of this invention. The explanatory view of the manufacturing method of the silicon single crystal in the 1st and 2nd embodiments. The figure which shows the result of Experiment 1 in the Example of this invention, and shows the relationship between the gap and the generation state of a vortex flow. The graph which shows the result of Experiment 3 in the said Example, and shows the relationship between the usage time of a heat shield and the adhesion state of particles.

[Embodiment]
[First Embodiment]
Hereinafter, the first embodiment of the present invention will be described with reference to the drawings.

{Structure of single crystal pulling device}
As shown in FIG. 1, the single crystal pulling device 1 is a device used in the CZ method (Czochralski method), and includes a pulling device main body 2 and a control unit 3.
The pulling device main body 2 includes a chamber 21, a crucible 22, a crucible driving unit 23, a heater 24, a pulling unit 25, and a heat shield 26.

The chamber 21 includes a main chamber 211 that houses a crucible 22, a heater 24, and a heat shield 26, and a pull chamber 213 that is connected to the upper part of the main chamber 211 via a gate valve 212. The pull chamber 213 is provided with a gas introduction port 213A as an introduction portion for introducing an inert gas such as Ar gas into the main chamber 211. A gas exhaust port 211A for discharging the gas in the main chamber 211 is provided below the main chamber 211.
The crucible 22 melts a polycrystalline silicon raw material to obtain a silicon melt M. The crucible 22 is configured to be able to rotate and move up and down at a predetermined speed by driving the crucible driving unit 23.

The heater 24 is arranged around the crucible 22 and melts the silicon raw material in the crucible 22.
The pull-up unit 25 includes a pull-up cable 251 to which a seed crystal is attached to one end, and a pull-up drive unit 252 that raises and lowers and rotates the pull-up cable 251.
The heat shield 26 is provided above the crucible 22 so as to surround the silicon single crystal SM, and is formed of graphite. A SiC coat 261 is applied to the surface of the heat shield 26. The thickness of the SiC coat 261 is preferably 40 μm or more and 200 μm or less. This is because if it is less than 40 μm, the SiC coat 261 is easily damaged, and if it exceeds 200 μm, the SiC coat 261 may be cracked. The heat shield 26 blocks radiant heat radiated upward from the heater 24.

{Manufacturing method of silicon single crystal}
Next, the method for producing a silicon single crystal in the first embodiment will be described. In the first embodiment, a method of opening the main chamber 211 for maintenance each time a single silicon single crystal is manufactured will be described.
It is preferable that the resistivity of the silicon single crystal produced in the first embodiment and the second embodiment described later is adjusted to a predetermined value by the dopant added to the silicon melt M. Further, the diameter of the straight body portion of the silicon single crystal after grinding the outer circumference may be any value such as 200 m, 300 mm, 450 mm.

First, the control unit 3 of the single crystal pulling device 1 sets the manufacturing conditions of the silicon single crystal SM, for example, the heating temperature, the Ar flow rate, the furnace pressure, the crucible 22 and the rotation speed of the silicon single crystal SM.
Next, the control unit 3 heats the crucible 22 to melt the solid silicon raw material in the crucible 22 and generate a silicon melt M. After that, the control unit 3 maintains the main chamber 211 communicated with the pull chamber 213 by opening the gate valve 212 in an inert atmosphere under reduced pressure, and while rotating the crucible 22, the neck portion SM1, the shoulder portion SM2, and the straight body. A silicon single crystal SM having a portion SM3 and a tail portion SM4 (see FIG. 2) is grown (growth step). At the time of this training, the control unit 3 receives the height GP of the first gas flow path R1 which is the gap between the lower end of the heat shield 26 and the liquid level of the silicon melt M (gap GP (lower end of the heat shield 26). The position of the crucible 22 is adjusted (gap is adjusted) so that the distance between the silicon melt M and the liquid level of the silicon melt M)) is substantially constant.

Next, as shown in FIG. 2, the control unit 3 separates the tail portion SM4 of the silicon single crystal SM from the silicon melt M (separation step), and pulls up the silicon single crystal SM separated from the silicon melt M. Cool (low temperature cooling step). In this low temperature cooling step, the control unit 3 reduces the power of the heater 24 so that the silicon melt M solidifies. The low temperature cooling step constitutes the cooling step of the present invention.
Then, when the cooled silicon single crystal SM is housed in the pull chamber 213, the control unit 3 closes the gate valve 212 (closed step). After that, the silicon single crystal SM is taken out from the pull chamber 213 (take-out step), and when the entire silicon melt M is solidified and the temperature in the main chamber 211 is sufficiently lowered, the main chamber 211 is opened (opening step). Perform maintenance.

In the step after the disconnection step, the control unit 3 moves the crucible 22 up and down until the solidified layer ML in which the surface of the silicon melt M is solidified is formed as shown by the alternate long and short dash line in FIG. Adjusts the gap so that the gap GP of the first gas flow path R1 is 10 mm or more and 150 mm or less. Further, after the solidified layer ML is formed, the gap is adjusted so that the gap GP is 40 mm or more and 150 mm or less.
By such gap adjustment, the inert gas GS is located outside the first gas flow path R1, the second gas flow path R2 located outside the heat shield 26, the outside of the crucible 22, and the outside of the heater 24. It passes through the gas flow path R3 of No. 3 and is discharged to the outside of the main chamber 211. Further, the SiO gas evaporating from the silicon melt M does not accumulate in the first gas flow path R1, but rides on the flow of the inert gas GS and passes through the second and third gas flow paths R2 and R3. It is discharged to the outside of the main chamber 211.

As shown by the alternate long and short dash line in FIG. 2, the gap adjustment to reduce the gap GP to 150 mm or less is performed until the solidified layer ML in which the surface of the silicon melt M is solidified is formed, or the entire silicon melt M is solidified. After that, the gap GP may be made larger than 150 mm. This is because after the solidified layer ML is formed or after the entire silicon melt M is solidified, the SiO gas does not evaporate and the reactions of the above formulas (1) and (2) do not occur.
Further, the opening step may be performed while maintaining the gap GP at 40 mm or more and 150 mm or less. Further, the gap GP may be made larger than 150 mm before forming the solidified layer ML.

{Action and effect of the first embodiment}
According to the first embodiment, the following effects can be obtained.
In order to adjust the gap so that the gap GP of the first gas flow path R1 in the step after the disconnection step is 150 mm or less, the inert gas GS is mainly used via the second and third gas flow paths R2 and R3. It can be discharged to the outside of the chamber 211, and the generation of the vortex flow GW of the inert gas GS in the first gas flow path R1 as shown by the alternate long and short dash line in FIG. 2 can be suppressed. As a result, the SiO gas from the silicon melt M can be suppressed from accumulating in the first gas flow path R1, and the reactions of the above formulas (1) and (2) can be suppressed. Therefore, the wear of the surface of the heat shield 26, particularly the SiC coat 261 at the lower end, is suppressed, and particles are less likely to accumulate on the surface.
Therefore, it is possible to produce a high-quality silicon single crystal SM that does not have twin defects or dislocations.

If the gap is adjusted so that the gap GP is 150 mm or less until after the solidified layer ML is formed or after the entire silicon melt M is solidified, the reactions of the above formulas (1) and (2) after the separation step are ensured. It is possible to further suppress the wear of the heat shield 26.

If the gap is adjusted so that the gap GP is 40 mm or more, even if the entire silicon melt M solidifies and expands, damage to the heat shield 26 due to contact with the solidified material can be suppressed.

Since the heat shield 26 coated with the SiC coat 261 is used, it is possible to suppress the accumulation of particles on the surface of the heat shield 26, and it is possible to suppress the occurrence of twin crystal defects and dislocations of the silicon single crystal SM.

[Second Embodiment]
Next, the second embodiment of the present invention will be described with reference to the drawings.
In the second embodiment, a method of producing a silicon single crystal after recharging the silicon raw material each time one silicon single crystal is produced using the same single crystal pulling device 1 as in the first embodiment will be described.
Unless otherwise specified, each step performed in the second embodiment is performed in the same manner as in the first embodiment.

First, after performing the same growing step and separating step as in the first embodiment, the control unit 3 cools while pulling up the silicon single crystal SM while controlling the power of the heater 24 so that the silicon melt M does not solidify. (High temperature cooling process). Then, after the closing step and the taking-out step are performed, the control unit 3 charges the crucible 22 with a silicon raw material in order to grow the next silicon single crystal SM (recharge step). The high temperature cooling step constitutes the cooling step of the present invention.
In the recharging step, first, the control unit 3 controls the power of the heater 24 to solidify only the surface of the silicon melt M, and then puts a solid silicon raw material into the crucible 22 using a raw material supply device (not shown). .. At this time, since the silicon raw material is charged onto the solidified surface, it is possible to prevent the silicon melt M from scattering and adhering to the inner peripheral surface of the crucible 22 at the time of charging. As a result, the deposits fall on the silicon melt M, and the occurrence of twin defects and dislocations of the silicon single crystal SM caused by the deposits can be suppressed.
After that, the control unit 3 controls the power of the heater 24 to melt the surface of the silicon melt M and the silicon raw material, and grows the next silicon single crystal.
After that, the separation step, the high temperature cooling step, the closing step, the taking-out step, the recharging step, and the growing step are repeated as necessary, and when the growing of the last silicon single crystal before the opening step is completed, the same as in the first embodiment. In addition, a separation process, a low temperature cooling process, a closing process, a taking-out process, and an opening process are performed.

When the high temperature cooling step is performed in the step after the disconnection step, the silicon melt M does not solidify and the surface of the solidified product does not rise. Therefore, in the control unit 3, the gap GP of the first gas flow path R1 is 10 mm. Adjust the gap so that it is 150 mm or less. Further, when the low temperature cooling step is performed, the control unit 3 adjusts the gap so that the gap GP is 10 mm or more and 150 mm or less until the solidified layer ML is formed, as in the first embodiment, and the solidified layer is formed. After the ML is formed, the gap is adjusted so that it is 40 mm or more and 150 mm or less.
Such gap adjustment is preferably performed at least until the gate valve 212 is closed in the closing step, after which the gap GP may be made larger than 150 mm.
Further, the take-out step may be performed while maintaining the gap GP at 40 mm or more and 150 mm or less. Further, the gap GP may be made larger than 150 mm before the closing step.

{Action and effect of the second embodiment}
According to the second embodiment, from the disconnection step to the closing step through the high temperature cooling step or the low temperature cooling step, the gap GP of the first gas flow path R1 is set to 40 mm or more and 150 mm or less from the disconnection step. The generation of eddy current GW can be suppressed in a long time until the closing step, and the effect of suppressing wear of the SiC coat 261 can be enhanced.

[Modification example]
The present invention is not limited to the above embodiment, and various improvements and design changes can be made without departing from the gist of the present invention.

For example, in the first and second embodiments, the heat shield 26 without the SiC coat 261 may be used. With such a configuration, it is possible to prevent the graphite constituting the heat shield 26 from being worn by the reaction of the above formula (2) and the holes on the surface of the heat shield 26 from becoming large. Therefore, particles are less likely to accumulate on the surface of the heat shield 26, and the occurrence of twin defects and dislocations due to the particles can be suppressed.

Next, the present invention will be described in more detail with reference to Examples, but the present invention is not limited to these examples.

[Experiment 1: Investigation of the relationship between the gap and the occurrence of eddy currents]
Using a gas flow simulation using a computer, the flow of the inert gas near the first gas flow path R1 was investigated after the separation step until the silicon melt solidified. At this time, as shown in FIG. 3, the gap GP was set to 280 mm (Comparative Example 1), 170 mm (Comparative Example 2), 150 mm (Example 1), and 100 mm (Example 2).
In Comparative Examples 1 and 2, while a part of the inert gas GS passes through the first and second gas flow paths R1 and R2 and is discharged from the main chamber, the vortex flow in the first gas flow path R1. It was found that GW occurs. On the other hand, in Examples 1 and 2, it was found that all the inert gas GS passed through the first and second gas flow paths R1 and R2 and was discharged from the main chamber.
From the above, it was confirmed that the generation of vortex flow GW can be suppressed by setting the gap GP to 150 mm or less.

[Experiment 2: Investigation of the relationship between the presence or absence of SiC coating in the heat shield, the gap, and the incidence of twin defects]
[Sample production]
{Comparative example 3}
A new heat shield was attached to the same device as the single crystal pulling device 1 of the first embodiment. The heat shield of Comparative Example 3 is composed only of graphite without SiC coating.
Then, a silicon single crystal was produced by the same production method as in the first embodiment. The gap adjustment was performed so that the gap of the first gas flow path was 280 mm after the disconnection step until the silicon melt solidified. Using one heat shield, 100 silicon single crystals were produced by the same method.

{Comparative example 4}
100 silicon single crystals were produced by the same method as in Comparative Example 3 except that a new heat shield having the same configuration as that in Comparative Example 3 was coated with SiC having a thickness of 60 μm.

{Examples 3 to 6}
Same as Comparative Example 3 except that a new heat shield having the same configuration as that of Comparative Example 3 was used and the gap of the first gas flow path in the gap adjustment was 150 mm in Example 3 and 100 mm in Example 4. By the method, 100 silicon single crystals were produced for each.
Same as Comparative Example 4 except that a new heat shield having the same configuration as that of Comparative Example 4 was used and the gap of the first gas flow path in the gap adjustment was 150 mm in Example 5 and 100 mm in Example 6. By the method, 100 silicon single crystals were produced for each.

[Evaluation]
The number of silicon single crystals in Comparative Example 3 in which twin defect occurred was investigated, and the value obtained by dividing the number in which twin defect occurred by the total number in Comparative Example 3 was obtained as the twin defect occurrence rate. .. In Comparative Example 4 and Examples 3 to 6, the twinning defect occurrence rate was determined in the same manner. Then, when the twinning defect occurrence rate of Comparative Example 3 is set to 100%, the ratios of Comparative Examples 3 and 4 and Examples 3 to 6 are standardized as the twinning defect occurrence rate (hereinafter, "standardized defect occurrence rate"). ”). The results are shown in Table 1.

As shown in Table 1, when a heat shield having no SiC coating is used, the standardized defect occurrence rate of Comparative Example 3 having a gap of more than 150 mm is 1 of Examples 3 and 4 having a gap of 150 mm or less. It was confirmed that it was more than 5.5 times.
Further, when a SiC-coated heat shield is used, the normalized defect occurrence rate of Comparative Example 4 having a gap of more than 150 mm is more than twice that of Examples 5 and 6 having a gap of 150 mm or less. It could be confirmed.
The reason why the normalized twin defect occurrence rate of Comparative Examples 3 and 4 is higher than that of Examples 3 to 6 is considered as follows.

In Comparative Examples 3 and 4, a vortex flow is generated in the first gas flow path between the heat shield and the surface of the silicon melt, and the vortex flow is generated, as in the case of Comparative Example 1 shown in FIG. SiO gas evaporated from the silicon melt accumulates in the region. Then, the accumulated SiO gas causes the reaction of the above formula (2) in Comparative Example 3 and the reaction of the above formulas (1) and (2) in Comparative Example 4, and the SiC coat and the surface of the heat shield are worn. , Particles in the main chamber tend to collect in the holes on the surface.

On the other hand, in Examples 3 to 6, similar to the drawings of Examples 1 and 2 shown in FIG. 3, the vortex flow as in Comparative Examples 3 and 4 does not occur, so that between the heat shield and the surface of the silicon melt. It is suppressed that SiO gas is accumulated in the first gas flow path of the above. As a result, the reactions of the above formulas (1) and (2) are less likely to occur, wear of the surface of the SiC coat and the heat shield is suppressed, and particles are less likely to accumulate on the surface. As a result, it is considered that twin defect is less likely to occur.

From the above, it was confirmed that twinning defects are less likely to occur by setting the gap to 150 mm or less in the step after the cutting step.
Further, since there was no difference in the normalization defect occurrence rate between Example 3 and Example 4 or between Experimental Example 5 and Experimental Example 6, if the gap is 150 mm or less, the size thereof. Regardless, it was confirmed that the difficulty of twinning defects did not change much.
Further, when Comparative Example 3 and Comparative Example 4 were compared, the twin crystal defect occurrence rate was higher in Comparative Example 3 in which the SiC coating was not applied. From this, it was confirmed that the SiC coat has an effect of suppressing the occurrence of twinning defects.

[Experiment 3: Investigation of the relationship between the usage time of the heat shield and the state of particle adhesion]
{Comparative example 5}
Using the new thermal shield of Comparative Example 5 having the same configuration as that of Comparative Example 4, a plurality of silicon single crystals were produced under the same conditions as in Comparative Example 4 (gap is 280 mm). Then, the amount of adhesion of particles was measured at 16 points along the circumferential direction of the lower end of the heat shield at each predetermined elapsed time from the start of using the heat shield of Comparative Example 5. For the measurement of particles, an aerial particle counter (model KC-03A) manufactured by RION was used, and the measurement target was particles of 0.3 μm or more.

{Example 7}
Using the new heat shield of Example 7 having the same configuration as that of Comparative Example 5, a plurality of Comparative Example silicon single crystals were produced under the same conditions as in Comparative Example 5 except that the gap was set to 80 mm. Then, as in Comparative Example 5, the amount of particles adhering was measured at predetermined elapsed times from the start of using the heat shield.

{Evaluation}
FIG. 4 shows the relationship between the usage time of the heat shield of Comparative Examples 5 and 7 and the amount of particles adhered to the heat shield. From this result, in Example 7, the ^{number of particles was 6 or less per 13 cm 2} even if the usage time exceeded 10,000 hours, whereas in Comparative Example 5, the average usage time was almost the same as that of Example 7. There were about 2500 pieces per 13 cm ^2.
Further, when the lower ends of the heat shields of Comparative Examples 5 and 7 at the above-mentioned usage time were confirmed, discoloration was observed in Comparative Example 5, but no discoloration was observed in Example 7.
From this, if the gap is made larger than 150 mm in the step after the separation step, the SiC coat is worn by the vortex flow, and the wear tends to cause particles to collect, and the collected particles fall into the silicon melt. Therefore, it is considered that twin defect and dislocation are likely to occur.
On the other hand, when the gap is set to 150 mm or less, vortex flow does not occur, so that the wear of the SiC coat is suppressed, particles are less likely to accumulate in the heat shield, and twin defect or presence is caused by the fall of particles into the silicon melt. It is considered that dislocation is less likely to occur.

1 ... Single crystal pulling device, 22 ... Crucible, 23 ... Crucible drive, 24 ... Heater, 26 ... Heat shield, 211 ... Main chamber, 211A ... Gas exhaust port, 212 ... Gate valve, 213 ... Pull chamber, 213A ... Gas Introduction port (introduction part), 261 ... SiC coat, GS ... inert gas, M ... silicon melt, SM ... silicon single crystal.

Claims (5)

A method for producing a silicon single crystal by the Czochralski method.
Crucible and
The crucible drive unit that raises and lowers the crucible,
A heater that heats the crucible to produce a silicon melt,
A heat shield provided above the crucible so as to surround a silicon single crystal and formed of graphite,
A main chamber accommodating the crucible, the heater and the heat shield,
Using a single crystal pulling device including an introduction part for introducing the inert gas into the main chamber,
A growing step of growing a silicon single crystal by bringing the seed crystal into contact with the silicon melt and then pulling it up while introducing an inert gas into the main chamber.
A separation step of separating the silicon single crystal from the silicon melt, and
It is provided with a cooling step of cooling while pulling up a silicon single crystal separated from the silicon melt.
In the step after the separation step, the distance between the lower end of the heat shield and the surface of the silicon melt is adjusted to 10 mm or more and 150 mm or less by moving the crucible up and down, and the lower end of the heat shield and the lower end of the heat shield are adjusted. A method for producing a silicon single crystal, which comprises suppressing the generation of a vortex flow of the inert gas in the first gas flow path which is a gap between the silicon melt and the liquid surface. In the method for producing a silicon single crystal according to claim 1,
Using the single crystal pulling device including a pull chamber connected to the upper part of the main chamber via a gate valve.
The cooling step includes a low-temperature cooling step of cooling the silicon single crystal while adjusting the power of the heater so that the silicon melt solidifies.
The manufacturing method is
A closing step of closing the gate valve when the silicon single crystal cooled in the low temperature cooling step is housed in the pull chamber.
It is provided with an opening step of opening the main chamber after the closing step.
A method for producing a silicon single crystal, which comprises adjusting the distance to 40 mm or more by moving the crucible up and down in a step after the cutting step. In the method for producing a silicon single crystal according to claim 1 or 2.
Using the single crystal pulling device including a pull chamber connected to the upper part of the main chamber via a gate valve.
The cooling step includes a low-temperature cooling step of cooling the silicon single crystal while adjusting the power of the heater so that the silicon melt solidifies.
The manufacturing method is
A closing step of closing the gate valve when the silicon single crystal cooled in the low temperature cooling step is housed in the pull chamber.
It is provided with an opening step of opening the main chamber after the closing step.
In the step after the disconnection step, the interval is adjusted to 10 mm or more and 150 mm or less by moving the crucible up and down until the surface of the silicon melt is solidified by adjusting the power of the heater. A characteristic method for producing a silicon single crystal. In the method for producing a silicon single crystal according to claim 1 or 2.
Using the single crystal pulling device including a pull chamber connected to the upper part of the main chamber via a gate valve.
The cooling step includes a high-temperature cooling step of cooling the silicon single crystal while adjusting the power of the heater so that the silicon melt does not solidify.
The manufacturing method is
A closing step of closing the gate valve when the silicon single crystal cooled in the high temperature cooling step is housed in the pull chamber.
In order to grow the next silicon single crystal, a recharge step of charging the crucible with a silicon raw material after the closing step and a recharge step of charging the crucible with a silicon raw material.
After the recharge step, the step of growing the next silicon single crystal is provided.
A method for producing a silicon single crystal, which comprises adjusting the interval to 10 mm or more and 150 mm or less by moving the crucible up and down at least until the gate valve is closed in the step after the disconnection step. .. In the method for producing a silicon single crystal according to any one of claims 1 to 4.
A method for producing a silicon single crystal, which comprises using the heat shield having a surface coated with SiC.