TWI602958B - Manufacturing method of single-crystalline silicon - Google Patents

Description

矽Single crystal manufacturing method

The present invention relates to a method for producing a single crystal, and more particularly to a necking process which is one of the steps of cultivating a single crystal using a Czochralski method (hereinafter referred to as "CZ method").

Many single crystals are produced by the CZ method. In the CZ method, it is carried out in sequence: a liquid-contacting process to make the crystallized contact with the mash; the necking process, which reduces the crystal diameter by the so-called dash neck method; the shoulder cultivation process, the single The diameter of the crystal is gradually widened to reach the desired diameter; and the body cultivation process is continued while cultivating the single crystal while maintaining the crystal diameter as a whole.

The necking process is a necessary process for removing a high-density slippage row introduced into the seed crystal by thermal shock when the seed crystal is contacted with the melt. The slippage row can be removed from the single crystal by cultivating a single crystal of a certain length which reduces the crystal diameter to about 5 to 6 mm.

Various necking methods are disclosed in Patent Documents 1 to 5. For example, in the method disclosed in Patent Document 1, the shape of the solid-liquid interface between the mash and the single crystal is deformed in the necking process to make the difference in the difficulty of removal or to progress to the outside of the radial direction of the single crystal. It is made into a ring shape and is removed by elimination. Further, in the method disclosed in Patent Document 2, a single crystal having a crystal axis orientation of [110] is grown, and a necking operation for reducing the diameter of the seed crystal is performed to form a first neck portion, and a shoulder is formed. The shape of the solid-liquid interface is changed from the upper convex to the lower convex portion before the front portion, and then the necking operation is performed again to form the second neck portion, thereby removing the axially-shaped row remaining in the central axis portion of the neck portion.

The method disclosed in Patent Document 3 is that the diameter-increased portion which is formed by increasing the diameter of the neck portion and increasing the diameter of the neck portion is formed by continuously forming a plurality of times in the final stage of the necking process. Thereby, the shape of the solid-liquid interface is frequently changed, and the chance of changing the moving direction of the difference row is given several times, and the axial difference row is efficiently removed. The method disclosed in Patent Document 4 is also to remove the slip difference row by forming the tapered portion and the enlarged diameter portion in the necking process.

The method disclosed in Patent Document 5 is to reduce the cultivation speed of the neck from a general speed (for example, 2 to 5 mm/min) to an extreme low speed (for example, 0.8 mm/min) for a certain period of time, and perform this operation at least once. More than three times is preferred, and the shape of the solid-liquid interface is changed to a concave shape and a convex shape in order to remove the slip difference row.

On the other hand, the thin neck formed by the Darth necking method is used to support the strength of the single crystal ingot which becomes heavier due to the large diameter of the single crystal of the recent years, and there is a single crystal in the pull tab. The middle neck is broken and causes a major accident such as falling single crystals. Therefore, the method disclosed in Patent Document 6 is: the distance between the radiation cover of the seed crystal contact liquid and the liquid surface of the crucible when the drawing of the single crystal is performed without the necking process without the necking process. (Interval) is set to 80 mm or more, and the radiant heat is increased to sufficiently preheat the seed crystals, thereby suppressing the occurrence of slippage due to thermal shock at the time of liquid contact.

[First technical literature]

[Patent Document 1] International Publication No. 2010/146853

[Patent Document 2] Japanese Patent Laid-Open Publication No. 2011-57460

[Patent Document 3] Japanese Patent Laid-Open Publication No. 2009-263142

[Patent Document 4] Japanese Patent Laid-Open No. Hei 11-199384

[Patent Document 5] Japanese Patent Laid-Open Publication No. 2001-316198

[Patent Document 6] Japanese Patent Laid-Open Publication No. 2005-281018

The conventional necking method disclosed in Patent Document 1 is a method of removing a shaft-shaped difference which is introduced in parallel with a crystal growth direction when a crystal contact liquid is introduced, but the method of determining the applicable interface shape is not clear, and the shaft is not clear. When the difference row exists near the center of the seed crystal, it is very difficult to remove it.

Similarly, in the conventional necking method disclosed in Patent Document 2, even if the shape of the solid-liquid interface is convex, the angle is extremely small, and in the case where the crystal diameter is sharply thick, the shaft-shaped row may not be removed or Even if the shaft-shaped row is removed, it is necessary to form a problem such as a very long neck. Further, when the crystal diameter is further reduced by the Darth necking method after the process for making the crystal diameter thicker in order to remove the axial difference, the axial difference is guided before reaching the outer peripheral surface of the single crystal and being destroyed. Back to the center direction, the shaft-shaped difference row cannot be removed, and the intentionally implemented shaft-shaped difference elimination step becomes invalid. The conventional necking method disclosed in Patent Documents 3 to 5 also has the same problems as those of Patent Document 2.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a method for producing a single crystal which can reliably remove a shaft-shaped row in a necking process.

In order to solve the above problems, the system of single crystal according to the present invention In the method for producing a single crystal by the CZ method, after the seed crystal is brought into contact with the mash, the above-mentioned seed crystal is pulled, and the cerium single crystal containing the neck is grown at the lower end of the seed crystal. In the above-mentioned neck cultivation process, the first step comprises: pulling the above-mentioned crystal which has contacted the mash liquid at a first pulling speed, and controlling the shape of the solid-liquid interface to be convex, and forming the first neck And a second step of pulling the seed crystal at a second drawing speed slower than the first drawing speed, while controlling the shape of the solid-liquid interface to be convex and making the solid-liquid interface angle larger than the crystal diameter increasing angle Forming a second neck.

According to the present invention, the solid-liquid interface shape in the second step can be made into an appropriate downward convex shape, and the axial difference which cannot be removed in the first step can be removed in the second step. Therefore, it is possible to improve the probability that the single crystal is not poorly discharged by the necking process, and it is possible to produce a high quality single crystal.

In the invention, it is preferable that the first pulling speed is 3 mm/min or more and 5 mm/min or less, the target diameter of the first neck portion is 4 mm or more and 7 mm or less, and the second drawing speed is 1.5 mm/ Above min and below 4.5 mm/min, the target diameter of the second neck portion is 7 mm or more and 11.5 mm or less. In this case, the difference between the first pulling speed and the second pulling speed is 0.5 mm/min. Thereby, in the first step, the main difference row can be surely removed by the Darth necking method. Further, in the second step, the solid-liquid interface angle can be made larger than the crystal diameter increasing angle, and the probability of excluding the axial difference in the direction perpendicular to the solid-liquid interface to the outside of the single crystal can be greatly improved.

In the second step, the solid-liquid interface angle is preferably 4 degrees or more and 10 degrees or less, and the crystal diameter increasing angle is preferably 1 degree or more and 4 degrees or less. Thereby, the interface angle of the solid-liquid interface shape in the second step can be set Set in the appropriate range. Therefore, the probability of removing the shaft-shaped row in the second step can be improved, and the reliability of the second step can be improved.

In the present invention, it is preferable that the length of the second neck portion is 150 mm or more. Further, it is preferable that the length of the first neck portion is 200 mm or less. When the length of the second neck portion is 150 mm or more, the axial difference row can be surely removed. Further, when the length of the first neck portion is 200 mm or less, it is possible to prevent the axial difference from being displaced to the central axis in the first step, and it is possible to prevent a situation in which it is difficult to remove the axial difference.

According to the method for producing a single crystal of the present invention, the above-described single crystal is pulled while the heat shielding body surrounding the single crystal in the cultivation is placed above the above-mentioned molten liquid, and the second step is performed. The distance from the mash to the lower end of the heat shield is set to be smaller than the distance from the mash to the lower end of the heat shield in the first step, and the second neck is preferably incubated. In this case, it is preferable to set the second neck portion by setting the distance from the mash to the lower end of the heat shield in the second step to 80 mm or less. Thereby, the convex solid-liquid interface shape can be easily formed in the first step, and the process of making the crystal diameter thin can be easily performed. Further, in the second step, the solid-liquid interface shape tends to be convex, and the convex-solid solid-liquid interface shape can be easily formed.

According to the method for producing a single crystal of the present invention, after the end of the second step of the neck culture process, the diameter of the single crystal is not further reduced, and the shoulder cultivation process is directly performed, and the single sheet is processed. It is preferred that the diameter of the crystal gradually widens to reach the desired diameter. Thereby, it is possible to prevent the axial difference moved to the vicinity of the outer peripheral surface from being discharged back to the center. Therefore, the second step can be prevented The implementation became an invalid state of affairs.

According to the present invention, it is possible to provide a method for producing a single crystal which can surely remove a shaft-shaped row in a necking process.

1‧‧‧矽Single crystal pulling device

2‧‧‧矽Single crystal

2a‧‧‧ neck

2a ₁ ‧‧‧First neck

2a ₂ ‧‧‧second neck

2b‧‧‧Shoulder

2c‧‧‧ Body

2d‧‧‧ tail

2s‧‧ ‧ kinds of crystal

3‧‧‧矽融液

10‧‧‧Reaction room

10A‧‧‧Main Reaction Room

10B‧‧‧La crystal room

11‧‧‧Insulation

12‧‧‧Quartz

13‧‧‧Base

14‧‧‧Rotary support shaft

15‧‧‧heater

16‧‧‧Hot shield

17‧‧‧ cited lead

18‧‧‧Lead take-up mechanism

D‧‧‧Axis difference

G‧‧‧ interval width

r ₀ ‧‧‧ radius of single crystal

r ₁ ‧‧‧ radial position of the shaft

r ₂ ‧‧‧ radius of single crystal

S ₁ ‧‧‧Single crystal outer surface

S ₂ ‧‧‧ solid-liquid interface

S101‧‧‧Liquid process

S102‧‧‧Necking process

S103‧‧‧Shoulder cultivation process

S104‧‧‧ Body cultivation process

S105‧‧‧Tail cultivation process

S11‧‧‧ first step

S12‧‧‧ second step

θ ₁ ‧‧‧ solid-liquid interface angle

θ ₂ ‧‧ crystallization angle increase angle

[Fig. 1] Fig. 1 is a schematic cross-sectional view showing the structure of a single crystal pulling device 1.

[Fig. 2] Fig. 2 is a flow chart showing the cultivation process of the single crystal.

[Fig. 3] Fig. 3 is a schematic cross-sectional view showing the shape of a single crystal ingot.

[Fig. 4] Fig. 4 is a schematic view showing the growth of a single crystal in the necking process.

[Fig. 5] Fig. 5 is a flow chart for explaining the necking process.

Fig. 6 is a schematic view showing the relationship between the cross-sectional shape of the neck of the single crystal and the position of the axial dislocation in the neck.

[Formation for implementing the invention]

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

Fig. 1 is a schematic cross-sectional view showing the structure of a single crystal pulling device 1.

As shown in Fig. 1, the single crystal pulling device 1 has a reaction chamber 10, a heat insulating material 11 disposed inside the reaction chamber 10, and a susceptor 13 supported in the reaction chamber 10. Quartz crucible 12; rotating branch The shaft 14 is supported to support the base 13 so as to be movable up and down; the heater 15 is disposed around the periphery of the base 13; and the heat shield 16 having a substantially inverted truncated cone shape is disposed above the base 13; A wire 17 referred to by the crystal pulling is disposed coaxially with the rotation support shaft 14 above the susceptor 13 and a lead winding mechanism 18 disposed above the reaction chamber 10.

The reaction chamber 10 is composed of a main reaction chamber 10A and a crystal pulling chamber 10B. The crystal pulling chamber 10B is connected to an upper opening of the main reaction chamber 10A. The quartz crucible 12, the susceptor 13, the rotating support shaft 14, the heater 15, and the heat are provided. The shielding body 16 is provided in the main reaction chamber 10A. The heat shielding body 16 is provided above the mash liquid 3 in a state of surrounding the singular crystal 2 in the cultivation. The lead wire take-up mechanism 18 is disposed above the crystal pulling chamber 10B, and the lead wire 17 extends downward from the lead wire take-up mechanism 18 through the inside of the crystal pulling chamber 10B, and the leading end portion of the lead wire 17 passes through the internal space of the main reaction chamber 10A. In the first drawing, the state in which the single crystal 2 in the cultivation is hung by the lead wire 17 is shown.

In the drawing step of the single crystal, first, a quartz crucible 12 is placed in the susceptor 13, and the crucible material is filled in the quartz crucible 12, and the seed crystal is attached to the front end portion of the lead wire 17 via the seed crystal clip. Next, the crucible material is heated by the heater 15 to form the crucible liquid 3, and the seed crystal is lowered to contact the crucible liquid 3. Thereafter, the seed crystals are gradually raised while rotating the seed crystal and the quartz crucible 12, respectively, and the substantially single crystal monocrystal 2 is grown. At this time, the diameter of the single crystal 2 is controlled by controlling the drawing speed, the power of the heater 15, and the like.

Figure 2 is a flow chart showing the cultivation process for single crystals. Further, Fig. 3 is a schematic cross-sectional view showing the shape of a single crystal ingot.

As shown in the second and third figures, the cultivation of the single crystal 2 is carried out in sequence: a liquid contact process (S101), the seed crystal is brought into contact with the crucible melt; and the necking process (S102), Forming a neck portion 2a having a reduced crystal diameter; a shoulder cultivation process (S103), widening the crystal diameter into a conical shape to form a shoulder portion 2b; and a body cultivation process (S104), in which the single crystal has grown to a predetermined value At the time point of the diameter, the body portion 2c is continuously pulled at a fixed diameter; and the tail cultivation process (S105) is made to narrow the diameter at the end of the drawing, and finally cut off with the liquid surface. By the above, a single crystal ingot having the neck portion 2a, the shoulder portion 2b, the body portion 2c, and the tail portion 2d is completed.

Next, explain the necking process. The necking process is a step of refining the diameter when cultivating the single crystal, and removing the difference (thermal shock difference row, misfit dislocation) introduced by the seed crystal when it contacts the mash.

Among the difference rows introduced during the liquid contact, the crystal growth is extended in the direction of the crystal pulling, and the crystal diameter is not removed by a simple change such as the Darth necking method. Since the axial difference is until the necking process, the shoulder cultivating process, and the body cultivating process, it will continuously exist in the growth direction of the crystal, and will be limited to a specific area of each wafer when the wafer fabrication is completed. Further, the axially-shaped row is difficult to remove by a conventional method of forming a neck portion by expanding the diameter, reducing the diameter, and lengthening the neck portion by a length such as a length of the neck. However, in the present embodiment, the necking process is divided into two steps as shown below, the main slip row is removed in the first step, and the shaft row is removed in the second step, whereby the neck can be eliminated. Poorly discharged.

Figure 4 is a schematic diagram showing the growth of a single crystal in a necking process. Further, Fig. 5 is a flow chart for explaining the necking process.

In the cultivation of a single crystal, the seed crystal 2s is first contacted with the mash 3 (Fig. 4(a)). At this time, in the seed crystal 2s, a plurality of rows including the axial difference row are introduced (Fig. 4(b)). Thereafter, the necking process is performed in order to remove the difference row (Fig. 4) (c) ~ Figure 4 (f)).

As shown in Fig. 5, the necking process according to the present embodiment is constituted by a first step S11 of removing the main difference row other than the axial difference row and a second step S12 of removing the axial difference row. In Fig. 4, Figs. 4(c) and (d) show the first step S11, and Figs. 4(e) and (f) show the second step S12.

In the first step S11, the first neck portion 2a ₁ having a fine crystal diameter is grown by pulling the seed crystal 2s at a relatively fast drawing speed (refer to Figs. 4(c) and (d)). The target diameter of the first neck portion 2a ₁ is preferably 4 to 7 mm. By refining the crystal diameter to about 4 to 7 mm in this manner, it is possible to remove the main difference between the single crystals which continue to be crystallized from the seed crystal 2s.

The pulling speed in the first step S11 is preferably 3 to 5 mm/min. When the pulling speed is fast, the cooling of the center portion is accelerated as compared with the outer peripheral portion of the single crystal, and thus the shape of the solid-liquid interface becomes convex. The length of the first neck portion 2a ₁ is preferably limited to 200 mm or less. This is because when the shape of the solid-liquid interface is convex, the axial difference in the vertical direction of the solid-liquid interface moves to the center direction of the single crystal, and if the first neck 2a _{1 is} too long, the axial difference is arranged. When it is completely moved to the center axis, it becomes impossible to remove the shaft-shaped row in the second step S12.

In the second step S12, the seed crystal 2s is pulled at a pulling speed which is relatively slower than the first step S11, and the solid-liquid interface shape is controlled to be convex (refer to Figs. 4(e) and 4(f)). Further, in the second step S12, the second neck portion 2a _{2 is} cultivated while controlling the pulling condition so that the solid-liquid interface angle becomes larger than the crystal diameter increasing angle. When the solid-liquid interface angle is larger than the crystal diameter increasing angle, the axial difference that moves toward the outside in the radial direction together with the crystal growth increases the crystal diameter, and the axial difference can be excluded to the outside of the single crystal.

Here, the solid-liquid interface angle refers to an angle indicating how much the solid-liquid interface is inclined with respect to the horizontal plane; the crystal diameter increasing angle refers to the inclination angle of the crystal cross-section, and shows that the outer circumference of the single crystal is increased relative to the central axis by the crystal diameter How much is tilted. When the crystal diameter at the start of the second step is R ₁ and the crystal diameter and crystal length at the end of the second step are R ₂ and L, respectively, the crystal diameter increase angle is defined as arctan {(R ₂ -R _{1 )} )/2L}.

In the present embodiment, the solid-liquid interface angle is preferably 4 to 8 degrees. If the angle of the solid-liquid interface is within this range, the axial difference row can be surely removed. In order to make the solid-liquid interface angle 4 to 8 degrees, for example, the pulling speed is set to 1.5 to 4.5 mm/min, and the rising speed of the crucible is set to 1 mm/min. Further, the target diameter of the second neck portion 2a ₂ is preferably set to 7 to 11.5 mm.

Further, in the second step S12, it is preferable to reduce the distance (interval width G) from the liquid surface of the mash liquid 3 to the heat shielding body 16 (refer to FIG. 1), and it is particularly preferable to set it as 80 mm or less. This is because if the gap width G is narrowed, it becomes a cold-cold condition, and the radiant heat from the heater 15 is blocked, and the single crystal is easily cooled. By suppressing the influence of the radiant heat as described above, the cooling of the outer peripheral portion is accelerated as compared with the central portion of the single crystal, and the solid-liquid interface shape tends to be convex. Conversely, in the first step S11, by setting the gap width G larger than that in the second step S12, it is possible to enhance the tendency of the solid-liquid interface shape to be convex, and it is possible to promote the elimination of the axial gap.

The shape of the solid-liquid interface is a downward convexity when the crystal diameter gradually becomes coarse as the crystal grows. At this time, it is necessary to show that the crystal sharpening angle θ _{2 of the} single crystal is coarser than the solid-liquid interface angle θ ₁ . This is because if the single crystal is sharply thickened and the crystal diameter increasing angle θ ₂ is excessively large, the rate of increase in the diameter of the single crystal is too fast with respect to the speed at which the axial difference is discharged to the outside of the radial direction of the single crystal. It is impossible to chase the axial difference to the outside of the single crystal.

Fig. 6 is a schematic view showing the relationship between the cross-sectional shape of the neck portion 2a of the single crystal 2 and the position of the axial difference in the neck portion 2a.

As shown in Fig. 6, it is assumed that the axial displacement row D exists at the center of the single crystal 2 (r = 0) at the start of the second step S12, and this position is the origin of the cylindrical coordinate system. In the second step S12, the solid-liquid interface S _{2 has} a downward convex shape, and when the inclination angle thereof is the solid-liquid interface angle θ ₁ , the radial difference from the origin to the axial displacement row D at the position of the crystal length L is radial. shift position becomes r ₁ r ₁ = Ltanθ _1. Further, the radius of the single crystal 2 at the origin is r ₀ , and when the radius of the single crystal 2 increases by Δr as it grows from the origin to the crystal length L, the crystal diameter increasing angle θ ₂ becomes tan θ ₂ = Δr / L, 2 single crystal radius r ₂ be _{r 2 = r 0 + △ r} = r 0 + Ltanθ 2.

In order to move the axial displacement row D to the outer side in the radial direction as the crystal grows, it is possible to trace the outer side of the outer peripheral surface S ₁ of the single crystal 2, and it is necessary that r ₁ >r ₂ . That is, Ltan θ ₁ > r ₀ + Ltan θ ₂ is necessary.

For example, when the radius r ₀ of the single crystal 2 at the start of the second step S12 is 2.5 mm (diameter: 5 mm), the solid-liquid interface angle θ ₁ = 4 degrees, and the crystal diameter increasing angle θ ₂ =0 (not increasing the diameter) , a length of 35.7mm or more in the crystal, dislocation D-shaped shaft can overtake the outer peripheral surface S 2 of the outer single crystal ₁ extinguished.

Further, the increase in the crystal length L is crystallized 150mm 5mm diameter increased from 11.5mm to become _{tanθ 2 = (11.5-5) / (} 2 × 150) = 0.022 ( crystal diameter by an angle θ ₂ ≒ 1.26 degrees). Therefore, for example, when the crystal diameter increasing angle θ ₂ = 1.26 degrees, in order to exclude the axial row D from the outer side surface S ₁ of the single crystal 2, the necessary crystal length L is as follows.

L>5.75/(0.07-0.022)=119.8(mm)

That is, as long as the crystallized neck 2a ₂ of the second length L of 120mm or more, the outer axial dislocation single crystal D out chase extinguished. However, in practice, considering the error of the solid-liquid interface angle θ ₁ , it is preferable to grow 30 to 50 mm longer than the theoretical crystal length L, and it is preferable that the crystal length L of the second neck portion 2a ₂ is 150 mm or more. .

After the end of the second step S12 of the necking process S102, it is preferable to make the diameter of the single crystal 2 thinner and to directly perform the shoulder growing process S103. This is because if the diameter of the single crystal 2 is again made fine, the axially-displaced row D which has been intentionally moved to the vicinity of the outer peripheral surface S ₁ is returned to the center direction, and the necking process to this point is in vain. Since the step S12 that even by a second single crystal can not be excluded from the second shaft-shaped dislocations still present in the vicinity of the outer peripheral surface S _1, in turn, will be moved to the shoulder incubated process S103 immediately after the outer peripheral surface S ₁ of The possibility of exclusion from the outside is possible, even if it is not excluded, and there is a possibility that it is removed by peripheral grinding during the processing of the product. Therefore, after the end of the second step S12, it is preferable to perform the shoulder cultivation process S103 directly.

As described above, according to the method for producing a single crystal of the present embodiment, the first step and the second step are performed in order to remove the difference between the seed crystals 2s and the necking process S102. The step is to make the diameter of the single crystal thinner and to make the solid-liquid interface shape convex and to cultivate. The second step is to increase the diameter of the single crystal to make the solid-liquid interface shape convex and to make the solid-liquid interface angle Since θ _{1 is} larger than the crystal diameter increasing angle θ ₂ and the single crystal is grown, it is possible to remove the axial difference in the single crystal, and it is possible to make the single crystal undifferentiated before the shoulder is grown.

The preferred embodiments of the present invention have been described above, but the present invention is not limited to the embodiments described above, and various modifications may be made without departing from the spirit and scope of the invention. Inside.

For example, in the above embodiment, in addition to the pulling speed, the interval is widened, and the solid-liquid interface shape is more easily changed. Alternatively, the heater power may be changed instead of the variation interval, and other drawing conditions may be changed.

S11‧‧‧ first step

S12‧‧‧ second step

Claims (9)

A method for producing a single crystal, which is a method for producing a single crystal by using a CZ method, wherein after the seed crystal is brought into contact with the mash, the crystallization of the crystallization of the neck is performed by pulling the crystallization of the seed crystal. The lower end grows, characterized in that the neck cultivation process comprises: a first step of pulling the above-mentioned crystal which has contacted the mash liquid at a first pulling speed, and controlling the shape of the solid-liquid interface to be convex, while Forming a first neck; and a second step of pulling the seed crystal at a second pulling speed slower than the first pulling speed, while controlling the shape of the solid-liquid interface to be convex and making the solid-liquid interface angle larger than the crystal Increasing the diameter angle, forming a second neck portion; wherein the second step of the above-mentioned neck cultivation process is finished, the diameter of the single crystal is not further reduced, and the shoulder cultivation process is directly performed, and the single crystal is formed. The diameter gradually widens to the desired diameter. The method for producing a single crystal according to the first aspect of the invention, wherein in the first step, the first pulling speed is 3 mm/min or more and 5 mm/min or less, and the target diameter of the first neck portion is 4 mm or more and 7 mm or less; and in the second step, the second drawing speed is 1.5 mm/min or more and 4.5 mm/min or less, and the target diameter of the second neck portion is 7 mm or more and 11.5 mm or less. The method for producing a single crystal according to the first aspect of the invention, wherein in the second step, the solid-liquid interface angle is 4 degrees or more and 10 degrees or less. A method for producing a single crystal according to item 3 of the patent application, wherein The crystal diameter increasing angle is 1 degree or more and 4 degrees or less. The method for producing a single crystal according to the third aspect of the invention, wherein the length of the second neck portion is 150 mm or more. The method for producing a single crystal according to the fourth aspect of the invention, wherein the length of the second neck portion is 150 mm or more. The method for producing a single crystal according to any one of claims 1 to 6, wherein in the first step, the length of the first neck portion is set to 200 mm or less. The method for producing a single crystal according to any one of claims 1 to 6, wherein a heat shielding body surrounding the single crystal of the single crystal in the cultivation is disposed above the above-mentioned molten liquid to pull the single crystal At the same time, the distance from the mash to the lower end of the heat shield in the second step is set to be smaller than the distance from the mash to the lower end of the heat shield in the first step, and the second is cultivated. neck. The method for producing a single crystal according to claim 7, wherein the second neck is cultivated by setting a distance from the mash to the lower end of the heat shield in the second step to be 80 mm or less. .