JPH0444215B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention uses the Czyochralski method (CZ method) to measure the level of the melt surface of a single crystal raw material in an apparatus that mainly manufactures single crystals such as semiconductors. Regarding how to.

[Conventional technology]

This type of single crystal growth apparatus is generally constructed as shown in FIG.

FIG. 4 is a schematic vertical sectional view of a conventional device, and numeral 1 in the figure indicates a chamber. Inside the chamber 1, a double-structured crucible 2 in which a quartz container 2b is arranged inside a graphite container 2a is disposed at the center.
A heater 3 and a heat insulating wall 4 are disposed concentrically on the outer periphery of the crucible 2 with a gap forming an exhaust flow path between them. A radiation screen 5 is provided over the area.

The upper wall of the chamber 1 has a pulling port 1a for the single crystal 6.
is opened, and a pulling shaft constituting a pulling device is introduced into the chamber 1 through this pulling port 1a,
The single crystal 6 is grown on the lower end of the seed crystal 7 fixed thereto and pulled up.

The radiation screen 5 consists of a flat annular rim 5a and a truncated conical tapered part 5b whose diameter decreases downward from the inner edge of the rim 5a. It is placed on the chamber 1 to block radiant heat from the crucible 2, heater 3, melt surface, etc., increase the temperature gradient in the pulling direction of the single crystal 6, and at the same time feed the single crystal 6 from above to the crucible 2. A carrier gas such as Ar is guided into the crucible 2, and SiO gas etc. generated from the crucible 2 are guided below the crucible 2 and discharged from there to the outside of the chamber.

By the way, in such a single crystal growth apparatus, changes in the growth conditions of the single crystal 6 can cause polycrystalization, so it is necessary to always maintain a constant temperature as well as the level of the melt surface in the crucible 2. , the melt level is also constantly monitored, and the single crystal 6
The change in the level of the melt surface caused by the pulling up of the crucible 2 is compensated for by adjusting the position of the crucible 2 itself.

That is, assuming that the diameter of the single crystal being pulled is φ _C and the diameter of the melt surface is φ _L , the crucible rising speed V for maintaining the melt surface at a constant height is determined by the following equation (1).

V=crystal rising speed×φ _C ² /φ _L ² ×ρ _C /ρ _L …(1) However, ρ _C : Specific gravity of crystal ρ _L : Specific gravity of melt [Problem to be solved by the invention] However, The inner diameter of the quartz container 2b that makes up the crucible 2 varies, and the inner diameters of both the quartz container 2b and the graphite container 2a change due to corrosion due to the reaction with SiO, so the inner diameter varies by several mm for each pulling batch.
fluctuations occur, making it difficult to make accurate corrections, making it difficult to accurately control the rise of the crucible 2, causing polycrystalline phenomena to occur frequently, and fluctuations in the melt surface level causing the lower end of the melt and radiation screen to There were problems such as contact with the edges and contaminating the melt, and this also blocking the passage of exhaust gases such as SiO.

The present invention was made in view of the above circumstances, and its purpose is to directly detect the melt level by using a part of the existing equipment as is, without making any special modifications. The present invention provides a method for measuring a melt level.

[Means for solving problems]

A method for measuring the level of a melt surface in a single crystal growth apparatus according to the present invention is provided in a single crystal growth apparatus in which a radiation screen is provided above a crucible that is moved up and down and around a single crystal pulling area. In the method of measuring the level of the melt surface in the radiation screen, a reference point set on a part of the radiation screen and a reflected image of the reference point with respect to the melt surface are measured using a linear sensor that emits a signal according to the light intensity. The level of the melt surface is determined based on the distance between the reference point and its reflected image on the linear sensor.

[Effect]

This makes it possible for the method of the present invention to directly grasp changes in the level of the melt surface, regardless of the deformation or volume change of the crucible itself.

[Example] Hereinafter, the method of the present invention will be specifically explained based on drawings showing examples thereof. Fig. 1 is a schematic cross-sectional view showing the implementation state of the method of the present invention, Fig. 2 is an explanatory diagram of the measurement principle, and Fig. 3 A shows the positional relationship between the linear sensor, the reference point captured by it, and its reflected image. Explanatory diagram,
FIG. 3B is a waveform diagram showing the signal level, in which 1 indicates the chamber, 2 the crucible, 3 the heater, and 4 the heat retaining wall. A heat retaining wall 4 is disposed inside the chamber 1 along its side periphery, a crucible 2 is disposed in the center surrounded by the heat retaining wall 4, and a heater 3 is installed between the crucible 2 and the heat retaining wall 4. are arranged with a gap forming an exhaust air passage between them, and a radiant screen 5 is arranged extending from the heat retaining wall 4 to the peripheral edge of the melt surface in the crucible 2. There is.

The crucible 2 has a double structure in which a quartz container 2b is fitted inside a graphite container 2a, and the upper end of a shaft 2c passing through the bottom wall of the chamber 1 is connected to the center of the bottom. The shaft 2c is rotated and raised and lowered by motors M ₁ and M ₂ connected to the lower end of the shaft 2c.

At the center of the upper wall of the chamber 1 is opened a single crystal pulling port 1a which also serves as a supply port for the atmospheric gas inside the chamber 1, and at one location around it is an observation window 1.
b is opened, a protective tube 8 is erected at the pulling port 1a, and a camera 11 is disposed facing the outside of the observation window 1b.

A pulling shaft 7a is suspended from the upper end of the protection tube 8, and a seed crystal 7 is suspended from the lower end of the shaft via a chuck.
It is connected to the rotation and lifting functions equipped with M ₃ and M ₄ , and after blending the seed crystal 7 with the melt, by rotating and raising it, the single crystal 6 is placed at the lower end of the seed crystal 7. It is designed to encourage growth.

The radiation screen 5 is constructed by providing a tapered part 5b on the inner peripheral edge of a metal annular rim 5a, the diameter of which decreases downward from here and is inclined to form a hollow inverted truncated cone shape. The outer peripheral edge of 5a is held on the heat retaining wall 4 by abutting and suspending it on the heat retaining wall 4.

The camera 11 has a lens 11 as shown in FIGS.
The lower edge 5 of the tapered portion 5b of the radiation screen 5 is placed in the field of view of the linear sensor 12 at the imaging position a.
The installation position and installation angle (approximately 30 degrees) are set so as to capture the reflected image 5 of the lower edge 5c with respect to the lower edge 5c and the melt surface.

The linear sensor 12 is oriented perpendicularly to the tangent to the curved edge of the lower edge 5c as shown in FIG.
It is set to a length that can capture both the image Nb of the reflected image 5, and the image captured by each light receiving element.
It is designed to output an electrical signal according to the brightness of Na and Nb, and the output is read into the arithmetic control section 13.

Among the images captured by the linear sensor 12, the image Na of the lower edge 5c is relatively dark, while the reflected image Nb from the melt surface shows that the melt surface itself is at a high temperature.
The back surface of C reflects the light from the heater 3, etc., and the surface of the melt is illuminated by the reflected light, making it extremely bright, and the signal level of the electrical signal output from the light receiving elements of each part of the linear sensor 12 is The result will be as shown in Figure 3B. That is, the signal level is highest at the image receiving position of image Nb, exhibits a substantially constant level from there to the image receiving position of image Na, and decreases rapidly after passing the image receiving position of image Na.

The calculation control unit 13 detects the positions of the images Na and Nb based on the signal levels from these linear sensors 12, and calculates the distance h between the lower edge 5c of the tapered portion 5b and the reflected image 5 based on these positions. .

That is, from the lens 11a of the camera 11 to the lower edge 5
When the distance to c is a, the distance from the lens 11a to the linear sensor 12 is b, and the inclination angle of the optical axis of the lens 11a is θ, the following relational expression holds true.

h sinθ+h・Nb−Na/bΔd cosθ=a/b・(Nb−Na)・Δd ∴h=a/b・(Nb−Na)・Δd/sinθ+Nb−Na/b
・Δd cosθ h=a・(Nb−Na)・Δd/b sinθ−cosθ(Nb−N
a)・Δd…(2) However, Δd: Size of a single light receiving element in the longitudinal direction of the linear sensor 12 Nb−Na: Image Na on the linear sensor,
Dimension between Nb Therefore, the distance l from the lower edge 5c of the tapered portion 5b to the melt surface is given by the following equation (3).

l=1/2h...(3) The calculation control unit 13 compares the calculated l with a predetermined reference level value, and adjusts the crucible 2 to eliminate the difference.
A control signal is output to the motor M ₂ for adjusting the position of the crucible 2.

In the above embodiment, a case has been described in which the level of the melt surface is detected using the lower edge 5c of the tapered portion 5b of the radiation screen 5 as a reference point, but the present invention is not limited to this, and instead of the radiation screen 5. It is also possible to use another immovable member as a reference point or to arrange a special reference member.

[Test example]

Using an 18-inch crucible, 60 kg of silicon polycrystalline grains as a raw material were put into it, and during the process of pulling a silicon single crystal with a diameter of 150 mm at a speed of 1.0 mm/min, the melt surface level was lowered by the method of the present invention as described above. When the position of the crucible was controlled based on the measurements, it was confirmed that the change in the distance between the lower edge of the radiation screen and the melt surface could be kept within a range of ±0.3 mm.

The above embodiment is based on the image Na of the lower edge 5c of the tapered portion 5b in the radiation screen 5 and its reflected image 5.
Although the case has been described in which the melt surface level is detected by determining the separation dimension from the image Nb and the position of the crucible is controlled based on this, in reality, the position of the lower edge 5c of the radiation screen 5 does not move. , it is sufficient to focus only on the position of the image Nb on the linear sensor 12.

Therefore, when the image Nb moves to the opposite side of the image Na on the linear sensor 12, it means a decrease in the melt level, and when it moves towards the image Na side, it means an increase in the melt level. Therefore, the position and movement speed of the crucible 2 may be controlled so that the position of the image Na does not change from the initially determined position.

〔effect〕

As described above, in the method of the present invention, a reference point is set on the radiation screen covering above the melt surface, and the distance between this reference point and the reflected image of the reference point from the melt surface is detected to determine the distance between the two. This makes it possible to accurately detect changes in the melt surface without being affected by changes in the volume of the crucible itself, and to adjust the crucible height accordingly, accurately and quickly. Not only can single-crystal growth conditions be maintained and the frequency of polycrystalization phenomenon can be greatly suppressed, but also the contact between the melt and the radiation screen due to changes in the melt surface level, and the resulting obstruction of the exhaust gas flow path. The present invention has excellent effects such as being able to prevent inconveniences such as closure.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram showing the implementation state of the method of the present invention,
FIG. 2 is a partially enlarged explanatory diagram, FIG. 3 is an explanatory diagram showing the principle of measurement, and FIG. 4 is a schematic vertical sectional view of a conventional device. 1... Chamber, 2... Crucible, 3... Heater, 4... Heat insulation wall, 5... Radiation screen, 5b... Taper part, 5
c...lower edge, 6...single crystal, 7...seed crystal, 11...camera, 12...linear sensor, 13...arithmetic control unit,
Na, Nb...image.

Claims (1)

[Claims] 1. In a method for measuring the level of a melt surface in a crucible in a single crystal growth apparatus in which a radiation screen is provided above a crucible that is moved up and down and around a single crystal pulling area, the radiation screen is A reference point defined in a part and a reflected image of the reference point on the melt surface are captured by a linear sensor that emits a signal according to the light intensity, and the separation between the reference point and the reflected image on the linear sensor is determined. A method for measuring the level of a melt surface in a single crystal growth apparatus, characterized in that the level of the melt surface is determined based on dimensions.