JP2013049586A - Continuous casting method of silicon ingot - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a continuous casting method of silicon ingots using electromagnetic induction, by which occurrence of dislocations can be suppressed while sufficiently suppressing production of residual stress.SOLUTION: A plurality of heaters for heat retention and a plurality of soaking heaters are continuously arranged in stages under a bottomless cooling crucible. In accordance with a coordinate system with the lower end position of the cooling crucible as the origin and the vertical lower side as the positive, the vertical setting position of a thermometer for monitoring the temperature of the uppermost heater for heat retention is defined as a first position Z, the heater temperature at the first position Zis represented by T, the vertical setting position of a thermometer for monitoring the temperature of the uppermost soaking heater is defined as a second position Z, and the heater temperature at the second position Zis represented by T. The vertical setting position of each thermometer used for monitoring the temperature of each heater for heat retention between the first position Zand the second position Zis defined as Z. The output of each heater for heat retention is controlled so that the heater temperature T at each position Z satisfies conditions of expression (1).

Description

  The present invention relates to a silicon ingot continuous casting method in which a silicon ingot that is a material of a substrate for a solar cell is continuously cast using electromagnetic induction.

  A polycrystalline silicon wafer used for a substrate of a solar cell is manufactured by slicing a unidirectionally solidified silicon ingot as a raw material. As a method for producing a silicon ingot, for example, as disclosed in Patent Documents 1 and 2, a continuous casting method using electromagnetic induction (hereinafter also referred to as “electromagnetic casting method”) has been put into practical use. The electromagnetic casting method is a very useful technique in that a high-quality silicon ingot can be manufactured at low cost.

  FIG. 3 is a longitudinal sectional view schematically showing a configuration of a typical electromagnetic casting apparatus used in the electromagnetic casting method. As shown in FIG. 1, the electromagnetic casting apparatus includes a chamber 1. The chamber 1 is a water-cooled container having a double wall structure in which the inside is isolated from the outside air and maintained in an inert gas atmosphere suitable for casting. A raw material supply hopper 2 is connected to the upper wall of the chamber 1. The chamber 1 is provided with an inert gas inlet 3 in the upper part and an exhaust port 4 in the lower side wall.

  In the chamber 1, a bottomless cooling crucible 5, an induction coil 6 and an after heater 7 are arranged. The cooling crucible 5 functions not only as a melting container but also as a mold, and is a rectangular tube made of metal (for example, copper) excellent in thermal conductivity and conductivity, and is suspended in the chamber 1. . The cooling crucible 5 is formed with a plurality of slits (not shown) in the vertical direction, leaving the upper and lower portions, and is divided into a plurality of strip-shaped pieces in the circumferential direction by the slits, and is forced by cooling water flowing through the inside. To be cooled. The induction coil 6 is provided around the cooling crucible 5 so as to surround the cooling crucible 5 and is connected to a power supply device (not shown).

  The after heater 7 is provided continuously below the cooling crucible 5 and surrounds the silicon ingot 16 that is pulled down from the cooling crucible 5. The after-heater 7 includes a plurality of stages of heat retaining heaters 8 and a plurality of stages of soaking heaters 9 in order from the top near the cooling crucible 5. FIG. 3 shows an example in which four stages of heat retaining heaters 8 and 15 stages of soaking heaters 9 are provided.

  In the chamber 1, a raw material introduction pipe 10 is disposed below the raw material supply hopper 2. Granular or lump silicon raw material 14 is supplied from the raw material supply hopper 2 to the raw material introduction pipe 10 and is introduced into the cooling crucible 5 through the raw material introduction pipe 10.

  Immediately below the cooling crucible 5, a support bar 11 that can be moved up and down penetrating the bottom wall of the chamber 1 is provided, and a support base 12 is attached to the upper end of the support bar 11. The ingot 16 is pulled down as the support bar 11 is lowered while being supported by the support base 12.

  A plasma torch 13 is provided directly above the cooling crucible 5 so as to be movable up and down. The plasma torch 13 is connected to one pole of a plasma power supply device (not shown), and the other pole is connected to the ingot 16 side. The plasma torch 13 is inserted into the upper part of the cooling crucible 5 by descending.

  In the electromagnetic casting method using the electromagnetic casting apparatus having such a configuration, the silicon raw material 14 is charged into the cooling crucible 5, an alternating current is applied to the induction coil 6, and the plasma torch 13 inserted at the top of the cooling crucible 5 is applied. Energize. At this time, since the strip-shaped pieces constituting the cooling crucible 5 are electrically divided from each other, an eddy current is generated in each piece due to electromagnetic induction by the induction coil 6, and the cooling crucible 5 The eddy current on the inner wall side generates a magnetic field in the cooling crucible 5. As a result, the silicon raw material 14 in the cooling crucible 5 is melted by electromagnetic induction heating to form molten silicon 15. In addition, a plasma arc is generated between the plasma torch 13 and the molten silicon 15, and also by this plasma arc heating, the silicon raw material 14 is heated and melted, and the burden of electromagnetic induction heating is reduced to efficiently melt the molten silicon 15. Is formed.

  The molten silicon 15 has a force (pinch force) in the inner normal direction of the surface of the molten silicon 15 due to the interaction between the magnetic field generated along with the eddy current on the inner wall of the cooling crucible 5 and the current generated on the surface of the molten silicon 15. ), The cooling crucible 5 is held in a non-contact state. When the support base 12 supporting the molten silicon 15 is gradually lowered while the silicon raw material 14 is melted in the cooling crucible 5, the induction magnetic field decreases as the distance from the lower end of the induction coil 6 decreases. Further, the molten silicon 15 is solidified from the outer peripheral portion by cooling from the cooling crucible 5. Then, the silicon raw material 14 is sequentially introduced into the cooling crucible 5 as the support base 12 is lowered, and melting and solidification are continued, so that the molten silicon 15 is solidified in one direction and the ingot 16 is continuously cast. it can.

  According to such an electromagnetic casting method, contact between the molten silicon 15 and the cooling crucible 5 is reduced, so that impurity contamination from the cooling crucible 5 due to the contact is prevented, and a high-quality ingot 16 can be obtained. it can. And since it is continuous casting, it becomes possible to manufacture the ingot 16 at low cost.

  Here, in the electromagnetic casting method described above, in order to suppress the occurrence of residual stress that causes cracking of the ingot 16, the cast ingot 16 is not immediately cooled to room temperature, but is once subjected to predetermined soaking using the after heater 7. The temperature is maintained and then cooled to room temperature. That is, the ingot 16 pulled down from the lower end of the cooling crucible 5 is gradually cooled to the soaking temperature in the process of passing through the installation area (hereinafter also referred to as “insulating zone”) of the heat insulating heater 8 immediately below the ingot 16, and continues the soaking heater. The temperature is maintained at a predetermined soaking temperature in nine installation areas (hereinafter also referred to as “soaking tropics”), and after a predetermined time has passed, the liquid is taken out of the chamber 1 and cooled.

  Another technique for suppressing residual stress is, for example, an electromagnetic casting method described in Patent Document 3. In the electromagnetic casting method described in Patent Document 3, as the after heater that surrounds the ingot below the cooling crucible, the heating means and the cooling means are alternately arranged, or the heating means and the heat insulating means are alternately arranged. With this structure, this structure alternately exhibits a heat retaining action and a heat removal action, effectively giving a temperature gradient along the ingot pulling direction, and as a result, keeping the shape of the solid-liquid interface of the ingot flat. The residual stress can be reduced.

International Publication WO02 / 053496 Pamphlet Japanese Patent Laid-Open No. 7-138012 International Publication WO2006 / 088037 Pamphlet

  The conventional electromagnetic casting method described above is useful for preventing cracking of an ingot because it can suppress the occurrence of residual stress. However, an ingot manufactured by a conventional electromagnetic casting method cannot stably obtain high conversion efficiency when configured as a substrate of a solar cell. Although details will be described later, although the thermal stress in the vicinity of the solid-liquid interface of the ingot is determined according to the cooling pattern in the heat retention zone before reaching the soaking zone, the thermal stress is large in the conventional electromagnetic casting method, This is due to the frequent occurrence of dislocations in the ingot.

  In particular, in the electromagnetic casting method described in Patent Document 3, since there is no soaking in the area of the after heater, the effect of suppressing the residual stress is not sufficient, as will be described in detail later. Further, as the after heater, since the heating means and the cooling means are alternately arranged, or the heating means and the heat insulating means are alternately arranged, a plurality of stages of heat retaining heaters and a plurality of stages of soaking heaters are simply provided. There is also a drawback in that the structure becomes complicated as compared with a configuration in which they are continuously arranged.

  The present invention has been made in view of the above-mentioned problems, and as the after heater under the cooling crucible, a plurality of stages of heat retaining heaters and a plurality of stages of soaking heaters are simply arranged in succession, and residual stress An object of the present invention is to provide a continuous casting method of a silicon ingot using electromagnetic induction that can sufficiently suppress generation and can suppress generation of dislocation.

The gist of the present invention is the following method for continuously casting a silicon ingot. That is, a silicon raw material is put into a conductive bottomless cooling crucible disposed in a chamber, and the silicon raw material is melted by electromagnetic induction heating from an induction coil surrounding the bottomless cooling crucible, and this molten silicon is bottomless cooled. A silicon ingot is continuously solidified while being pulled down from a crucible to cast a silicon ingot, and after the casting, the silicon ingot is kept at a predetermined soaking temperature and then cooled.
Below the bottomless cooling crucible, surrounding the silicon ingot, the heat insulation heater is connected in multiple stages, and the soaking heater is connected in multiple stages,
According to the coordinate system in which the lower end position of the bottomless cooling crucible is the origin and the vertical downward direction is positive,
The vertical installation position of the thermometer used for temperature monitoring of the uppermost insulation heater as a first position Z _0, with represents the heater temperature at the first position Z ₀ at T _0,
If the vertical installation position of the thermometer used for temperature monitoring of the uppermost soaking the heater second and position Z _1, showing the heater temperature at the second position Z ₁ in T _1,
The vertical installation position of each thermometer used for monitoring the temperature of each insulation heater between the first position Z ₀ and the second position Z ₁ is Z, and the heater temperature T at each position Z is (1) This is a silicon ingot continuous casting method characterized in that the output of each heat insulation heater is controlled so as to satisfy the condition of the equation.

In the above continuous casting method, wherein the heater temperature T ₀ is in the range of 1200 to 1350 ° C., the heater temperature T ₁ of the range of 900 to 1150 ° C. in the second position Z ₁ in the first position Z ₀ It is preferable to be inside. More preferably, the said heater temperature T ₀ at the first position Z ₀ is set within the range of 1250 to 1300 ° C., the heater temperature T ₁ of the at the second position Z ₁ in the range of 1000 to 1100 ° C. .

In the continuous casting method, the heater temperature T between the first position Z ₀ and the second position Z ₁ is preferably lower than the heater temperature T ₀ at the first position Z _0. .

  According to the continuous casting method of the silicon ingot using the electromagnetic induction of the present invention, as the after heater under the cooling crucible, a structure in which a plurality of stages of heat retaining heaters and a plurality of stages of soaking heaters are simply arranged continuously, By optimizing the output control of the heater, the occurrence of residual stress can be sufficiently suppressed, and at the same time, the occurrence of dislocation can be suppressed.

It is a figure which shows the cooling pattern in the heat retention zone at the time of continuous casting. It is a figure which shows the thermal stress in a solid-liquid interface for every cooling pattern shown by FIG. It is a longitudinal cross-sectional view which shows typically the structure of the typical electromagnetic casting apparatus used with the electromagnetic casting method.

  As described above, in the electromagnetic casting method, in order to suppress the occurrence of residual stress, a plurality of stages of heat insulation heaters are provided below the cooling crucible to form a heat insulation zone, and subsequently, a plurality of stages of soaking are performed. A heater is installed to form a soaking zone, and the ingot pulled down from the lower end of the cooling crucible is not immediately cooled to room temperature, but is maintained at a predetermined soaking temperature in the soaking zone, and then cooled to room temperature. Like to do. In actual operation, the soaking temperature is set to about 1100 ° C. The retention time in the soaking zone varies depending on the site of the ingot, but the upper end portion of the ingot that is the shortest is 10 hours or longer.

  Both the heat retaining heater and the soaking heater are resistance heating type heaters. Specifically, the heat retaining heater is a so-called carbon heater using carbon as a heating element, and the soaking heater is a so-called Kanthal heater using a metal wire of a heat-resistant alloy such as a Kanthal wire as a heating element. The reason for changing the types of the heat retaining heater and the soaking heater in this way is as follows. Since the heat insulation zone is high temperature, the heat insulation heater installed here is insufficient if it is not a carbon heater. On the other hand, because the soaking zone is relatively cold, Kanthal heaters are sufficient for the soaking heater installed here. Of course, a carbon heater can be used as a soaking heater, but a Kanthal heater is used from the viewpoint of cost reduction.

  The purpose of maintaining the ingot cast as described above at a constant temperature in the soaking zone is to remove the thermal strain accumulated in the ingot, suppress the generation of residual stress, and prevent cracking of the ingot. Thermal strain accumulates in the ingot with plastic deformation due to thermal stress generated in the high temperature range, but by holding the ingot at a predetermined soaking temperature for a long time, the thermal strain is further plastically deformed and relaxed. be able to. If the thermal strain is not removed, a residual stress is generated inside the ingot cooled to room temperature, so that a crack or the like is likely to occur due to the residual stress. Therefore, holding the ingot at a soaking temperature is indispensable for preventing cracking of the ingot.

  Here, the soaking temperature in the soaking zone is set to about 1100 ° C. for the following two reasons. The soaking temperature must firstly be a temperature range in which silicon can be easily deformed, and secondly, it must be a temperature that can be output by a cantal heater (soaking heater). This will be described in more detail. In general, the higher the temperature of a substance, the faster the dislocation movement speed. Therefore, if the same stress is applied, the amount of deformation increases at a higher temperature. The same applies to silicon, and the higher the temperature, the faster the dislocation movement speed. However, the dislocation movement speed of silicon does not change linearly with respect to the temperature, but becomes much faster in a temperature range above that temperature with a certain temperature as a threshold. The temperature is about 1100 ° C. The maximum temperature that can be generated by a Kanthal heater used in the soaking zone is about 1150 ° C.

  On the other hand, it is preferable to set the temperature of the uppermost stage (immediately under the crucible) of the heat retaining zone as high as possible in order to ensure the quality of the ingot. The reason for this is that the ingot in that part is immediately close to solidification and therefore has a high temperature close to the melting point. By keeping the temperature close to the solidification temperature, the thermal stress on the outer surface of the ingot can be reduced, and the occurrence of dislocation due to plastic deformation is suppressed. This is because it is considered possible. However, if the temperature is set too high, a serious accident may occur in which the solidified silicon is re-dissolved and the molten silicon inside the ingot flows out. For this reason, the uppermost temperature of the heat retention zone must be set to a temperature at which silicon remelting does not occur. For these reasons, in actual operation, the uppermost temperature of the heat retention zone is set to around 1300 ° C.

  By the way, from a recent survey, the conversion efficiency when the ingot is configured as a solar cell substrate depends on the temperature at which the ingot immediately after solidification is gradually cooled to the soaking temperature, that is, before reaching the soaking zone. It has been found that the temperature fluctuates significantly depending on the cooling pattern in the heat retention zone. Then, it has been found that in order to evaluate the conversion efficiency according to the cooling pattern of the ingot in the heat retaining zone, the magnitude of the thermal stress in the vicinity of the solid-liquid interface obtained by numerical calculation may be evaluated.

  In the electromagnetic casting method, the larger the thermal stress at the solid-liquid interface, the greater the amount of displacement that the ingot deforms due to the yield phenomenon, and the more dislocations that are generated. In a solar cell substrate, the longer the lifetime of carriers generated by light, the higher the conversion efficiency. However, since dislocations become recombination centers that eliminate the carriers, the more the number of dislocations, the shorter the lifetime of minority carriers. The conversion efficiency deteriorates. Therefore, it can be said that the smaller the thermal stress at the solid-liquid interface during continuous casting, the smaller the amount of dislocations generated, resulting in improved conversion efficiency.

  The inventors assumed various cooling patterns in the heat insulation zone during continuous casting, and calculated and compared the thermal stress at the solid-liquid interface of the ingot by analysis for each of these cooling patterns.

FIG. 1 is a diagram showing a cooling pattern in a heat retaining zone during continuous casting, and FIG. 2 is a diagram showing thermal stress at a solid-liquid interface for each cooling pattern shown in FIG. In the analysis, as shown on the horizontal axis of FIG. 1, a coordinate system was defined in which the lower end position of the cooling crucible was the origin and the vertical downward direction was positive. According to this coordinate system, the vertical installation position of the thermometer used for monitoring the temperature of the uppermost heater is defined as the first position Z _0, and the vertical direction of the thermometer used for monitoring the temperature of the uppermost temperature equalizing heater is set. The installation position was defined as the second position Z _1, and the vertical installation position of each thermometer used for monitoring the temperature of each heat retaining heater between the first position Z ₀ and the second position Z ₁ was defined as Z. Then, the heater temperature at the first position Z ₀ at T _0, the heater temperature at the second position Z ₁ represents respectively T _1, the heater temperature T ₀ at the first position Z ₀ and 1280 ° C., the second the heater temperature T ₁ of the at position Z ₁ as 1100 ° C., was changed heater temperature T at each position Z in three conditions. Here, the heater temperature T ₁ at the second position Z ₁ corresponds to a soaking temperature. Each of the thermometers described above is installed at a substantially central position in the vertical direction of each of the heat retaining heaters and the soaking heaters of the corresponding stages.

  The analysis was performed by the following method. First, the heating conditions at the time of ingot cooling by the after heater, that is, the heater temperature T at each position Z is changed under three conditions (conventional example, example of the present invention, and reference example), heating by the after heater and heating by electromagnetic induction The temperature distribution in the ingot was calculated based on the heat transfer analysis considering the above. Next, the time change of the temperature distribution is input to general-purpose finite element method software, the stress distribution of the cross section perpendicular to the pulling direction (equivalent stress of Mises) is calculated, and the thermal stress at the solid-liquid interface is calculated from this stress distribution. Calculated. And the average value of the calculated thermal stress was made into the evaluation parameter | index, and it showed in FIG.

As shown in FIG. 1, the cooling pattern of the conventional example is a pattern in which the heat insulation zone from the first position Z ₀ to the second position Z ₁ immediately below the crucible is gradually cooled with a linear temperature gradient. Is assumed. The cooling pattern of the example of the present invention is a pattern that gradually cools by following the high temperature side, and assumes the operating conditions of the present invention. The cooling pattern of the comparative example is a pattern that gradually cools down the low temperature side, contrary to the example of the present invention, and assumes operating conditions for comparison.

  The following is shown from the results shown in FIG. It can be seen that the thermal stress generated in the vicinity of the solid-liquid interface is remarkably reduced in the cooling pattern of the present invention in which the cooling is performed while following the high temperature side as compared with the cooling pattern of the conventional example that is gradually cooled with a linear temperature gradient. . On the contrary, it can be seen that the thermal stress generated near the solid-liquid interface increases in the cooling pattern of the comparative example in which the cooling is performed while following the low temperature side.

  From the above, at the time of continuous casting, the cooling pattern of the conventional example in the heat retaining zone is expressed by an equation of a linear function, and the ingot is gradually cooled following the high temperature side which is the upper region of this equation, that is, each position By controlling the output of each heat insulation heater so that the heater temperature T at Z satisfies the condition of the following formula (1), the thermal stress at the solid-liquid interface can be reduced, thereby causing dislocation generation. Can be suppressed. As a result, if the ingot is configured as a solar cell substrate, high conversion efficiency can be stably obtained. Moreover, by maintaining the ingot after such slow cooling at a constant soaking temperature in the soaking zone, the occurrence of residual stress can be sufficiently suppressed, and as a result, cracking of the ingot can also be prevented. It becomes possible.

In the continuous casting method in which the ingot is gradually cooled in the heat retaining zone using the above equation (1), the heater temperature T ₀ at the first position Z ₀ is set within the range of 1200 to 1350 ° C., and the heater at the second position Z ₁ It is preferable that the temperature T ₁ is in the range of 900 to 1150 ° C. More preferably, the heater temperature T ₀ at the first position Z ₀ is set within the range of 1250 to 1300 ° C., the heater temperature T ₁ of the second position Z ₁ in the range of 1000 to 1100 ° C..

As described above, the heater temperature T ₀ at the first position Z ₀ is preferable in the set as high as possible is to ensure the quality of the ingot, setting too high temperatures, once it solidified silicon redissolved, This is because molten silicon inside the ingot may flow out. Further, the heater temperature T ₁ of the second position Z ₁ corresponds to soaking temperature, the first, it is easy temperature range of deformation of the silicon, the second, with Kanthal heater (soaking heater) This is because the temperature needs to be able to be output.

Further, in the continuous casting method in which the ingot is gradually cooled in the heat retaining zone using the above equation (1), the heater temperature T between the first position Z ₀ and the second position Z ₁ is the heater at the first position Z _0. It is preferable that the temperature is lower than T ₀ . Heater temperature T exceeding the heater temperature T ₀ at the first position Z ₀ is because can lead to redissolution of silicon. Practically, the heater temperature T is preferably 20 to 70 ° C. above the temperature calculated from the right side of the above equation (1).

  According to the continuous casting method of a silicon ingot using electromagnetic induction according to the present invention, the generation of residual stress can be sufficiently suppressed, and at the same time, the generation of dislocations can be suppressed. When configured as a substrate, an improvement in conversion efficiency can be realized. Therefore, the continuous casting method of the present invention is extremely useful in that a silicon ingot for a solar cell excellent in quality can be produced.

1: chamber, 2: raw material supply hopper, 3: inert gas inlet,
4: exhaust port, 5: bottomless cooling crucible, 6: induction coil,
7: After heater, 8: Insulating heater, 9: Soaking heater,
10: Raw material introduction pipe, 11: Support rod, 12: Support base,
13: Plasma torch, 14: Silicon raw material, 15: Molten silicon,
16: Ingot

Claims (4)

Silicon raw material is put into a conductive bottomless cooling crucible placed in the chamber, and the silicon raw material is melted by electromagnetic induction heating from an induction coil surrounding the bottomless cooling crucible, and this molten silicon is removed from the bottomless cooling crucible. A silicon ingot is continuously solidified while being pulled down to cast a silicon ingot, and after the casting, the silicon ingot is kept at a predetermined soaking temperature and then cooled.
Below the bottomless cooling crucible, surrounding the silicon ingot, the heat insulation heater is connected in multiple stages, and the soaking heater is connected in multiple stages,
According to the coordinate system in which the lower end position of the bottomless cooling crucible is the origin and the vertical downward direction is positive,
The vertical installation position of the thermometer used for temperature monitoring of the uppermost insulation heater as a first position Z _0, with represents the heater temperature at the first position Z ₀ at T _0,
If the vertical installation position of the thermometer used for temperature monitoring of the uppermost soaking the heater second and position Z _1, showing the heater temperature at the second position Z ₁ in T _1,
The vertical installation position of each thermometer used for monitoring the temperature of each insulation heater between the first position Z ₀ and the second position Z ₁ is Z, and the heater temperature T at each position Z is (1) A continuous casting method of a silicon ingot, wherein the output of each heat insulation heater is controlled so as to satisfy the condition of the formula.

And characterized in that said said heater temperature T ₀ at the first position Z ₀ is set within the range of 1200 to 1350 ° C., in the range of heater temperatures T ₁ to 900 to 1150 ° C. at the second position Z ₁ The method for continuously casting a silicon ingot according to claim 1. And characterized in that said said heater temperature T ₀ at the first position Z ₀ is set within the range of 1250 to 1300 ° C., in the range of heater temperatures T ₁ and 1000 to 1100 ° C. at the second position Z ₁ The method for continuously casting a silicon ingot according to claim 1. Any of claims 1 to 3, the heater temperature T being lower than the heater temperature T ₀ in the first position Z ₀ between the first position Z ₀ and the second position Z ₁ A method for continuously casting a silicon ingot according to claim 1.

Images