JP4868747B2 - Silicon casting method - Google Patents

Description

The present invention relates to silicon casting how for the production of polycrystalline silicon ingot.

  Solar cells are expected to be put to practical use in a wide range of fields, from small households to large-scale power generation systems, as a clean petroleum alternative energy source. These are classified into crystalline, amorphous, and compound types depending on the type of raw materials used, and most of them currently on the market are crystalline silicon solar cells. This crystalline silicon solar cell is further classified into a single crystal type and a polycrystalline type. Single crystal silicon solar cells have the advantage that the conversion efficiency is easy to increase because the quality of the substrate is good, but the disadvantage is that the manufacturing cost of the substrate is high.

  On the other hand, polycrystalline silicon solar cells have been distributed in the market for a long time, but in recent years, the demand is increasing as interest in environmental issues is increasing, and higher conversion efficiency is required at lower cost. . In order to cope with these demands, it is necessary to reduce the cost and quality of the polycrystalline silicon substrate, and it is required to produce a high-purity silicon ingot with a high yield.

  A polycrystalline silicon substrate used for a polycrystalline silicon solar cell is generally manufactured by a method called a casting method. In this casting method, a silicon melt that has been melted by heating at a high temperature is poured into a mold made of quartz or the like coated with a release material to solidify in one direction from the bottom of the mold, or a silicon raw material is placed in the mold and melted once. Then, the silicon ingot is formed by unidirectionally solidifying again from the bottom. The ends of the silicon ingot are removed, cut into a desired size, and the cut silicon ingot is sliced to a desired thickness to obtain a polycrystalline silicon substrate for forming a solar cell.

  As a means for producing such a polycrystalline silicon ingot, a pouring method in which a melting furnace is provided to melt a silicon raw material, and the molten silicon melt is poured into a mold and cooled and solidified. In the casting apparatus, it is necessary to prepare the melted part and the solidified part separately, which increases the size of the apparatus. However, in the method by melting in the mold, in which the silicon raw material is melted in the mold and solidified in the mold as it is, an expensive member such as a high-purity quartz crucible does not need to be provided because there is no need to provide a melting furnace. Can be eliminated, and the cost of the polycrystalline silicon ingot can be reduced. Therefore, it is more advantageous to produce a polycrystalline silicon ingot using the method described later.

  Here, FIG. 7 shows a conventional silicon casting apparatus for in-mold melting.

  In FIG. 7, 11 is a mold, 12 is a cooling unit, 13 is a heat insulating wall, 14 is a mold heating unit, 15 is a silicon raw material (silicon melt), and 18 is a release material. As the mold 11, one in which a mold release material 18 is coated on the inner surface of the mold 11 made of quartz or graphite is usually used. In general, as the release material 18, powders such as silicon nitride which is silicon nitride, silicon carbide which is silicon carbide, silicon dioxide which is silicon oxide, and the like are used, and these powders are mixed with an appropriate binder and solvent. It is known as a well-known technique to mix and stir into a solution composed of a slurry to form a slurry and coat the inner wall of the mold by means such as coating or spraying. In this silicon casting apparatus, a silicon raw material 15 is supplied into the mold 11, the silicon raw material 15 is heated and melted to about 1500 ° C. by the mold heating unit 14, and the mold 11 is gradually lowered to melt the silicon melt. The mold 11 is gradually separated from the mold heating unit 14 to cool, or the position of the mold 11 is left as it is, and the temperature is lowered by the cooling unit 12 to cool, and the silicon melt 15 is solidified in the mold 11 to form a silicon ingot. (For example, see Patent Document 1 and Non-Patent Document 1). This silicon ingot can be taken out of the mold 11 by breaking the mold 11 or can be taken out by disassembling the mold 11. These are all arranged in a vacuum vessel (not shown).

The bulk specific gravity of the silicon raw material in the mold used in such a silicon casting method is usually 1.0 to 1.5, which is about half that of the specific gravity 2.33 of silicon. Therefore, in the conventional silicon casting apparatus as shown in FIG. 7, the height of the silicon ingot formed by solidifying the silicon melt 15 is only about half that of the mold 11. The mold 11 and the mold release material 18 applied to the inner wall surface of the mold 11 are expensive consumables and have a large proportion of the running cost. Therefore, the conventional technique has a problem that the necessary mold 11 is large with respect to the size of the ingot, and the mold 11 and the release material 18 per weight of the silicon ingot are expensive. FIG. 8 shows a conventional additional raw material supply type silicon casting apparatus. As shown in FIG. 8, a movable door 23b is provided above the heat insulating wall 23a on the mold 21, and a raw material supply unit 26 is provided on the movable door 23b. The additional silicon raw material 27 can be supplied into 21 (see, for example, Patent Document 2). As the silicon raw material 25 filled in the mold 21 is thus melted, the silicon melt 25 is increased by additionally supplying an additional silicon raw material 27 into the mold 21 after the bulk of the silicon raw material 25 is reduced. Thus, the cost of the mold 21 and the release material 28 per weight of the silicon ingot can be reduced.
U.S. Pat. No. 3,898,051 Japanese Patent Laid-Open No. 10-139586 JP-A-3-220440 5th International PVSEC p303-305

  However, in the conventional additional silicon raw material supply method, the silicon raw material previously charged in the mold is heated to about 1500 ° C. and melted. When the silicon raw material is additionally supplied from the raw material supply means 26 to the silicon melt in the mold melted in this state, the silicon melt is rapidly cooled, particularly at the liquid surface of the silicon melt. Peeled off from the substrate and mixed into the silicon melt.

  Further, in severe cases, there is a problem that the mold material comes into contact with the silicon melt due to peeling of the release material, and the silicon ingot is cracked to reduce the product yield.

  The inventor is that the temperature in the furnace is drastically decreased when an additional silicon raw material is charged, or the silicon melt adheres to the inner surface of the mold due to the scattering of the silicon melt at the time of charging, etc. As a result, the mold material and the release material coated on the mold material are peeled and damaged due to the difference in thermal expansion coefficient between the mold material and the mold release material due to a rapid thermal history, and enter the silicon melt in the mold. It was found that this was a factor that greatly reduced the product yield.

The present invention has been invented in view of such conventional problems, aims to provide a casting how that solves the problem of decreasing the yield by the incorporation of cracks and release material of the ingot And

Silicon Silicon casting how according to the present invention, in order to achieve the above object, the silicon melt held in the mold a silicon casting method of cooling and solidifying, the silicon melt is held in said mold It is formed by supplying an additional silicon raw material while heating and melting the raw material, and the timing of supplying the additional silicon raw material is determined when the silicon melt and unmelted silicon are mixed in the mold. The temperature rise rate is calculated from the temperature of the silicon raw material held inside or the silicon melt temperature, and the timing for supplying the additional silicon raw material is determined based on this temperature rise rate .

Also, a silicon casting method for cooling and solidifying a silicon melt held in a mold, wherein the silicon melt is formed by supplying an additional silicon raw material while heating and melting the silicon raw material held in the mold The timing for supplying the additional silicon material is when the silicon raw material held in the mold is heated and melted when the silicon melt and unmelted silicon are mixed in the mold. The silicon raw material on the side is preferentially heated and melted.

  With the above-described method, the liquid surface of the silicon melt in the mold can be kept at a particularly low temperature. Therefore, even when additional silicon raw material is added, the thermal history of the liquid surface in the silicon melt can be reduced. it can. Since the silicon melt temperature is low, the viscosity of the silicon melt is also high, so that even when an additional silicon raw material is added, the silicon melt in the mold is difficult to splash. As a result, the mold release material on the inner surface of the mold caused by the rapid temperature drop in the mold accompanying the supply of the additional silicon raw material and the adhesion of the silicon melt can suppress peeling damage. Therefore, it is possible to provide a silicon casting method that suppresses a decrease in yield of the ingot due to cracks in the mold and mixing of the release material.

  Hereinafter, the silicon casting method of the present invention will be described in detail with reference to the accompanying drawings. Here, FIG. 1 shows a first embodiment of the present invention, and FIG. 2 shows a second embodiment of the present invention.

  FIG. 1 is a schematic cross-sectional view of a first embodiment of a silicon casting apparatus according to the present invention, wherein 1 is a mold, 2 is a cooling unit, 3a is a heat insulating wall, 3b is a movable door, and 4a is a mold upper heating unit. 4b is a mold side heating unit, 5 is a silicon raw material or a silicon melt formed by melting silicon material, 6 is a material supply unit, 6a is an introduction member, 6b is an opening / closing port, 6c is a material supply tank, 7 Is an additional silicon raw material, and 8 is a release material.

  The mold 1 is made of graphite, silicon dioxide, or the like, and is configured as an integral structure mold. Such a monolithic mold is disposable to destroy and remove the mold 1 when the silicon ingot is taken out, and is added after melting the silicon raw material in the mold to effectively use the height of the mold. It is preferable to supply the silicon raw material.

  A mold release material 8 is applied to the inner surface side of the mold 1 shown in FIG. 1, and the ingot can be taken out after such a mold release material 8 solidifies the silicon melt 5 inside the mold 1. At this time, if the mold release material 8 is not present, the ingot and the mold 1 are fused, and occupies an important position because cracks and cracks occur when taken out.

  As such a release material 8, each powder of silicon nitride, silicon carbide, silicon oxide or the like or mixed powder is mixed and stirred in a solution composed of an appropriate organic binder and solvent to form a slurry, and a spatula. The inner surface of the mold 1 is coated and dried by using a brush, a brush, a spray method or the like. Common binders include PVA (polyvinyl alcohol), PVB (polyvinyl butyral), PEG (polyethylene glycol), MC (methyl cellulose), CMC (carboxymethyl cellulose), EC (ethyl cellulose), HPC (hydroxypropyl cellulose), and wax. Used for. In addition, when coating the release material 8 mixed with an organic binder aqueous solution to form a slurry, in order to prevent the thermally decomposable product of the organic binder from being mixed into the silicon melt 5 by subsequent heating, A degreasing process is performed. Moreover, as a drying method, the mold release material can be adhered to the mold 1 by performing a degreasing treatment by natural drying or drying using a conventionally known method such as a hot plate or an oven.

  And the casting_mold | template 1 is arrange | positioned inside the heat insulation wall 3a which consists of graphite. In addition, the silicon raw material 5 is charged into the mold and heated by the heating unit 4 to obtain a silicon melt 5. Next, an additional silicon raw material 7 is supplied into the mold 1 by the raw material supply unit 6 provided on the upper part of the mold 1. In addition, this casting_mold | template 1, the heat insulation wall 3a, the heating part 4, and the raw material supply part 6 are installed in a vacuum vessel (not shown).

  The raw material supply unit 6 includes a raw material supply tank 6c for holding an additional silicon raw material 7 made of, for example, stainless steel, and an introduction member for guiding the additional silicon raw material 7 to the mold 1 below the raw material supply tank 6c. 6a is provided. The introduction member 6a has a through hole h, and an opening / closing port 6b is provided between the through hole h and the additional silicon raw material supply tank 6c.

  When the additional silicon raw material 7 is supplied into the mold 1, the additional silicon raw material 7 can be supplied through the introduction member 6a by opening the opening / closing port 6b from the raw material supply tank 6c. The heating unit 4 is a resistance heating type heater or an induction heating type coil.

In this case, the density of the silicon ingot (solid) is 2.33 g / cm ³ whereas the density of the silicon melt (liquid) is 2.55 g / cm ^3, when moving from a liquid to a solid In consideration of the volume expansion of about 9% occurring in the above, it is desirable to add the additional silicon raw material so that the silicon melt becomes about 90% of the height of the mold 1.

  And in order to make silicon melt into a silicon ingot, the casting_mold | template 1 is heat-removed using the cooling part 2 from the lower part. By removing heat from the lower part of the mold 1, unidirectional solidification of the silicon ingot is realized.

  However, since the silicon melt 5 in the mold has a large heat removal from the liquid surface, the liquid surface of the silicon melt 5 is first solidified, and the liquid silicon melt is left inside. Thereafter, the silicon melt 5 left in the mold is solidified and expanded, and the surface of the silicon ingot is erupted, and the silicon ingot is cracked. In order to prevent this problem, it is desirable to heat the liquid surface of the silicon melt 5 by the upper heating part 4a located above the silicon melt 5 so as not to solidify the liquid surface.

  In this way, the silicon melt 5 can be unidirectionally solidified by removing heat from the lower part of the mold 1 by contacting or approaching the cooling unit 2 provided at the lower part of the mold 1. At this time, as the cooling unit 2, for example, a metal plate such as stainless steel (SUS) can be used, and a coolant such as water is circulated in the inside to effectively remove the silicon melt 5 inside the mold 1. It is configured to remove heat. Note that the silicon ingot obtained by cooling and solidification is cut and washed to a desired size and used as a polycrystalline silicon substrate material for solar cells.

  Here, the first embodiment for carrying out the silicon casting method of the present invention will be described in detail.

  In the first embodiment according to the present invention, while the silicon raw material held in the mold is heated and melted, the silicon melt is formed by supplying the additional silicon raw material, and the timing of supplying the additional silicon raw material is as follows: It was when the silicon melt and unmelted silicon were mixed in the mold.

FIG. 3A shows a state of the silicon casting method according to the first embodiment of the present invention. So that the unmelted silicon raw material floats on the liquid surface of the silicon melt for less than the density it is of the silicon melt density of unmelted silicon raw material ^{(2.33g / cm 3) (2.55g} / cm 3) . As shown in FIG. 3A, the additional silicon material 7 is dropped toward the silicon melt 5 in the mold by dropping the additional silicon material 7 onto the solid unmelted silicon material 5a. Scattering of the silicon melt 5 to the inner surface of the mold can be suppressed.

  Since the unmelted silicon raw material is present in the silicon melt in the mold as shown in FIG. 3 (a), the temperature of the silicon melt in the mold is within the temperature distribution of the silicon melt. In particular, the temperature is low. Therefore, even when an additional silicon raw material is charged, it is possible to suppress the release material from being peeled off due to the rapid cooling of the silicon melt in the mold.

  Thus, in the present invention, when supplying an additional silicon raw material, the mold material and the mold release material coated on the inner surface of the mold material are not heated more than necessary. Further, by supplying an additional silicon raw material on the unmelted silicon raw material, it is difficult for the silicon melt to splash and the silicon melt can be prevented from adhering to the release material.

  As a result, it is possible to eliminate the problem that the release material damaged due to peeling is taken into the silicon melt or the silicon ingot is cracked.

  In addition, as long as silicon melt and unmelted silicon are mixed in the mold, the effect of the present invention can be obtained at any time by supplying silicon raw material, but immediately before the unmelted silicon melts and disappears. It is preferable to supply an additional silicon raw material because many additional silicon raw materials can be supplied into the mold.

  Further, not only when the silicon raw material 5 charged in advance is melted and the additional silicon raw material 7 is supplied, but also when the additional silicon raw material 7 is supplied and the additional silicon raw material 7 is supplied again. It may be supplied when the silicon melt and unmelted silicon are mixed in the mold.

  Next, a second embodiment for carrying out the silicon casting method of the present invention will be described.

  In the second embodiment according to the present invention, while the silicon raw material held in the mold is heated and melted, an additional silicon raw material is supplied to form a silicon melt, and the timing of supplying the additional silicon raw material is within the mold. From the state where the silicon melt and the unmelted silicon raw material were mixed together, it was immediately after the unmelted silicon raw material melted and disappeared.

  The term “immediately according to the present invention” refers to 60 seconds when the unmelted silicon material melts and disappears from the state where the silicon material and unmelted silicon material are mixed in the mold.

  FIG. 2 is a schematic diagram showing the silicon casting method of the second embodiment, wherein 1 is a mold, 2 is a cooling unit, 3a is a heat insulating wall, 3b is a movable door, 4a is a mold upper heating unit, 4b. Is a mold side heating part, 5 is a silicon raw material or silicon melt formed by melting silicon raw material, 6 is a raw material supply part, 6a is an introduction member, 6b is an opening / closing port, 6c is a raw material supply tank, and 7 is an additional The silicon raw material 8 is a release material.

  FIG. 3B shows a silicon casting method according to the second embodiment of the present invention. As shown in FIG. 3B, the timing for supplying the additional silicon raw material 7 is immediately after the molten silicon raw material melts and disappears from the state in which the silicon melt and the unmelted silicon raw material are mixed in the mold. did. By supplying the additional silicon raw material in this way, the additional silicon raw material 7 is formed at a temperature lower than the case where the silicon raw material is melted by general melting in the mold (about 1500 ° C.), that is, at a temperature near the melting point of the silicon raw material. Therefore, the influence of the heat history on the mold material and the mold release material can be suppressed. Further, even if the silicon raw material in the mold is all in the melt state, the silicon melt 5 is scattered on the inner surface of the mold because the viscosity η of the silicon melt is low at a temperature near the melting point of the silicon raw material. Can be suppressed.

  Further, if the viscosity η of the silicon melt 5b in the mold 1 after the unmelted silicon raw material melts and disappears is 0.8 mPa · s or more, the silicon melt 5 is effectively scattered on the inner surface of the mold. This can be suppressed. Therefore, if the mold 1 is heated so as to maintain the above viscosity range, the silicon material can be melted even when the additional silicon raw material 7 is charged without heating the mold material and the mold release material 8 coated on the inner surface of the mold material more than necessary. The viscosity of the liquid 5 as a whole, particularly the liquid surface, becomes high, and the silicon melt 5 is difficult to splash. Moreover, the peeling damage of the mold release material 8 resulting from a heat history can also be prevented.

  Therefore, it is possible to eliminate the problem that the release material 8 which is peeled and damaged is taken into the silicon melt 5 or cracks are generated in the silicon ingot.

  However, when the viscosity η of the silicon melt is less than 0.8 mPa, the viscosity is low. Therefore, when the additional silicon raw material 7 is supplied, the silicon melt 5 is likely to splash on the liquid surface and scatters on the inner surface of the mold. Since there is a problem that it is easy, it is preferable to set it immediately after the unmelted silicon raw material melts and disappears as much as possible.

  Here, as a means for measuring the viscosity, for example, there is a vibration type viscometer described in Patent Document 3, and the viscosity η of the silicon melt has a correlation between the silicon melt temperature and the viscosity η. The silicon melt temperature may be calculated as an index.

  In the first and second embodiments according to the present invention, the silicon raw material held in the mold mainly using contact temperature measurement using a thermocouple or non-contact temperature measurement using an infrared radiation thermometer or the like. It is desirable to supply the additional silicon raw material 7 by measuring the temperature or the silicon melt temperature to calculate the rate of temperature rise.

  FIG. 4 shows a schematic sectional view of an embodiment of the silicon casting apparatus according to the present invention. With this apparatus, the first embodiment and the second embodiment according to the present invention described above can be implemented.

  1 is a mold, 2 is a cooling means, 3a is a heat insulating wall, 3b is a movable door, 4a is a mold upper heating means, 4b is a mold side heating means, and 5 is a silicon raw material or silicon formed by melting a silicon raw material Melt, 6 is a raw material supply means, 6a is an introduction member, 6b is an opening and closing port, 6c is a raw material supply tank, 7 is an additional raw material, 8 is a release material, 9 is a temperature measuring means, 10 is a display, and 11 is a calculator. , 12 are raw material supply control means.

  As shown in FIG. 4, a temperature measuring means 9 such as a thermocouple is provided to determine the temperature of the silicon raw material in the mold or the silicon melt 5. Then, the temperature of the temperature measuring means 9 is taken into the calculator 11 through the display 10 and is calculated. Thereafter, the state of the silicon raw material in the mold can be grasped based on the temperature of the silicon raw material in the mold 1 obtained by the computing unit 11 or the temperature increase rate which is the temperature per unit time of the silicon melt temperature. . The state of the silicon raw material in the mold is a state where the silicon melt and the unmelted silicon raw material are mixed, or immediately after the unmelted silicon raw material melts and disappears is set as the timing for supplying the additional silicon raw material 7. In addition, a mechanism for automatically opening and closing the raw material supply means 6 b is provided by sending an open / close signal from the raw material supply control means 12 to the raw material supply means 6.

  For example, a programmable controller (sequencer) or the like is used as a mechanism for supplying the additional silicon raw material 7 described above. Such a programmable controller is very well known in the field of machine control, and can be used as a unit by combining modules having a single function, or using a single module specializing in functions. Composed. In the case of the arithmetic unit 11 according to the present invention, the output from the temperature measuring means 9 is, for example, an analog voltage value (for example, a voltage that becomes a value of 1/100, such as 10 V when the temperature value is 1000 ° C. Therefore, modules having a function of converting them into digital values may be used in appropriate combination.

  Further, if the additional silicon material 7 is supplied following the temperature change by PID control, the additional silicon material 7 can be supplied with higher accuracy. For example, if the angle of the opening / closing port 6b is changed based on the temperature information obtained from the temperature measuring unit 9 by linking the material supply unit 12 and the opening / closing port 6b, the supply amount of the additional silicon material is adjusted. can do.

  Then, if the heating temperature of the mold upper part heating means 4a and the mold side part heating means 4b is adjusted based on the information obtained by the temperature measuring means 9, the mold material and the release material 8 that are also the cause of the release material 8 peeling. It is also possible to minimize the heat history.

  Furthermore, instead of using a programmable controller, a general-purpose personal computer may be provided with an interface mechanism for control, and the same result can be obtained.

  Here, a graph of the temperature profile of the temperature measuring means 9 and the temperature increase rate calculated by the calculator 11 is shown in FIG. In FIG. 5, a straight line shows a temperature change, and a plot shows a temperature increase rate. For example, the temperature of the silicon raw material held in the mold or the silicon melt temperature is measured every 5 seconds, and the temperature change rate is obtained by time-differentiating the amount of change in this temperature.

  The temperature measuring means 9 preferably directly measures the silicon raw material 5 or the silicon melt 5 inside the mold 1, but the additional silicon raw material 7 is measured by measuring the temperature of the mold 1 and taking into account the thermal conductivity of the mold member. There is no problem even if the timing of supplying is delayed. Further, the measurement object of the temperature measuring means 9 may be the silicon melt 5 and the silicon raw material 5, or the intermediate state between the silicon raw material 5 and the silicon melt 5.

  As shown in FIG. 5, the temperature of the mold 1 rises with the heating of the mold upper heating means 4a (A section), and when the temperature in the furnace becomes stable, the temperature hardly changes. section). Thereafter, when the silicon raw material 5 in the mold 1 is almost melted, the rate of temperature rise is slightly faster (C section), and when the silicon raw material 5 is completely melted, the temperature rise rate is rapidly increased (D section).

  Therefore, when the additional silicon raw material 7 is supplied in a state where the silicon melt 5 and the unmelted silicon 5a are mixed in the mold, a signal for automatically opening the raw material supply tank 6b in the section B or C is sent. The additional silicon raw material 7 can be supplied before the silicon raw material 5 is completely melted. In particular, if the additional silicon material 7 is added when moving from B to C, more additional silicon material 7 can be supplied.

  In the case where the silicon melt 5 and the unmelted silicon raw material 5a are mixed in the mold and the additional silicon raw material is supplied immediately after the unmelted silicon raw material 5a melts and disappears, the section C to D is used. What is necessary is just to supply the additional silicon raw material 7 within 60 seconds after moving.

  In the present invention, when the silicon raw material 5 is heated and melted, it is desirable to preferentially heat and melt the silicon raw material 5 on the inner wall surface of the mold. FIG. 6 shows the temperature distribution in the mold when heated from the side of the mold.

  1 is a mold, 4a is a mold upper part heating means, 4a is a mold side part heating means, 8 is a release material, and A, B, and C show the temperature distribution of the silicon melt in the mold. Here, when the silicon raw material 5 held in the mold is heated and melted, the mold upper part heating means 4a surrounded by a dotted line in FIG. 6 is not necessarily required, and it may not be heated from the upper part of the mold. Moreover, although it may be heated from the upper part of the mold, it is desirable that the amount of heat heated from the side of the mold is larger.

  As shown in FIG. 6, when heated from the side of the mold, the temperature distribution of the silicon melt is highest in the A region of the inner wall surface of the mold. And a mold center part and an opening part become C area | region where temperature is lower than A area | region. The B region is the temperature between the A region and the C region. According to the temperature distribution shown in FIG. 6, the inner wall surface of the mold contains a large amount of liquid in the form of a silicon melt 5, and an unmelted solid silicon raw material 5 exists near the center of the mold opening. Therefore, if the silicon material on the inner surface of the mold is preferentially heated by the mold side heating unit 4b so that the position where the additional silicon material 7 is dropped is near the center of the mold 1, the additional silicon material 7 is introduced. Even if it becomes on unmelted silicon, it can reduce that silicon melt 5 jumps.

  Therefore, the mold release material on the inner surface of the mold caused by the rapid temperature drop in the mold accompanying the supply of the additional silicon raw material and the adhesion of the silicon melt can suppress peeling damage. The cause of the peeling of the release material 8 caused by the silicon melt 5 adhering to the inner surface of the mold, and the peeling damage of the semiconductor ingot based thereon can be further suppressed.

  In the first embodiment and the second embodiment according to the present invention described above, when the additional silicon raw material 7 is supplied, the release material 8 coated on the inner surface of the mold is not heated more than necessary. The thermal history of the mold material and the release material 8 can be suppressed, and peeling damage of the release material 8 when the additional silicon raw material 7 is charged can be prevented. Further, it is possible to suppress the problem that the release material 8 that has been peeled and damaged is taken into the silicon melt 5 or cracks are generated in the silicon ingot.

  As a result, it is possible to provide a casting method that suppresses the yield reduction of the silicon ingot due to cracks in the mold 1 and mixing of the mold release material 8 without causing the mold release material 8 on the mold inner surface to be peeled and damaged.

  It should be noted that the embodiment of the present invention is not limited to the above-described example, and it is needless to say that various modifications can be made without departing from the gist of the present invention.

  For example, the position where the additional silicon raw material 7 is supplied does not have to be located at the center of the mold 1, so that the mold 1 and the additional silicon raw material 7 do not collide and the additional silicon raw material 7 does not spill from the mold 1. It can be anywhere.

  Further, the mold 1 used in the casting apparatus may be a divided mold 1. Even when such a divided mold 1 is used, the mold release material 8 applied to the inner wall surface of the upper part of the mold 1 is used. In order to use it effectively, it is desirable to supply the additional silicon raw material 7 after melting the silicon raw material 5 in the mold 1, and the same effect can be obtained by using the silicon casting method of the present invention.

  Moreover, if additional charging is repeated so that the amount of the melt when all the silicon raw materials are melted is 90% of the effective depth of the mold, the silicon melt can be converted into a silicon melt when cooled and solidified. Compared to the effective depth of the mold 1, it is ideal because the height of the resulting silicon ingot is about 10% larger.

It is a cross-sectional schematic diagram which shows 1st embodiment of the silicon | silicone casting method of this invention. It is a cross-sectional schematic diagram which shows 2nd embodiment of the silicon casting method of this invention. (A) is a conceptual diagram of the silicon casting method shown in FIG. 1, (b) is a conceptual diagram of the silicon casting method shown in FIG. It is a cross-sectional schematic diagram which shows one Embodiment of the silicon casting apparatus which concerns on this invention. It is a graph of the temperature increase rate calculated by the calculator of FIG. This is a temperature distribution in the mold when heated from the side of the mold. It is a cross-sectional schematic diagram which shows the silicon | silicone casting apparatus in the conventional melt in a mold. It is a cross-sectional schematic diagram which shows the conventional additional raw material supply type silicon casting apparatus.

Explanation of symbols

1: Mold 2: Cooling means 3, 3a: Thermal insulation wall 3b: Movable door 4: Mold heating means 4a: Mold upper heating means 4b: Mold side heating means 5: Silicon raw material (silicon melt)
6: Raw material supply means 6a: Introducing member 6b: Opening / closing port 6c: Raw material supply tank 7: Additional silicon raw material 8: Release material 9: Temperature measuring means 10: Display 11: Calculator 12: Raw material supply control means

Claims (2)

A silicon casting method for cooling and solidifying a silicon melt held in a mold,
The silicon melt is formed by supplying additional silicon material while heating and melting the silicon material held in the mold,
Said additional timing for supplying the silicon raw material, the silicon melt and the unmelted silicon in the case of a mixed state in said mold,
A silicon casting method for calculating a temperature rising rate from a temperature of a silicon raw material held in the mold or a silicon melt temperature and determining a timing for supplying the additional silicon raw material based on the temperature rising rate . A silicon casting method for cooling and solidifying a silicon melt held in a mold,
The silicon melt is formed by supplying additional silicon material while heating and melting the silicon material held in the mold,
The timing of supplying the additional silicon raw material is when the silicon melt and unmelted silicon are mixed in the mold.
Wherein the silicon raw material held in the mold when is heated and melted, Cie silicon casting method preferentially by heating and melting the silicon raw material of the mold inner wall.

Images