JP5077966B2 - Method for producing silicon ingot - Google Patents

Description

  The present invention relates to a method for manufacturing a silicon ingot, and more particularly to a method for manufacturing a single crystal silicon ingot and a polycrystalline silicon ingot suitable for manufacturing a solar cell.

  Solar cells that convert light energy such as sunlight into electrical energy are expected to develop significantly as energy technologies that can cope with global environmental problems. Solar cells with various structures and configurations have been developed. Among them, a crystalline silicon solar cell using a polycrystalline or single crystal silicon wafer is one of the oldest and has a high conversion efficiency. The production volume is still increasing because it is obtained and the manufacturing cost is relatively low.

  At present, single crystal silicon ingots for obtaining silicon wafers for solar cells are manufactured by reusing waste materials produced when manufacturing semiconductor wafers from semiconductor silicon ingots.

  For example, Patent Document 1 describes that a silicon single crystal that is out of specification in the production of a silicon single crystal substrate for a semiconductor device is used for the production of a silicon single crystal substrate for a solar cell. Patent Document 2 describes that a raw material for a solar cell is obtained from silicon remaining in a quartz crucible after manufacturing silicon for semiconductors.

  Such reuse is possible because the purity (impurity concentration) required for solar cell silicon is higher than that for semiconductor silicon, as shown in the table published in Non-Patent Document 1 (Table 1 below). This is because it is sufficiently low.

  There are various methods for producing a single crystal silicon ingot used for a semiconductor device or a solar cell, and one of them is the Czochralski (Cz) method. The Cz method is a method in which a single crystal ingot is pulled up from a silicon melt obtained by melting raw material silicon.

  The raw material silicon in the Cz method is a rod-like or piece-like material produced by the Siemens method for the production of silicon ingots for semiconductors, or an irregular shape such as a lump that is crushed. In many cases, it is a polycrystalline silicon, and for manufacturing a silicon ingot for a solar cell, the end material at the time of manufacturing a silicon ingot for a semiconductor is often used as described above.

  Of course, it is possible to use high-purity polycrystalline silicon by the Siemens method as a raw material for solar cells, but silicon for semiconductors (either high-purity polycrystalline silicon itself or scraps from its manufacturing process) When used as battery silicon, more doping is required to obtain silicon crystals having a resistivity of usually about 3-6 Ωcm.

  This “doping” is described in detail in Non-Patent Document 2, and an appropriate dopant (impurities such as boron and phosphorus) is added to the raw material polycrystalline silicon using a correlation graph between impurity concentration and resistivity. There is a need.

  Along with this, in the single crystal silicon ingot for solar cells manufactured by the Cz method, a dopant concentration distribution occurs in the length direction (vertical direction when pulled up).

It is known that the dopant concentration distribution C _s when the solidification rate of silicon is g is represented by the following formula (1).
C _s = k × C ₀ × (1-g) ^k−1 (1)
In equation (1), k is the equilibrium segregation coefficient, C ₀ is the initial dopant concentration in the silicon melt, and the equilibrium segregation coefficient of boron (B) most commonly used as a p-type dopant is 0.8, n The equilibrium segregation coefficient of phosphorus (P) most commonly used as a type dopant is 0.35.

  In the Cz method, a single crystal ingot is grown while maintaining the balance between the heat capacity of the silicon melt and the solidification phenomenon, but it is difficult to use up all of the silicon melt in the crucible, In this case, the crucible residue contains an excessive dopant concentration not only for semiconductors but also for solar cells, which is higher than the initial dopant concentration of the silicon melt.

  For this reason, unlike the silicon remaining in the quartz crucible after the production of silicon for semiconductors, the crucible residue after the production of the single crystal silicon ingot for solar cells has been conventionally discarded without being reused.

JP 2002-76382 A JP 2002-37617 A

Incorporated administrative agency, National Institute of Advanced Industrial Science and Technology (AIST), Photovoltaic Power Generation Research Center, Supplementary Materials “Types and Abbreviations of Silicon”, [online], last updated on December 26, 2008, [April 20, 2009 Day search], Internet <URL: http://unit.aist.go.jp/rcpv/ci/about_pv/supplement/SiTypes.html> Fumio Shimura, “Semiconductor Silicon Crystal Engineering”, 1st Edition, Maruzen Co., Ltd., September 30, 1993, p. 40

  The present invention has been made in view of such circumstances, and by providing a resistivity range of a single crystal silicon ingot by the Cz method, it is possible to provide a silicon ingot manufacturing method that can reuse a crucible residue that has been conventionally discarded. Objective.

  In the method for producing a silicon ingot according to the present invention, a single crystal ingot for a solar cell is obtained from a silicon melt containing a first dopant that defines a first conductivity type held in a crucible by a Czochralski method. And a step of forming an ingot using a residue in the crucible (crucible residue) after pulling up the single crystal silicon ingot as a raw material of the polycrystalline silicon ingot.

  The concentration of the first dopant is adjusted so that the resistivity at the head portion of the single crystal silicon ingot is 20 Ω · cm or more, and the resistivity of the tail termination portion of the single crystal silicon ingot is 8 Ω · cm or more and 16 Ω · cm or less. It is preferable to pull up the single crystal silicon ingot to be used, and use the remaining crucible having a resistivity of 3 Ω · cm to 6 Ω · cm as a raw material of the polycrystalline silicon ingot.

  In addition, the method includes a step of adding a second dopant that defines a conductivity type different from that of the first dopant to the crucible residue, whereby the resistivity of the polycrystalline silicon obtained from the crucible residue is 1 Ω · cm. It is preferable to adjust to 2 Ω · cm or less.

  Furthermore, it is preferable that the first dopant is phosphorus and the second dopant is boron.

  The silicon ingot manufacturing method of the present invention is a silicon ingot manufacturing method including a step of manufacturing a p-type polycrystalline silicon ingot by a casting method using a crucible residue as a raw material of the polycrystalline silicon ingot.

  In the silicon ingot manufacturing method of the present invention, a crucible residue is used as a raw material for a polycrystalline silicon ingot, and a p-type polycrystalline silicon ingot having a resistivity of 1 Ω · cm to 2 Ω · cm is manufactured from the raw material by a casting method. Preferably, the method includes a step and a step of manufacturing a solar battery cell from the polycrystalline silicon ingot.

  Since the crucible residue, which is a by-product during the production of a single crystal silicon ingot by the Cz method, can be reused as a raw material for a cast ingot for a polycrystalline silicon solar cell, the use efficiency of the raw material silicon can be improved, and a low-cost solar cell Can be provided.

It is sectional drawing which shows typically the crystal growth of the single crystal ingot by CZ method. It is sectional drawing which shows typically an example of the shape of a single crystal ingot. (A) is a graph which shows an example of the relationship between the solidification rate and resistivity which concern on embodiment of this invention, and a residual phosphorus density | concentration, (b) is the solidification rate, resistivity and residual phosphorus which concern on embodiment of this invention. It is a graph which shows another example of the relationship of density | concentration. It is sectional drawing which showed the conventional solar cell typically.

  Hereinafter, the present invention will be described in more detail. In the following description of the embodiments, the description is made with reference to the drawings. In the drawings of the present application, the same reference numerals denote the same or corresponding parts.

  The present invention includes a step of forming a single crystal silicon ingot by the Czochralski (Cz) method. FIG. 1 schematically shows a cross-sectional view of crystal growth of a single crystal ingot by the Cz method. In the growth of the single crystal ingot by the Cz method, a quartz crucible 1 having a cylindrical shape is filled with a raw material containing polycrystalline raw material silicon and a dope material, and the raw material is melted from the outside of the quartz crucible 1 to obtain a melt state. The silicon (silicon melt 21) is formed. Thereafter, a seed crystal 23 (also referred to as a seed crystal) set on the seed holder 22 is immersed in the silicon melt 21, and then the seed crystal 23 is slowly pulled up while rotating to grow a single crystal ingot 24.

  FIG. 2 shows a schematic cross-sectional view of an example of the shape of a single crystal ingot formed by pulling up by the Cz method. In pulling up a single crystal ingot of the Cz method, first, after contacting a seed crystal with a silicon melt, a diameter of 3 mm is used in order to eliminate dislocations propagated from slip dislocations generated in the seed crystal at a high density by thermal shock. A necking step (squeezing step) for forming the narrowed portion 101 once narrowed to about ˜5 mm is performed. FIG. 2 shows the shape of a single crystal ingot already cut from the seed crystal, and the seed crystal is not shown. The length of the narrowed portion 101 is not particularly limited, but is usually 5 cm or more and 30 cm or less in order to completely prevent the influence of the dislocation.

  After the squeezing step, the crystal diameter is increased to a desired diameter (10 cm to 30 cm, or about 4 inches to 12 inches). Such enlargement of the crystal diameter can be achieved by a known method by slowing the pulling rate. Then, such enlargement of the crystal diameter creates a crown 102 and a shoulder portion 103 (also referred to as a shoulder portion) shown in FIG. Thereafter, a main body portion 104 which is a single crystal straight cylinder (also referred to as a body or a constant diameter portion) having a desired diameter is grown to a desired length. Here, the uppermost end of the constant diameter portion (directly after the transition from the shoulder portion to the constant diameter portion) is hereinafter referred to as a head or a head portion (indicated by H in FIG. 2). The crown 102 refers to a portion formed at a time zone where the pulling speed is slow, and the shoulder portion 103 refers to a portion between the crown 102 and the head portion. The portion formed in the time zone in which preparation for forming the straight body portion is performed, and this time is about 5 minutes.

  After the body portion 104 is grown to a desired length, as shown in FIG. 2, the diameter of the single crystal body is gradually reduced from the constant diameter portion as shown in FIG. There is a method to keep away from the propagation of thermal dislocations when separated from the silicon melt. Usually, the inverted conical part at one end of this single crystal is called tail 105, and the length in the cone height direction escapes from the propagation of thermal dislocations occurring at the end of the apex, It is preferable that the diameter is at least equal to or larger than the diameter of the main body portion 104.

  In the manufacturing method of the present invention, in the step of pulling up the single crystal silicon ingot, the concentration of the first dopant in the silicon melt is adjusted so that the resistivity at the head portion H of the single crystal silicon ingot is 20 Ω · cm or more. It is preferable to do. When the resistivity in the head portion H is 20 Ω · cm or more, the resistivity on the tail side can be increased as follows, which is preferable. In this case, the ingot is pulled up so that the resistivity at the tail end of the single crystal silicon ingot is 8 Ω · cm or more and 16 Ω · cm or less, and the resistivity of the residual crucible is 3 Ω · cm or more and 6 Ω · cm or less. Is desirable. When the resistivity in the head portion H is 20 Ω · cm or more, the portion to be pulled up as a single crystal is long, and the yield of the single crystal manufactured from a single filling amount of raw material silicon into a crucible is good. As a result, the manufacturing cost can be reduced.

  The relationship between such dopant concentration and resistivity will be described below. In the single crystal silicon ingot for solar cells manufactured by the Cz method, a dopant concentration distribution is generated in the length direction (vertical direction at the time of pulling).

The concentration distribution C _s of the dopant when the solidification rate of silicon is g is expressed by the formula (1) as described above.
C _s = k × C ₀ × (1-g) ^k−1 (1)
(In equation (1), k is the equilibrium segregation coefficient, and C ₀ is the initial dopant concentration in the silicon melt.) The solidification rate is the mass of crystallized silicon relative to the total mass of raw silicon contained in the silicon melt. The ratio of The equilibrium segregation coefficient of boron (B), which is most commonly used as a p-type dopant, is 0.8, and the equilibrium segregation coefficient of phosphorus (P), which is most commonly used as an n-type dopant, is 0.35. is there.

In order to set the resistivity within the above range, the relationship between the dopant concentration and the solidification rate is determined in the production of a silicon ingot having a predetermined shape, and the head portion and tail end portion described later are adjusted by adjusting the dopant concentration based on these relationships. The resistivity may be adjusted to be in a desired range. For example, in order to make the resistivity of the head part 20 Ω · cm or more, when phosphorus is used as a dopant, 0.5 to 2 g of dopant (about 0.1 to 0.5 mΩ · cm in resistivity) is applied to about 60 kg of silicon. It is preferable to introduce a silicon fragment containing phosphorus at a high concentration (10 ⁻¹⁹ cm ⁻³ unit). Further, in order to increase the resistivity of the head portion, the dopant input amount is decreased.

  Using an n-type single crystal silicon ingot with phosphorus as a representative example, the resistivity (phosphorus concentration) of the ingot and the resistivity of the remaining melt (phosphorus concentration) will be examined.

  3A and 3B, in the n-type single crystal pulling by the Cz method using phosphorus as a dopant, the resistivity of the head part is 30 Ω · cm (FIG. 3A) and 20 Ω · cm ( FIG. 3B shows the change in resistivity inside the ingot with respect to the solidification rate. Further, the concentration of phosphorus remaining in the quartz crucible (residual phosphorus concentration) was examined. Table 2 shows the residual phosphorus concentration when the head portion resistivity is 30 Ω · cm, and Table 3 shows the residual phosphorus concentration when the head portion resistivity is 20 Ω · cm. The “resistivity” in the present invention is a value at 25 ° C., and the measuring method is based on the four-probe method. Measurement by the four-probe method can be performed using, for example, a resistivity measuring device RT series manufactured by Napson. Further, the “concentration” of a dopant such as phosphorus in the present invention can be measured by the ICP method, for example, by a measuring device ICPE-9000 manufactured by Shimadzu Corporation.

  In FIGS. 3A and 3B, and Tables 2 and 3, the n-type resistivity (“resistivity” in the figure and table) is displayed as a negative value with a minus sign. In FIGS. 3 (a) and 3 (b), the range of the solidification rate sandwiched between solid lines parallel to the resistivity axis is an effective part of the body, and the range where the solidification rate is larger than the one-dot chain line is the remaining crucible. It is a part corresponding to.

  As can be seen from FIGS. 3 (a) and 3 (b), since the segregation coefficient of phosphorus is 0.35, the ingot resistivity gradually decreases from 30Ω · cm or 20Ω · cm together with the solidification rate ( That is, the phosphorus concentration is increased), and an ingot having a wide resistivity is manufactured. On the other hand, from the viewpoint of manufacturing solar cells and modules using a wafer processed from an ingot, the resistivity range of the ingot is preferably as narrow as possible.

  The inventors of the present invention not only narrowed the resistivity range as much as possible, but also intensively studied a method of applying the crucible residue to another application in order to improve the silicon use efficiency. It was found that if the resistivity of the residual crucible is 3 Ω · cm or more and 6 Ω · cm or less, it can be recycled as a cast ingot raw material for polycrystalline silicon solar cells. As can be seen from FIGS. 3 (a) and 3 (b), this is equivalent to rounding up the resistivity at the final end (tail end) of the tail portion of the ingot to 8Ω · cm or more and 16Ω · cm or less. is there. In addition, when the resistivity of the tail end portion is in the above range, the yield of the body part from the bottom of the head part to the upper end of the tail part can be taken up to a maximum of 80% when looking at the solidification rate. In the production of the above, the rounding up with the above resistivity does not cause a large loss, and the remaining 20% of the crucible is not discarded as in the prior art, and can be recycled into a polycrystalline ingot for solar cells.

  That is, a polycrystalline silicon ingot can be obtained from the crucible residue by a casting method, and a polycrystalline solar battery cell and a solar battery module can be produced. Thus, low-cost solar batteries can be mass-produced. As a result, the utilization efficiency of the raw material silicon is considered to be almost 100%.

  Furthermore, as a preferable method for manufacturing a polycrystalline ingot for a solar cell, doping n-type silicon doped with phosphorus having a resistivity of about 3 Ω · cm by doping boron with an order of magnitude of 1 Ω · A method of using p-type silicon of cm to 2 Ω · cm can be mentioned.

  Note that the silicon remaining in the crucible is partially fixed to the quartz crucible, and therefore, silicon oxide may be dissolved and recovered by hydrofluoric acid treatment.

  By the above manufacturing method, it is possible to recycle the crucible residue, which has been a waste material, so that a polycrystalline silicon solar cell can be manufactured at low cost.

  Examples The present invention will be described more specifically with reference to examples. However, the present invention is not limited to these examples.

A single crystal silicon ingot having a body part diameter of 16 cm and a tail part length of 16 cm was produced by the Cz method. The conditions were as follows.
<Conditions>
Raw material silicon: high purity polycrystalline silicon by Siemens method and single crystal silicon milling material for semiconductor, 60kg per lot
Crucible: quartz crucible resistivity range with inner diameter of 18 inches: upper end of ingot 20 Ωcm, tail end 8 Ωcm
Impurity controlling impurities: phosphorus (P)
Residual hot water volume at the time of disconnection: 15 kg (during filling 60 kg) resistivity 3 Ωcm
Atmospheric gas: Argon pressure: 10 Torr
Pulling speed: 1.2 mm / min Average seed rotation speed: 16 rpm
Crucible rotation speed: 12rpm
For pulling up the single crystal ingot, first, raw silicon is put in a quartz crucible and melted to form a silicon melt. After bringing the single crystal silicon into contact with the silicon melt as a seed crystal, the pulling is started while rotating the seed crystal. Thereafter, a single crystal body portion, which is a constant diameter portion having a diameter of 16 cm, was grown through a necking (drawing) step, a crown (cone) forming step, and then a shoulder forming step. In the necking step, the diameter was reduced to 5 cm, and the tail portion was pulled up by gradually increasing the pulling speed and controlling the program to gradually increase the temperature. The solidification rate was 75%.

  After pulling up this single crystal ingot, the melting crucible is taken out for each lot, and the remaining crucible is recovered, and the resistivity is 1 Ω · cm to 2 Ω · cm by doping boron into a total of 300 kg for a total of 20 lots collected and casting. A p-type polycrystalline ingot was prepared.

  Next, this ingot was processed into a wafer of 125 mm × 125 mm × 200 μm to produce a polycrystalline silicon solar battery cell.

  The solar cell production method was as follows. FIG. 4 shows a schematic diagram of an example of the polycrystalline solar battery cell 11 manufactured in this example.

  First, the surface of a 125 mm × 125 mm × 200 μm p-type silicon substrate 12 is etched, and thereafter, thermal diffusion of phosphorus (P) is performed at 900 ° C. on one side surface to be a light receiving surface, and the surface resistance value is about 50Ω. An n + type diffusion layer 13 was formed.

  A silicon nitride film having a thickness of about 60 nm was formed thereon as an antireflection film 14 by plasma CVD. Next, an aluminum paste is screen-printed on the back surface opposite to the light-receiving surface, dried at 150 ° C., then placed in an IR firing furnace and fired at 700 ° C. in air, and the back surface field layer 15 (BSF layer) and aluminum An electrode 16 was formed. Next, silver paste was screen-printed in a pattern on the back surface and dried to form a collecting electrode 17.

  Further, a silver paste was screen-printed on the light receiving surface so as to form a fishbone pattern, dried, and then fired at 600 ° C. for 2 minutes in an oxidizing atmosphere to form a silver electrode 18. Finally, a polycrystalline solar cell 11 was obtained by coating a silver electrode with a solder layer (not shown).

  As a result, characteristics equivalent to those of a polycrystalline silicon solar cell using a high-purity silicon raw material were obtained, and the significance of the present invention was confirmed. Table 4 shows the average measurement results obtained by evaluating the current-voltage characteristics of the solar cells obtained by the above treatment with a solar simulator. Evaluation was performed under the conditions of AM1.5 and room temperature of 25 ° C.

  As is apparent from the results of Table 4, by producing a single crystal ingot by the Cz method with a resistivity that satisfies the scope of the present invention, the remaining crucible can be effectively used, and other single crystal end materials such as tail portions, etc. Considering the reuse of wastewater, 100% utilization efficiency including the reused portion was demonstrated. Note that the top, bottom, and side end materials of the cast ingot can be reused in the cast ingot with a waste amount of almost 0% by sandblasting the surface for several mm without discarding.

  Although the embodiments and examples of the present invention have been described as described above, it is also planned from the beginning to appropriately combine the configurations of the above-described embodiments and examples.

  It should be understood that the embodiments and examples disclosed herein are illustrative and non-restrictive in every respect. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  ADVANTAGE OF THE INVENTION According to this invention, the manufacturing method which can implement | achieve effective utilization of raw material silicon can be provided.

  DESCRIPTION OF SYMBOLS 1 Quartz crucible, 11 Polycrystalline solar cell, 12 Silicon substrate, 13 n <+> type | mold diffused layer, 14 Antireflection film, 15 Back surface electric field layer, 16 Aluminum electrode, 17 Current collecting electrode, 18 Silver electrode, 21 Silicon melt, 22 Seed holder, 23 seed crystal, 24 single crystal ingot, 101 throttle part, 102 crown, 103 shoulder part, 104 body part, 105 tail.

Claims (6)

A step of pulling up a single crystal silicon ingot for a solar cell by a Czochralski method from a silicon melt containing a first dopant defining a first conductivity type held in a crucible;
And a step of forming an ingot using a residue in the crucible after pulling up the single crystal silicon ingot as a raw material of the polycrystalline silicon ingot. Adjusting the concentration of the first dopant in the silicon melt so that the resistivity at the head portion of the single crystal silicon ingot is 20 Ω · cm or more; and
The single crystal silicon ingot is pulled up so that the resistivity at the tail end portion of the single crystal silicon ingot is 8 Ω · cm to 16 Ω · cm,
The method for producing a silicon ingot according to claim 1, wherein the residue used as a raw material of the polycrystalline ingot has a resistivity of 3 Ω · cm to 6 Ω · cm.   A step of adding a second dopant defining a conductivity type different from that of the first dopant to the residue, and a resistivity of the polycrystalline silicon ingot obtained from the residue by the step is 1 Ω · cm or more The manufacturing method of the silicon ingot of Claim 1 or 2 adjusted to 2 ohm * cm or less.   The method for producing a silicon ingot according to claim 3, wherein the first dopant is phosphorus and the second dopant is boron.   5. The silicon ingot according to claim 1, wherein the step of forming an ingot using the residue as a raw material of the polycrystalline silicon ingot includes a step of manufacturing a p-type polycrystalline silicon ingot by a casting method. Manufacturing method.   The step of forming an ingot using the residue as a raw material for a polycrystalline silicon ingot is a step of producing a p-type polycrystalline silicon ingot having a resistivity of 1 Ω · cm or more and 2 Ω · cm or less from the raw material by a casting method. And a method for producing a solar battery cell from the polycrystalline silicon ingot. 6. A method for producing a silicon ingot according to claim 1.

Images