JP2002104898A - Silicon crystal and silicon crystal wafer and method of manufacturing them - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a silicon crystal and silicon crystal wafer and a method of manufacturing them able to reduce manufacturing cost to a large extent and to prevent conversion efficiency from lowering caused by light deterioration. SOLUTION: The silicon crystal is added with Ga and phosphorus as a doping agent. The method of manufacturing silicon crystal is that after adding Ga and phosphorus into silicon molten liquid in a crucible by Czochralski method, a silicon single crystal is grown by making a seed crystal contact with silicon molten liquid and pulling it up while revolving it. The method of manufacturing the silicon crystal is that after adding Ga and phosphorus into the silicon molten liquid in the crucible by Bridgeman method, silicon polycrystal is grown by pulling down the crucible out of heating region.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a silicon crystal, a silicon crystal wafer and a method for producing the same, and more particularly, to a silicon crystal and a silicon crystal wafer useful as a material for a solar cell, and a method for producing the same.

[0002]

2. Description of the Related Art The technical background of a silicon crystal used for a solar cell as an example will be described below. Solar cells are broadly classified into two types, “silicon-based solar cells” and “compound semiconductor-based solar cells,” based on the type of semiconductor material used in the power generation unit. Furthermore, silicon-based solar cells are classified into “crystalline silicon-based solar cells” and “amorphous (amorphous) silicon-based solar cells”, and crystalline silicon-based solar cells are referred to as “silicon single-crystal-based solar cells” and “silicon-rich solar cells”. Crystalline solar cell "
are categorized.

[0003] Focusing on the conversion efficiency, which is the most important characteristic of a solar cell, compound semiconductor solar cells have recently reached the highest of nearly 25% among these, and silicon single crystal solar cells have recently reached around 20%. Followed by 5 to 1 for silicon polycrystalline solar cells and amorphous silicon solar cells.
It is about 5%. On the other hand, paying attention to material costs, silicon is the second most common element on the earth after oxygen, and is much cheaper than compound semiconductors. Therefore, silicon-based solar cells are more widely used.

[0004] Here, the "conversion efficiency" is a value indicating "the ratio of energy that can be converted into electric energy and extracted by the solar cell with respect to the energy of the light incident on the solar cell" and expressed as a percentage. (%) (Also referred to as photoelectric conversion efficiency).

[0005] In recent years, demand for solar cells has been expanding as one of clean energy due to environmental problems.
The high energy cost compared to general commercial power is an obstacle to its widespread use. In order to reduce the cost of the silicon crystal solar cell, it is a major issue to reduce the production cost of the substrate and further increase the conversion efficiency.

Next, a method of manufacturing a general silicon single crystal solar cell will be briefly described. First, in order to obtain a silicon wafer serving as a substrate of a solar cell, a Czochralski (CZ) method or a floating zone melting (FZ) method is used.
A cylindrical silicon single crystal ingot is made. Further, the ingot is sliced, processed into a thin wafer having a thickness of, for example, about 300 μm, and the wafer surface is etched with a chemical solution to remove processing distortion on the surface, thereby obtaining a wafer (substrate) serving as a solar cell material. Can be The wafer is subjected to an impurity (dopant) diffusion process to form a PN junction surface on one side of the wafer, and electrodes are attached on both sides. Finally, light energy loss due to light reflection on the incident surface of sunlight is reduced. A solar cell is completed by providing an anti-reflection film for the purpose.

In a solar cell, it is important to manufacture a solar cell having a larger area in order to obtain a larger current. As a method for obtaining a large-diameter silicon wafer which is a substrate material for manufacturing a large-area solar cell, a large-diameter silicon single crystal can be easily produced, and the strength of the produced single crystal is excellent. The CZ method is suitable. For this reason, the production of silicon single crystals for solar cells is mainly performed by the CZ method.

On the other hand, a silicon wafer used as a substrate material of a silicon single crystal solar cell has a substrate lifetime (LT), which is one of its characteristics, of 10 μm.
s or more, it cannot be used as a solar cell substrate, and in order to obtain a solar cell with high conversion efficiency,
The substrate LT is required to be preferably 250 μs or longer.

However, when a single crystal made by the CZ method, which is the mainstream of the current single crystal rod manufacturing method, is irradiated with intense light when the solar cell is processed into a solar cell, the LT of the solar cell substrate is reduced.
Therefore, sufficient conversion efficiency cannot be obtained to cause light degradation, and improvement in the performance of solar cells is also required.

When a solar cell is manufactured using this CZ method silicon single crystal, when strong light is applied to the solar cell, LT
It is known that the cause of the photodegradation due to the decrease is caused by boron (hereinafter referred to as B) and oxygen existing in the single crystal substrate. At present, the conductivity type of a wafer used as a solar cell is mainly P-type.
B is added as a dopant to the mold wafer.
The single crystal rod as a material of the wafer can be manufactured by a CZ method including pulling under a magnetic field (hereinafter, referred to as MCZ) or an FZ method. In the FZ method or the MCZ method, a single crystal rod is used. Since the manufacturing cost is higher than that of the normal CZ method, the semiconductor device is currently manufactured exclusively by the normal CZ method that does not apply a magnetic field that can produce a single crystal at a relatively low cost.

However, a high concentration of oxygen is present in the crystal produced by the CZ method, so that the LT characteristics are affected by B and oxygen in the P-type CZ silicon single crystal,
There is a problem that light degradation occurs.

In order to solve such a problem, the applicant of the present application has proposed in a previous application to use gallium (hereinafter, referred to as Ga) instead of B as a P-type dopant (Japanese Patent Application No. 2002-214,972). Hei 11-264549 and Japanese Patent Application 200
0-06435). By using Ga as a dopant in this manner, it is possible to prevent a decrease in the lifetime due to the influence of B and oxygen.

[0013]

However, although the effects of B and oxygen are eliminated by using Ga as a dopant, the crystal characteristics become unstable for the following reasons, and the production cost tends to increase. .

FIG. 1A shows the resistivity in the length direction of a silicon single crystal when Ga or B is used as a dopant, and FIG. 1B shows the respective dopant concentrations in the length direction of the silicon single crystal. Show. As is clear from FIG. 1, Ga has an extremely small segregation coefficient ko compared to B (ko of Ga = 0.008, ko of B = 0.8), and thus the lengthwise direction of the silicon crystal rod obtained by growing the crystal. However, there is a problem that a large difference occurs in the resistivity (or the dopant concentration) in the method. Further, Ga has an extremely low melting point as compared with B (melting point of Ga = 29.78 ° C., melting point of B = 2080 ° C.), so that it is difficult to handle and the evaporation rate is high.
(Ga evaporation rate = 2 × 10 ⁻³ cm / s, B evaporation rate = 8 × 10 ⁻⁶
cm / s), it is difficult to control the concentration of Ga to be doped into the silicon crystal, and there is a problem that the yield in the ingot becomes low because the dispersion in the ingot becomes large. In the case where the standardized resistivity or concentration range is narrow, the region falling within the range is considerably shorter in the case of Ga doping than in the case of B doping, so that the yield is low, and as a result, the manufacturing cost is reduced. Is high.

A silicon wafer, which is a main raw material of a silicon single crystal solar cell mainly used at present, is the same as a silicon wafer used in semiconductor devices such as integrated circuits (ICs) and memories. However, semiconductor devices are sold at a price of several hundred yen or more, especially for an integrated circuit, at a price of several thousand yen for a chip size of 1 cm 2 or less.
Compared with a semiconductor device, it is said that the cost per unit area of a silicon wafer for a solar cell must be reduced by two to four orders of magnitude, and there is a large problem in cost.

The present invention has been made in view of such problems. Even when a solar cell is manufactured using a silicon crystal, the manufacturing cost can be reduced, and the conversion efficiency and the lifetime are not reduced. It is an object of the present invention to provide a silicon crystal and a silicon crystal wafer capable of preventing a decrease in conversion efficiency due to light degradation and reducing variations in characteristics, and a method for manufacturing the same.

[0017]

According to the present invention, there is provided a silicon crystal, wherein gallium (Ga) and phosphorus are added as dopants. (Claim 1).

A cone, tail, defective part, or the like that cannot be used as a phosphorus-doped silicon crystal product can be reused as a raw material, and is much less expensive than a silicon crystal added only with Ga. It can be. Further, since the N-type dopant phosphorus is also doped, the variation in resistivity of the Ga-doped crystal can be suppressed, and the resistivity distribution in the crystal length direction can be reduced by G.
The crystal becomes flatter than a crystal to which only a is added, and a region within a specified resistivity range or concentration range becomes longer, so that the yield can be improved.

In this case, it is preferable that the concentration of phosphorus contained in the silicon crystal is not more than 1/5 of the concentration of Ga. As a silicon crystal used for a solar cell, the present applicant has proposed Ga instead of B as a dopant without photodegradation of conversion efficiency in the solar cell. However, even if a silicon crystal doped with phosphorus (P) is doped with Ga, the conductivity type is a P-type silicon crystal having a predetermined resistivity, and the conductivity type is changed from P-type to N-type. By setting the phosphorus concentration so as not to be inverted, a stable solar cell having high conversion efficiency without light degradation can be manufactured. If a large amount of phosphorus is doped, the conductivity type becomes N-type, so that the concentration of phosphorus contained in the silicon crystal of the present invention is 1/1 / Ga that is negligible. It is better to be 5 or less.

In this case, the concentration of Ga contained in the silicon crystal is 2 × 10 ^{17 to} 3 × 10 ¹⁴ atoms /
cm ³ is preferable (claim 3). Or
The resistivity of the silicon crystal is 20Ω · cm to 0.1Ω ·
cm (claim 4).

This is because a substrate having a low resistivity and a long lifetime is desired as a substrate for a solar cell, but a substrate having an extremely low resistivity of a substrate wafer has an auger (A) inside the substrate.
This is because the lowering of the minority carrier lifetime (lifetime) due to recombination occurs and the conversion efficiency decreases. Therefore, the amount of Ga contained in the silicon crystal of the present invention is adjusted so that the concentration of Ga is 2 × 10 ¹⁷ atoms / cm ³ or less, or the resistivity is 0.1 Ω · cm or more. Is preferred.

On the other hand, if the substrate resistivity is too high, a problem arises. When the substrate resistivity increases, power is consumed by the internal resistance of the solar cell when the solar cell is used,
Similarly, conversion efficiency is reduced. For these reasons, if used as a substrate material for solar cells, the concentration of Ga in the silicon crystal is 3 × ¹⁰ 14 atoms /
cm ³ or more, or a resistivity of 20 Ω · cm or less.

In this case, it is preferable that the interstitial oxygen concentration in the silicon crystal is not more than 16 ppma (JEIDA; Japan Electronics Industry Development Association Standard) (claim 5).

As described above, according to the present invention, even if oxygen is contained in the crystal, since the resistivity of the crystal is controlled by Ga without adding B, there is no light degradation, and the concentration of oxygen contained in the crystal is usually May be contained in the single crystal by the CZ method, and may be a usual concentration such as 16 ppma (JEIDA). Therefore, there is an advantage that it is not necessary to forcibly reduce the oxygen, it can be easily manufactured, and the crystal strength is high because of an appropriate amount of oxygen.

On the other hand, when the oxygen concentration is 16 ppma (JEID
To obtain a silicon single crystal wafer exceeding A), a silicon single crystal having a high oxygen concentration is required. However, to obtain a single crystal having an oxygen concentration higher than necessary, a crucible rotation during growing a single crystal must be performed at a high speed. For example, it is necessary to select manufacturing conditions that make it difficult to grow a single crystal. Under such growth conditions, a crystal that cannot be processed into a solar cell substrate can be formed because slip dislocation occurs in the single crystal during the growth of the single crystal, or the single crystal cannot be pulled straight and the crystal is deformed. Therefore, the production cost of the wafer is increased, and it is difficult to obtain an economic advantage. Therefore, the oxygen concentration of the wafer used in the present invention is 16 ppma (JEI
DA) It is preferred to be as follows. Further, by setting the interstitial oxygen concentration to 15 ppma (JEIDA) or less,
It is more preferable because the lifetime can be prevented from being deteriorated by the initial interstitial oxygen or the oxygen precipitate formed by the heat treatment.

In this case, the silicon crystal may be a silicon single crystal manufactured by the Czochralski method.

Conventionally, especially when the diameter of a single crystal used for a substrate is large, a crystal formed by the CZ method or the MCZ method tends to show a high oxygen concentration, so that a solar cell having high conversion efficiency is obtained. If present, F to reduce oxygen
It has been common practice to make a single crystal by the Z method or to use a small diameter single crystal of the MCZ method. However, it is almost impossible to produce a single crystal having a diameter exceeding 6 inches at the maximum by the FZ method, and it is difficult to produce a single crystal having a low oxygen concentration when the diameter exceeds 4 inches by the MCZ method. Large diameter single crystals have been unsuitable for obtaining highly efficient solar cells.

Further, the silicon single crystal of the present invention can obtain a stable substrate lifetime without being affected by oxygen contained in the single crystal by using Ga as a main dopant, and can further reduce the cost. Because it can be manufactured,
A single crystal rod having a large diameter can be used as a substrate wafer for a solar cell, and a solar cell with high conversion efficiency can be manufactured at low cost regardless of the diameter of the single crystal rod. In addition, a large-diameter wafer that is not currently used can be used as a solar cell substrate, so that the size of the solar cell itself can be increased, and the use of the solar cell can be further expanded. .

In this case, the silicon crystal may be a polycrystalline silicon manufactured by the Bridgman method. As described above, the conversion efficiency of a silicon polycrystalline solar cell is slightly lower than that of a silicon monocrystalline solar cell.However, the time required for production is short, and the cost can be reduced. It is even more effective if applied.

Further, the present invention is a silicon crystal wafer obtained by slicing the silicon crystal (claim 8).

When such a silicon crystal wafer doped with Ga and phosphorus is used as a substrate material for a solar cell, the phosphorus-doped silicon crystal can be reused as a raw material, so that the manufacturing cost can be reduced. Since B is not contained in the crystal, B and oxygen coexist, so that a deep level does not occur. By suppressing the amount of phosphorus, N-type inversion is prevented, and other effects occur. Since a reduction in the lifetime can be prevented, it is possible to obtain a long lifetime required for a solar cell even for a single crystal wafer containing high oxygen. As a result, an appropriate lifetime can be obtained even with a cell having a low resistivity, so that even a solar cell using a wafer having a high oxygen concentration can produce a high-performance solar cell without impairing the conversion efficiency. became. In addition, by containing oxygen moderately, a merit in use that the wafer strength is high can be obtained.

Further, the silicon crystal wafer of the present invention can be used for a solar cell (claim 9).

As described above, the Ga and phosphorus-doped silicon crystal wafer of the present invention is particularly useful when used for solar cells. Further, a silicon single crystal solar cell manufactured from such a Ga and phosphorus-doped silicon crystal wafer can be inexpensive and have high energy conversion efficiency.

That is, for example, G grown by the CZ method
By processing a and phosphorus-doped silicon single crystal rods into a solar cell substrate and forming a solar cell from the wafer, it has a stable conversion efficiency without being affected by oxygen taken into the crystal during single crystal growth A solar cell can be manufactured at low cost. In addition, when Ga and phosphorus-doped silicon single crystal are used as the material of the solar cell, the substrate lifetime can be stabilized without being affected by the concentration of oxygen, so that the conversion efficiency is high even if the resistivity of the solar cell is low. A solar cell can be manufactured.

With the conventional B-doped CZ single crystal, when the resistivity is lowered, the life time is reduced accordingly, and a solar cell with high conversion efficiency and low resistivity cannot be manufactured. However, when the Ga and phosphorus-doped silicon crystal and silicon crystal wafer of the present invention are used, a solar cell with low cost and high conversion efficiency can be manufactured.

Next, according to the present invention, in a method for producing a silicon crystal, Ga and phosphorus are added to a silicon melt in a crucible by the Czochralski method, and then a seed crystal is brought into contact with the silicon melt. A method for producing a silicon crystal, characterized in that a silicon single crystal is grown by pulling while rotating.

Thus, a silicon single crystal to which Ga and phosphorus are added can be manufactured. In particular, in the manufacturing method of the present invention, an unnecessary portion such as a cone portion of a widely used phosphorus-doped silicon crystal can be reused as a raw material, and the production cost is much higher than a method of doping only Ga. Can be cheaper. Further, by adding phosphorus as an N-type dopant, the fluctuation of the resistivity can be suppressed. Further, since the region within the specified resistivity or concentration range becomes longer due to the segregation of phosphorus as compared with the case where only Ga is doped, the yield is improved, and as a result, the production cost can be reduced.

Further, according to the present invention, in the method for producing a silicon crystal, after adding Ga and phosphorus to the silicon melt in the crucible by the Bridgman method, the polycrystalline silicon is grown by lowering the crucible from the heated region. (Claim 11). Thus, a silicon polycrystal to which Ga and phosphorus are added can be manufactured. In general, polycrystals can be made cheaper than single crystals, so that the cost for solar cells can be further reduced.

In this case, Ga is added to the silicon melt in the crucible by growing a silicon crystal to which high-concentration Ga has been added in advance and using a doping agent formed by crushing the high-concentration Ga-doped silicon crystal. And G in the silicon melt
It is preferable to add a (claim 12).

As a method of doping Ga when producing a single crystal to which Ga is added in the present invention, gallium may be directly added before melting polycrystalline silicon or into a molten silicon melt. If the single crystal to which is added is industrially mass-produced, it is better to dope after once adjusting the dopant as described above. By using such a method, work can be performed efficiently. This is because the melting point of gallium is as low as 30 ° C. and handling is difficult. Therefore, rather than putting gallium directly into the crucible,
By using the method of doping after preparing the dopant, the Ga concentration can be easily and accurately adjusted, and an accurate dopant concentration can be obtained. In addition, the handling of the dopant itself becomes easier as compared with the case where gallium is directly injected into the silicon melt, which also leads to an improvement in workability.

Further, phosphorus is added to the silicon melt in the crucible by growing a silicon crystal to which phosphorus has been added in advance and removing the cone, tail, or defective part of the silicon crystal rod to which phosphorus has been added. Phosphorus can be added to a silicon melt by using a crystal mass formed by crushing as a silicon raw material (claim 13).

As described above, even if the silicon crystal is doped with phosphorus, the material to be discarded is used, so that the material cost can be greatly reduced, and the solar cell cost can be further reduced. An increase in demand can be expected. In addition, when the solar cell substrate is used by controlling the phosphorus amount, the conductivity type does not reverse to the N type, the conversion efficiency does not decrease, and the conversion efficiency does not decrease due to light degradation.

Hereinafter, the present invention will be described in detail.
The present invention is not limited to these. The present inventors have conducted intensive research on how to obtain a substrate with relatively high conversion efficiency and low cost as a solar cell, which is relatively easy to manufacture and mass-produced as a substrate material for a solar cell. The present invention has been completed as a result of repeated experiments and studies.

That is, in a conventional solar cell manufactured from a P-type silicon crystal to which B is added as a dopant, the energy level of the PN junction surface of the solar cell is determined by the simultaneous presence of oxygen and B in the crystal. Changes, and a deep energy level (deep level or trap)
Also called level. ) Is formed, and the carriers in the solar cell are captured at this deep energy level, so that the lifetime of the substrate is reduced and light degradation occurs. However, the present inventors have found that, by adding Ga as a dopant, photodeterioration is prevented, and furthermore, even if phosphorus is present, the concentration of phosphorus at which the conductivity type does not reverse to the N type even when phosphorus is present, is specified. For example, the present invention has been completed by focusing on the point that the conductivity type is not reversed to the N type and the viewpoint of effective use of discarded members.

That is, the present applicant has proposed that a P-type silicon crystal is formed by adding Ga as a dopant instead of B. However, even if Ga-doped crystal contains N
Even if phosphorus as a type dopant is present, if the amount of phosphorus is limited, the conductivity type will not be inverted to N-type, the substrate lifetime will not be reduced, and a solar cell without light degradation can be manufactured. As a result of repeating the experiment, a substrate was prepared by pulling up a P-type silicon single crystal by using a silicon crystal to which Ga was added as a dopant and phosphorus to be discarded was used as part or all of the silicon raw material. And
It has been confirmed that the cost can be reduced with a solar cell using this, and that even if a relatively high concentration of oxygen is present in the crystal, the lifetime is stable and a solar cell without light degradation can be produced. .

Thus, if the cone, tail, or defective part of the silicon crystal rod to be discarded is used as part or all of the silicon raw material, for example, a high oxygen concentration can be obtained in a silicon single crystal manufactured by the CZ method. Even if it is shown, a solar cell having a stable and high conversion efficiency without causing photodegradation can be produced at an extremely low cost, and the power generation cost of the silicon single crystal solar cell can be reduced. As a result, the contribution to solving the cost problem of silicon raw materials for solar cells has become significant.

Further, by adding phosphorus as an N-type dopant, the fluctuation of the resistivity is suppressed. Further, since the region within the specified resistivity or concentration range becomes longer due to the segregation of phosphorus as compared with the case of only Ga doping, the yield is improved, and the production cost is also reduced.

In addition, since a stable conversion efficiency can be obtained without being affected by the oxygen concentration even in the case of a single crystal formed by, for example, the CZ method, a wafer having a diameter larger than that of the present one is used. can do. Conventionally, a single crystal doped with B and having a large crystal diameter has a photodegradation of the solar cell substrate due to a high oxygen concentration contained in the crystal. However, B is not added, and the amount of phosphorus is regulated by defining the amount of phosphorus. The use of a single crystal whose resistance is controlled by adding Ga to prevent mold formation can provide a high conversion efficiency without being affected by the oxygen concentration even with a single crystal substrate having a large diameter. It has also become possible to develop battery cells. Moreover, since oxygen is appropriately contained, the crystal strength is increased, the workability is improved, and the durability of the solar cell is improved.

In general, a solar cell needs to extract a predetermined voltage or current over a long period of time. Therefore, when actually used, a plurality of solar cell elements (cells) are connected in series or in parallel. It is connected and modularized so that the desired power can be extracted. In particular, in order to extract a large amount of electric power, it is necessary to connect a large number of solar cell elements. Therefore, the area of the solar cell is large in order to simplify and downsize the solar cell module and further reduce the manufacturing cost. It is more advantageous. If a large-sized and high-conversion-efficiency solar cell can be used as a material for a solar cell module, it is possible to further reduce solar cell cost and increase demand.

However, conventionally, oxygen contained in the crystal and B
Due to the effect of the above, even if a silicon single crystal having a large diameter is used as the solar cell substrate, the solar cell using the CZ method silicon single crystal has a low conversion efficiency of the solar cell energy and a decrease in the conversion efficiency due to light degradation. Therefore, it has been difficult to obtain a solar cell having characteristics commensurate with the cost for producing a large-diameter silicon single crystal.

On the other hand, when a silicon crystal doped with Ga and phosphorus is used as a solar cell substrate, a solar cell with high conversion efficiency can be obtained without a decrease in conversion efficiency due to photodegradation, and at the same time, the crystal diameter can be reduced. A large silicon single crystal can be manufactured by the CZ method. As a result, even a large-sized solar cell having an area of 100 cm ² or more can be mass-produced at a low cost. And demand can be expected to increase.

[0052]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail, but the present invention is not limited thereto. First, as a silicon crystal of the present invention, Japanese Patent Application No. 11-264549 describes a configuration example of a single crystal pulling apparatus by the CZ method and a pulling method, and a configuration example and a manufacturing method of a polycrystal manufacturing apparatus by the Bridgman method. The contents are basically the same as those described in Japanese Patent Application No. 2000-061435, and may be according to a general method. However, since the silicon raw material and the dopant are different, a description thereof will be added below.

Generally, the resistivity of a silicon crystal doped with phosphorus is in the range of 0.006 to 10000 Ω · cm. Within this range, a resistivity of 0.1 Ω · cm or less is not used in the present invention since it becomes the same as the dopant adjuster. First, for example, part or all of the unusable tails, cones, and defective parts that do not conform to standards such as resistivity among phosphorus-doped single crystal rods manufactured for semiconductor integrated circuits A raw material (hereinafter, referred to as a raw material in the recycle unit) is used, and the resistivity and weight of the raw material in the recycle unit are measured. ASTM is determined from the resistivity of the recycle part raw material.
(American Society for Tes
ting and Materials) Stand
ard F723-82 can be converted to phosphorus concentration,
The weight of phosphorus in the raw material for the recycling section can be determined by the following equation. Phosphorus weight = Phosphorus concentration x Phosphorus atomic weight x Atomic mass unit x Recycling unit raw material weight / Silicon density ... Formula (1)

Next, when the raw material for the reuse portion and a new silicon material, which is undoped as required, are melted together and the silicon crystal rod is pulled up, the phosphorus concentration at the shoulder of the silicon crystal rod becomes lower. It is determined according to the formula. Phosphorus concentration = Phosphorus weight x Phosphorus segregation coefficient / (Atomic weight of phosphorus x Atomic mass unit x Weight of (recycling part raw material + new silicon raw material) / Silicon density) ... Formula (2)

As a dopant which does not cause photodegradation, G
a is added to control the resistivity.
The a concentration is also expressed in the same manner as in the equation (2).
Phosphorus in the inside will be replaced by Ga.

Here, the atomic weight of phosphorus is 30.97,
Has an atomic weight of 69.72 and an atomic mass unit of 1.66 × 10
^-24 g, the density of silicon is 2.33 g / cm ³ , the segregation coefficient of phosphorus is 0.35, and the segregation coefficient of Ga is 0.008.

For this reason, for example, Ga is weighed to be 0.167 g, and the resistivity as a silicon raw material is 10 Ω.
・ Recycling section of 10 cm Assume that 10 kg of a crystal mass obtained by finely crushing a raw material and 20 kg of a new non-doped silicon raw material are charged together in a quartz crucible. The temperature of the heater is raised to melt the silicon raw material. When all the raw materials have been melted, the seed crystal is brought into contact with the surface of the melt, and is pulled up while rotating to produce a P-type silicon single crystal rod. Will be.
Then, G at the shoulder of the manufactured silicon single crystal rod
The concentration of a is 8.97 × 10 ¹⁴ atoms / cm ³ , the resistivity is 15 Ω · cm, and the concentration of phosphorus is 5.19 × 1.
0 ¹³ atoms / cm ³ .

Based on the above formula, the phosphorus weight can be calculated from the resistivity and the weight of the recycle part raw material. Of about 0 to 80%) with a new high-purity silicon raw material to obtain a predetermined phosphorus concentration and a predetermined G
The a concentration is determined, and the resistivity of the silicon crystal rod can be controlled. This makes it possible to reduce the cost of the silicon raw material. Here, it goes without saying that the use of 100% of the recycled material can greatly contribute to the cost, but the use of at least 20% greatly contributes to the cost.

The oxygen concentration in the silicon crystal is adjusted by appropriately adjusting the rotation speed of the crucible, the pulling speed of the silicon crystal, the pressure and the flow rate of the inert gas in the chamber. It can be controlled by adjusting the temperature of the melt and the pulling speed of the silicon crystal.

In processing a silicon crystal rod, the cone and tail are cut, and then the periphery of the silicon crystal rod is cylindrically ground and cut into blocks of an appropriate size. It is also possible to reuse the cone portion and the tail portion to which Ga and phosphorus are added again as a silicon raw material. However, for the tail part, heavy metal impurities may become high concentration due to segregation when reused again and again.
It is preferable not to reuse more than a certain amount.

In the silicon crystal of the present invention to which Ga and phosphorus are added, the concentration of Ga contained in the silicon crystal is 2 × 1.
0 ^{17 to} 3 × 10 ¹⁴ atoms / cm ³ , the concentration of phosphorus is 1/5 or less of the concentration of Ga, and the phosphorus concentration is
Preferably, the amount does not cause N-type inversion.

The resistivity of the silicon crystal of the present invention is as follows:
It is controlled by adding Ga. The concentration of phosphorus is Ga
If the concentration exceeds 1/5, the effect of phosphorus on the resistivity and the function as a recombination center for minority carriers may have a large effect on the lifetime. Therefore, the concentration of phosphorus is set to be 1% of the concentration of Ga in accordance with the concentration of Ga.
It is preferable to control to / 5 or less.

The silicon crystal wafer for a solar cell manufactured by using the present invention is obtained by slicing a single crystal block having an appropriate size by a slicer to form a wafer, and further reducing the processing distortion by etching. Manufactured by removing. Further, a solar cell manufactured by using the present invention is, for example, a high conversion efficiency cell RP-PERC using the silicon crystal wafer.
(Random Pyramid-Passiva
ted Emitter and RearCell)
It is preferably manufactured as a photovoltaic cell.

Hereinafter, the present invention will be described with reference to specific experimental examples. (Experimental Examples 1 to 9) To pull up a silicon crystal rod,
As a silicon raw material, a recycled raw material of 10 Ω · cm and a new non-doped silicon raw material were prepared, and as a dopant, a silicon crystal to which high-concentration Ga was previously added was grown and crushed. Then, conditions were adjusted so that the resistivity at the shoulder portion of the silicon crystal rod was 1 Ω · cm, and a silicon single crystal rod having a crystal diameter of 6 inches and a crystal orientation of <100> to which Ga and phosphorus had been added was added with a diameter of 18 inches. Using a quartz crucible, five wires were pulled up by the ordinary CZ method, and four wires were pulled up by the MCZ method to reduce the oxygen concentration.

The nine silicon crystals thus pulled were sliced from a block into a wafer having a thickness of 2 to 3 mm, and the lifetime was measured. For the measurement of the lifetime, this sliced wafer was subjected to HF: HNO ₃ = 5.
%: Treated with a mixed acid of 95% to remove the slice damaged layers on both sides by etching, and then wash the wafer surface, and then irradiate the wafer surface with steady light for 30 hours under the condition of AM (Air Mass) 1.5. Then, the natural oxide film on the surface is removed with HF, and then a chemical passivation (CP) treatment using a mixed solution of iodine and ethanol is performed.
Reduces carrier recombination on the crystal surface,
The lifetime of a silicon single crystal wafer was measured using the D method (photoconductivity decay method). The results are shown in Table 1 and FIG.

[0066]

[Table 1]

As can be seen from Table 1 and FIG. 2, in Experimental Examples 1 to 5 in which the oxygen concentration in the silicon single crystal was 16 ppma or less and the phosphorus concentration was 1/5 or less of the Ga concentration, phosphorus was doped. Nevertheless, it can be seen that there is almost no decrease in the life time, indicating that the characteristics are stable. In addition, it was confirmed that the fluctuation of the resistivity was suppressed.

On the other hand, as shown in Table 1 and FIG. 2, in the wafers of Experimental Examples 6 to 9 in which the phosphorus concentration exceeds 1/5 of the Ga concentration, the lifetime was reduced, and particularly, the oxygen concentration was reduced to 11%.
It can be seen that even in Experimental Examples 8 and 9 having ppma or less, the lifetime was reduced. However, the fluctuation of the resistivity was suppressed.

Next, a silicon single crystal wafer having a uniform resistivity was selected from the silicon single crystal wafers used above, and a large solar cell having a size of 10 cm × 10 cm (cell area of 10 cm) was selected.
A solar cell of 0 cm ² ) was manufactured, and its conversion efficiency was measured. To measure the conversion efficiency of a solar cell, place the solar cell on a measuring table temperature-controlled to 25 ° C., and use a solar simulator with a halogen lamp as a light source to emit a steady light under the condition of AM (air mass) 1.5. , And the voltage and current that could be taken out of the cell were measured to calculate the conversion efficiency of the solar cell. The conversion efficiency according to the present invention refers to a value defined by the following equation, and is as follows. [Conversion efficiency] = [electric power extracted per unit area of cell] / [light energy irradiated per unit area of cell] × 100 (%) The measurement results are also shown in Table 1.

As shown in Table 1, the oxygen concentration in the silicon single crystal was 16 ppma or less and the phosphorus concentration was Ga.
At a place of 1/5 or less of the concentration (Experimental Examples 1 to 5), the conversion efficiency shows a high value of 20.2 to 21.1%, indicating that light energy is efficiently converted to electric energy. Understand. In addition, the conversion efficiency after irradiating the solar cell with light for 30 hours or more shows almost the same value as the initial value, showing a stable conversion efficiency, and shows that the silicon crystal wafer to which phosphorus is added is not changed. Even with the solar cell used, by controlling the resistivity with Ga and controlling the oxygen concentration and the phosphorus concentration, a stable large-sized solar cell with high conversion efficiency and high performance without light degradation was obtained. It was confirmed.

On the other hand, if the phosphorus concentration exceeds 1/5 of the Ga concentration (Experimental Examples 6 to 9), or if the oxygen concentration is 16 ppm
a and above (Experimental Example 6), the conversion efficiency of the solar cell shows a slightly low conversion efficiency of 18.9 to 19.8% before light degradation, and the conversion efficiency after 30 hours of steady light irradiation. Was not lowered, but a conversion efficiency exceeding 20% could not be stably obtained.

The present invention is not limited to the above embodiment. The above embodiment is an exemplification, and has substantially the same configuration as the technical idea described in the scope of the claims of the present invention. It is included in the technical scope of the invention.

For example, in the above description, a case has been described in which a silicon single crystal to which Ga and phosphorus are added is mainly produced by the CZ method. The present invention is also applicable to polycrystalline silicon. That is, even in the Bridgman method, in order to reduce the cost of the silicon raw material,
Needless to say, it is effective to add Ga and phosphorus shown in the present invention.

[0074]

According to the present invention, Ga and phosphorus are added to a silicon crystal and a silicon crystal wafer. However, a silicon crystal and a silicon crystal wafer for manufacturing a solar cell having a low manufacturing cost can be obtained. By controlling the concentration and the phosphorus concentration, a solar cell with high light energy conversion efficiency can be manufactured without inversion of the conductivity type and light deterioration. Furthermore, it is possible to obtain a material having a large diameter, a high crystal strength, and excellent durability, while contributing to a large diameter and low cost.

[Brief description of the drawings]

FIG. 1A shows the resistivity in the length direction of a silicon single crystal when Ga or B is used as a dopant, and FIG.
FIG. 3 is a diagram showing respective dopant concentrations with respect to the length direction of the silicon single crystal.

FIG. 2 is a graph showing the relationship between lifetime and initial oxygen concentration and phosphorus concentration.

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 4G072 AA02 BB11 BB12 GG01 HH01 JJ09 NN21 NN25 UU02 4G077 AA02 BA04 CF10 EB01 EJ02 HA20 5F053 AA11 AA13 DD01 FF04 GG01 GG02 KK10 LL05 RR04 RR07

Claims (13)

[Claims] 1. A silicon crystal, wherein gallium (Ga) and phosphorus are added as dopants. 2. The silicon crystal according to claim 1, wherein the concentration of phosphorus contained in said silicon crystal is 1/5 or less of the concentration of Ga. 3. The silicon crystal according to claim 1, wherein the concentration of Ga contained in the silicon crystal is 2 × 10 ^{17 to} 3 × 10 ¹⁴ atoms / cm ³ . 4. The silicon crystal has a resistivity of 20 Ω · c.
The silicon crystal according to any one of claims 1 to 3, wherein m is from 0.1 to 0.1 Ω · cm. 5. The silicon crystal according to claim 1, wherein an interstitial oxygen concentration in the silicon crystal is 16 ppma or less. 6. The silicon crystal according to claim 1, wherein the silicon crystal is a silicon single crystal manufactured by a Czochralski method. . 7. The silicon crystal according to claim 1, wherein the silicon crystal is a polycrystalline silicon manufactured by a Bridgman method. 8. A silicon crystal wafer obtained by slicing the silicon crystal according to any one of claims 1 to 7. 9. The silicon crystal wafer according to claim 8, wherein the silicon crystal wafer is for a solar cell. 10. In a method for producing a silicon crystal, Ga and phosphorus are added to a silicon melt in a crucible by the Czochralski method, and then a seed crystal is brought into contact with the silicon melt and pulled up while rotating. A method for producing a silicon crystal, which comprises growing a silicon single crystal. 11. A method for producing a silicon crystal, wherein Ga and phosphorus are added to a silicon melt in a crucible by the Bridgman method, and then the crucible is pulled down from a heated region to grow a silicon polycrystal. Manufacturing method of a silicon crystal. 12. The method according to claim 1, wherein Ga is added to the silicon melt in the crucible.
Adding the Ga to the silicon melt by growing a silicon crystal to which a high concentration of Ga has been added in advance and using a dopant made by crushing the high concentration of the Ga-doped silicon crystal. The method for producing a silicon crystal according to claim 10 or 11. 13. Addition of phosphorus to the silicon melt in the crucible includes growing a silicon crystal rod to which phosphorus has been added in advance, and forming a cone, tail, or defective part of the silicon crystal rod to which phosphorus has been added. 13. The method for producing a silicon crystal according to claim 10, wherein phosphorus is added to the silicon melt by using a crystal lump produced by crushing as a silicon raw material.