JPH06191820A - Production of silicon thin plate - Google Patents

Abstract

PURPOSE:To provide a production method of a silicon thin plate with good productivity and lower production cost. CONSTITUTION:A crystalline silicon thin plate 80 is produced by cooling the interface of a silicon liquid in a crucible 40 and a gas phase to precipitate silicon in the interface. The silicon liquid is obtd. by melting a silicon plate 80 in a flux selected from indium, tin and gallium. Compared to the conventional method using a silicon liquid obtd. by melting silicon, it is not necessary for this method to maintain the silicon higher than its melting point so that the process temp. during the production of silicon thin plate can be decreased. Thereby, inexpensive constitutional parts can be used and the energy used for the production can be decreased.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for producing a crystalline silicon thin plate applicable to a substrate for solar cells, etc.
The present invention relates to a method for manufacturing a silicon thin plate which has good productivity and can reduce manufacturing cost.

[0002]

2. Description of the Related Art In recent years, from the viewpoint of protecting the global environment, a solar cell has been attracting attention as a power generation means replacing thermal power generation, nuclear power generation and the like. In other words, this solar cell is expected to be a clean energy source because it generates power when exposed to sunlight and does not emit carbon dioxide or radioactive waste at all.

However, the cost of power generation by a solar cell is still high, and it is necessary to significantly reduce the manufacturing cost in order for the solar cell to be widely popularized.

By the way, as a typical example of this solar cell,
An example is a crystalline silicon solar cell manufactured by processing a single crystal silicon substrate or a polycrystalline silicon substrate. And, in the former, from a single crystal silicon ingot obtained by pulling from the silicon melt by the Czochralski method, etc., a single crystal silicon wafer is cut out with a diamond saw as a solar cell substrate, and,
In the latter case, a polycrystalline silicon ingot is produced by a casting method or the like, and a cast silicon wafer is cut out with a diamond saw to obtain a solar cell substrate.

However, in these methods, since the growth rate of the crystalline silicon ingot is slow, the productivity thereof is poor, and a cutting step with a diamond saw is required, which increases the manufacturing cost accordingly. There was a drawback that

As a solution to the above-mentioned problems associated with the production of crystalline silicon solar cells, thin-film solar cells typified by amorphous silicon have been eagerly studied. However, in terms of cost and performance, the crystalline silicon-based solar cell has not been surpassed as a material for a solar cell for electric power.

Therefore, some methods of manufacturing a polycrystalline silicon thin plate have been proposed which do not require the above cutting step with a diamond saw.

As an example, an EFG (Edge-defined Fil) is used to pull up a ribbon-shaped silicon thin plate from a silicon melt.
m-fed Growth) method [eg, “EFG cry
stalgrowth technology for low cost terrestrial pho
tovoltaics: review andoutlook ”, FVWald et al., Sol
ar Energy Materials and Solar Cells, vol.23, p.175 (1
992), see]. In this method, as shown in FIG. 7, a crystalline silicon thin plate c is pulled up from a bath b of molten silicon a in a direction perpendicular to the silicon melt surface to manufacture. In FIG. 7, d is a ribbon and e is a die.

However, in this manufacturing method, the crystal growth of silicon occurs only at the interface between the silicon melt and the silicon thin plate, and therefore, in order to obtain crystalline silicon of a quality usable as a solar cell, the growth rate is at most several. It has to be limited to about centimeters / minute, and therefore has the drawbacks of poor productivity, low cost of solar cells, and difficulty in mass production.

As another example, a method in which a tape made of a heat resistant material such as carbon fiber cloth or graphite sheet is passed through a molten silicon bath to deposit crystalline silicon on the tape. Have also been developed [eg, “Growth of Poly sheets on
a carbon shaper by the RAD process ”, C.Belouetet a
l., Journal of Crystal Growth, vol.61, p.615 (1983), see]. In this method, a heat-resistant tape is passed through the surface of molten silicon or the surface of molten silicon, and the silicon melt adhering to the tape is solidified to obtain polycrystalline silicon. In this method, the crystallization of silicon does not necessarily have to occur at the silicon melt interface and can be performed in the heat treatment process after attaching to the tape. Has the advantage that the growth rate can be increased.

However, this method also has the following drawbacks.

First, in this method, silicon crystallizes on a heat resistant tape such as carbon. At this time, during cooling from the growth temperature (1410 ° C. or higher which is the melting point of silicon) to room temperature, cracks or various crystal defects occur in the silicon due to the difference in thermal expansion coefficient between the silicon and the heat resistant tape. Easy to do. Then, these cracks and crystal defects in the silicon cause the characteristics to be greatly deteriorated when applied to the solar cell.

The heat resistant tape such as carbon must be of high purity. This is because if the heat-resistant tape contains impurities, these impurities will be diffused into the silicon during the growth and crystallization process of silicon and will cause a significant deterioration in its characteristics when applied to solar cells. .

As described above, the EFG method of pulling up the ribbon-shaped silicon thin plate has a difficulty in productivity, and the manufacturing method of growing silicon on the heat-resistant tape deteriorates the characteristics of the solar cell. was there.

Under such a technical background, a new manufacturing method has been proposed in which a supporting base material such as the above heat-resistant tape is not required and a crystalline silicon thin plate is required at high speed. For example, in Japanese Unexamined Patent Publication No. 57-205395, a cooling plate is arranged near the silicon melt in the crucible to cool the interface between the silicon melt and the vapor phase, and to cool the solid phase. A method for depositing silicon is disclosed, and Japanese Patent Application Laid-Open No. 55-140791 discloses a method for spraying a cooling gas onto the surface of the silicon melt to deposit solid phase silicon.

These methods of cooling the interface between the silicon melt and the vapor phase to precipitate solid phase silicon require a silicon thin plate in a one-step process, have a high raw material utilization rate, and have the above-mentioned cutting. It was a method having excellent advantages as compared with the other manufacturing methods described above, such as no steps required.

[0017]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention By the way, it has been disclosed in the above publication that a cooling plate is arranged in the vicinity of the above-mentioned silicon melt or a cooling gas is sprayed to cool the interface between the silicon melt and the vapor phase. In the conventional method, a silicon melt obtained by melting silicon has been applied.

For this reason, when manufacturing a silicon thin plate, it is necessary to maintain the system above the melting point of silicon (about 1410 ° C.), so that expensive components having high heat resistance are required for components such as an electric furnace. Moreover, there was a problem that the amount of energy input was large and the energy intensity was high.

The present invention has been made by paying attention to such a problem, and an object thereof is to provide a method for manufacturing a silicon thin plate which has good productivity and can reduce the manufacturing cost. .

[0020]

[Means for Solving the Problems] That is, according to the first aspect of the present invention, an interface between a silicon melt and a vapor phase in a crucible is cooled, and silicon is deposited at the interface to produce a thin plate of crystalline silicon. Based on the method of manufacturing a silicon thin plate to
It is characterized in that the silicon melt is composed of a melt of silicon dissolved in a flux selected from indium, tin, and gallium.

According to such technical means, silicon is dissolved in a flux selected from indium (melting point: 156 ° C.), tin (melting point: 232 ° C.), and gallium (melting point: 30 ° C.). Since the obtained silicon melt is applied, it is not necessary to keep the system above the melting point of silicon as in the conventional method, so the temperature can be reduced during the production of silicon thin plates, and therefore inexpensive components can be applied. As a result, the energy input can be reduced.

In the invention according to claim 1, when the silicon melted in the flux is consumed, it becomes difficult to manufacture a silicon thin plate thereafter. Therefore, when a large number of thin plates are continuously manufactured. The raw material management may be a little complicated. The invention according to claim 2 relates to a method for improving this point.

That is, the invention according to claim 2 is claim 1
Based on the method for manufacturing a silicon thin plate according to the above, solid silicon is put into the crucible, and the silicon concentration in the silicon melt is maintained in a supersaturated state during the manufacturing of the silicon thin plate.

In the second aspect of the present invention, as the solid silicon to be introduced, for example, a p-type or n-type single crystal silicon plate having a p-type or n-type dopant mixed therein (a By applying a crystalline silicon plate, it is possible to adjust the manufactured silicon thin plate to a p-type or n-type silicon thin plate. In particular, when indium or gallium is applied as the flux, the silicon thin plate produced becomes a weak p-type even if a high-purity silicon plate is added,
By inserting the p-type or n-type single crystal silicon plate (polycrystalline silicon plate), there is an advantage that the silicon thin plate can be adjusted to any type.

When the solid silicon is charged into the crucible, the charged solid silicon may float from the melt due to the specific gravity of the applied flux.
In such a case, it is necessary to provide a fixing means for fixing the solid silicon in the crucible.

Next, when cooling the interface between the silicon melt and the vapor phase, if the surface of the silicon melt is cooled uniformly, a large amount of fine crystal grains are generated instantaneously over the entire surface of the silicon melt. Therefore, it may be difficult to obtain a silicon thin plate having a large crystal grain size because crystal growth of silicon along the surface direction of the silicon melt surface is hindered. The invention according to claim 3 relates to the invention made from such a technical point of view.

That is, the invention according to claim 3 is claim 1
Alternatively, on the premise of the method for manufacturing a silicon thin plate according to the second aspect, a cooler having a temperature gradient along the surface direction of the interface is arranged near the silicon melt to cool the interface between the silicon melt and the vapor phase. It is characterized by doing.

According to the third aspect of the present invention, the interface between the silicon melt and the vapor phase is formed by using a cooler having a temperature gradient along the surface direction of the interface between the silicon melt and the vapor phase. Since it is cooled, it is difficult to generate fine crystal grains in the high cooling temperature portion of the silicon melt, and fine crystal grains are first generated only in the low cooling temperature portion.
Since these crystal grains will grow in sequence from the low cooling temperature region to the high cooling temperature region along the surface direction of the silicon melt surface, a thin plate such as polycrystalline silicon with a large average grain size and good crystallinity can be secured. It has the advantage of being manufacturable.

A specific example of the cooler having a temperature gradient is a hollow body whose bottom shape is substantially the same as the desired shape of the silicon thin plate. The hollow body circulates the refrigerant and the hollow body. One having a gradient on one surface of the body facing the silicon melt can be exemplified.

[0030]

According to the invention of claim 1, a silicon melt obtained by dissolving silicon in a flux selected from indium, tin, and gallium is applied. Since it is not necessary to keep the temperature above the melting point of silicon, it is possible to reduce the temperature at the time of manufacturing the silicon thin plate, and therefore, it is possible to apply inexpensive components and reduce the input energy.

According to the second aspect of the present invention, solid silicon is charged into the crucible and the silicon concentration in the silicon melt is maintained in a supersaturated state during the production of the silicon thin plate. Since it becomes possible to continuously manufacture a silicon thin plate until the charged solid silicon is consumed, the productivity can be improved.

Further, according to the third aspect of the present invention, a cooler having a temperature gradient along the surface direction of the interface between the silicon melt and the vapor phase is arranged near the silicon melt and the interface is provided. Is being cooled, fine crystal grains are less likely to be generated in a portion of the silicon melt having a high cooling temperature, and fine crystal grains are first generated only in a portion of a low cooling temperature. Since it grows in sequence from the low cooling temperature region to the high cooling temperature region along the surface direction of the liquid surface, it is possible to reliably manufacture thin plates such as polycrystalline silicon with a large average grain size and good crystallinity. Become.

[0033]

Embodiments of the present invention will now be described in detail with reference to the drawings.

[Embodiment 1] The manufacturing apparatus used in this embodiment includes an electric furnace 10 and an electric furnace 1 as shown in FIG.
Length 40c fixedly mounted on the support base 100 within 0
and a reaction tube 20 made of quartz and having a diameter of 10 cm
A crucible 40 made of quartz, which is arranged in the inner part of the crucible via a crucible base 30 made of graphite, and a fork-shaped tip portion is housed in the crucible 40 arranged in the reaction tube 20 and a base end side thereof serves as a lever. A main part is constituted by a functioning member 50 for taking out a thin plate made of quartz, and a cooler 60 made of glassy carbon, which is also arranged in the reaction tube 20 in the vicinity of the crucible 40 and is provided with an air supply pipe 61 and an exhaust pipe 62. And
The base end side of the air supply pipe 61 and the tip end side of the exhaust pipe 62 are slidably attached through an opening (not shown) of a lid portion 21 provided in the open portion of the reaction tube 20, and The base end side of the thin plate taking-out member 50 is also slidably attached to the opening (not shown) of the lid portion 21. By attaching these, the cooler 60 and the fork-shaped tip of the thin plate taking-out member 50 are attached. The part can be moved up and down. Further, an inlet 22 and an outlet 23 for exhausting the inside of the reaction tube 20 and introducing high-purity argon gas into the reaction tube 20 are attached to the upper side of the reaction tube 20. In addition, 90 in FIG. 1 shows a heat insulating material.

As shown in FIGS. 2 and 3, the cooler 60 is composed of a rectangular hollow body 63 having a bottom shape of 45 mm × 45 mm, and the air supply pipe 61 is provided in the hollow body 63.
The refrigerant (nitrogen gas set to normal temperature) is circulated through the exhaust pipe 62.

On the other hand, the quartz crucible 40 is shown in FIG.
As shown in (A) to (B), the bottom surface has a closed diameter of 80.
It is composed of a cylindrical body having a height of 72 mm and a height of 72 mm, and a fixing member 42 composed of two protrusions is attached to the lower side of the inner wall surface thereof.

The solid silicon plate 70 put into the crucible 40 has a disk shape as shown in FIGS. 5A to 5B, and is attached to the inner wall surface of the crucible 40. The strip-shaped notch 7 is provided at a portion corresponding to the fixing member 42.
1 is formed, the notch 71 is pushed into the crucible 40 aligned with the position of the fixing member 42, and
The disc-shaped silicon plate 70 is rotated in any direction to shift the positions of the notch 71 and the fixing member 42, thereby preventing the disc-shaped silicon plate 70 from floating.

After the disk-shaped silicon plate 70 (p-type single crystal silicon plate having a specific resistance of 1 Ω · cm) for supplying the raw material is placed in the crucible 40 according to the above-mentioned procedure,
Into the crucible 40, about 1500 g (gram) of high-purity tin (99.9999%) was charged. Then, the system inside the reaction tube 20 is evacuated through the exhaust port 23 to 10 ⁻⁶ Torr.
After that, high-purity argon from which oxygen and water were removed by passing through the purification column was introduced from the inlet 22 at 200 cc / min to bring the pressure in the system to normal pressure (1 atm).

Next, the temperature of the entire system is set to 950 in the electric furnace 10.
The temperature was raised to 0 ° C., and this temperature was maintained for 1 hour to dissolve silicon in the molten tin until saturation was reached.

Next, the cooler 60 is arranged at a position 5 mm from the liquid surface of the tin melt in which silicon is dissolved, and the refrigerant is introduced into the cooler 60 under the condition of 1000 cc / min to introduce the tin melt. The interface between the liquid and the gas phase was cooled, and silicon was deposited at the interface.

After 30 minutes, the cooler 60 and the thin plate removing member 50 are pulled up (silicon deposited by this operation is pulled up from the tin melt), the electric furnace 10 is turned off, and the inside of the reaction tube 20 is brought to room temperature. After cooling, the deposited silicon thin plate 80 was taken out to the outside using the above-mentioned thin plate taking-out member 50.

The thin plate 80 produced has an outer dimension of approximately 40 m.
The polycrystalline silicon plate was m × 40 mm and had a thickness of about 0.5 mm, and the specific resistance was p-type silicon having a resistivity of about 2 Ω · cm. The average particle size was about 300 μm.

In this embodiment using the tin flux, the temperature of the entire system is set to 950 ° C. in the electric furnace 10, but the set temperature does not need to be limited to this temperature and is 800 to
The temperature can be appropriately changed within the range of 1000 ° C.

However, if the temperature is set lower than this, the crystallinity of the silicon thin plate is lowered and the solubility of silicon is lowered to lower the productivity, which is not preferable.

[Embodiment 2] In this embodiment, the cooler 6 is used.
6 is substantially the same as that of the first embodiment except that the hollow body 63 has a structure in which the bottom surface of the hollow body 63 has a gradient of about 10 degrees with respect to the horizontal plane as shown in FIG.

When this cooler 60 is arranged in the vicinity of the tin melt, a temperature gradient is formed along the surface direction of the interface between the tin melt and the vapor phase, so that the surface of the tin melt is As a result, silicon was sequentially grown along the surface direction from a region having a low cooling temperature to a region having a high cooling temperature, and a thin plate of polycrystalline silicon having a larger crystal grain size (average grain size 600 μm) than that in Example 1 was obtained. .

[0047]

According to the first aspect of the present invention, it is not necessary to keep the system above the melting point of silicon as in the conventional method, so that the process temperature at the time of manufacturing a silicon thin plate can be reduced.

Therefore, inexpensive components can be applied and the energy input can be reduced, so that the manufacturing cost can be reduced.

According to the second aspect of the present invention, solid silicon is put into the crucible and the silicon concentration in the silicon melt is maintained in a supersaturated state during the production of the silicon thin plate. Since the silicon thin plate can be continuously manufactured until the charged solid silicon is consumed, the production efficiency can be dramatically improved.

Further, according to the third aspect of the present invention, a cooler having a temperature gradient along the surface direction of the interface between the silicon melt and the vapor phase is arranged near the silicon melt and the interface is provided. Is being cooled, fine crystal grains are less likely to be generated in a portion of the silicon melt having a high cooling temperature, and fine crystal grains are first generated only in a portion of a low cooling temperature. Since it grows in order from the part with low cooling temperature to the part with high cooling temperature along the surface direction of the liquid surface, it is possible to easily and reliably manufacture thin plates such as polycrystalline silicon with large average grain size and good crystallinity. Have

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a manufacturing apparatus used in an example.

FIG. 2 is a partially enlarged view of FIG.

3 (A) is a plan view of the cooler, and FIG. 3 (B) is a sectional view taken along the line BB of FIG. 3 (A).

4A is a cross-sectional view of the crucible, and FIG. 4B is a plan view of the crucible.

FIG. 5 (A) is a plan view of a silicon plate, and FIG. 5 (B).
Is this cross section.

FIG. 6 is a sectional view showing the configuration of a cooler according to another embodiment.

FIG. 7 is a perspective view according to a conventional example of manufacturing a silicon thin plate.

[Explanation of symbols]

 40 Crucible 60 Cooler 61 Air Supply Pipe 62 Exhaust Pipe 70 Silicon Plate 80 Thin Plate

Claims (3)

[Claims] 1. A method for producing a thin silicon plate, comprising: cooling an interface between a silicon melt in a crucible and a gas phase; and depositing silicon at the interface to produce a thin plate of crystalline silicon. A method for manufacturing a silicon thin plate, comprising a melt obtained by dissolving silicon in a flux selected from indium, tin, and gallium. 2. The silicon thin plate according to claim 1, wherein solid silicon is charged into the crucible and the silicon concentration in the silicon melt is maintained in a supersaturated state during the production of the silicon thin plate. Manufacturing method. 3. A cooler having a temperature gradient along the surface direction of the interface is arranged near the silicon melt to cool the interface between the silicon melt and the vapor phase.
Alternatively, the method for manufacturing a silicon thin plate according to the above item 2.