JP2005166994A - Manufacturing method of solar cell, and solar cell manufactured by the method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a high-quality solar cell at low costs by improving the diffusion length of the minority carrier of a silicon substrate by an effective method of calcination. <P>SOLUTION: The manufacturing method of the solar cell is for manufacturing the solar cell having a p-type silicon substrate and a rear-surface electric field layer formed by the calcination of aluminum on a surface opposite to the photodetective surface of the p-type silicon substrate. A temperature-lowering speed from the highest temperature in a heat treatment process of forming the rear-surface electric field layer is ≥55°C/s. The solar cell is manufactured by the method. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a solar cell using a semiconductor material.

  At present, solar cells using single crystal or polycrystalline silicon are widely used mainly for homes in terms of material costs and energy conversion efficiency. In order for solar cells to become more widespread in the future and contribute to the solution of environmental problems, it is required to reduce the manufacturing process and material costs and to increase the energy conversion efficiency.

A normal single crystal silicon or polycrystalline silicon solar cell is manufactured, for example, by a process as shown in FIG. 1 (see Patent Document 1). First, fine unevenness is formed chemically or mechanically on the p-type silicon substrate 1 after cleaning as shown in FIG. 1A in order to efficiently capture incident light over the entire light receiving surface side. . Next, as shown in FIG. 1B, an n-type semiconductor layer 2 is formed on the p-type silicon substrate 1 by thermal diffusion of POCl _{3 or} the like. Thereafter, as shown in FIG. 1C, after the light receiving surface is protected with a tape or the like, the n-type semiconductor layer 2 formed on a portion other than the light receiving surface, that is, the back surface and the side surface is removed by etching.

Subsequently, as shown in FIG. 1D, an antireflection film 3 such as a silicon nitride film is formed on the n-type semiconductor layer 2 on the light receiving surface side. Next, as shown in FIG.1 (e), in order to form an electrode, the aluminum paste 4 is printed over the wide range of a back surface by screen printing, and is dried. Here, in order to prevent the aluminum paste from adhering to the side surface and the diffusion layer from being formed on the side surface of the substrate, the end of the substrate is prevented from being attached with the aluminum paste. Thereafter, as shown in FIG. 1 (f), the substrate is baked in a belt furnace or the like to form the p ⁺ layer 5 and the back electrode 6 having a back surface field layer (BSF) effect. Finally, as shown in FIG. 1G, a solar cell can be manufactured by printing a silver paste on the light receiving surface side by screen printing, firing in a belt furnace, and forming the light receiving surface side electrode 7. .

  In order to further improve the efficiency of conventional solar cells, it is an important theme to improve the quality of silicon as a material. In general, the diffusion length or lifetime of minority carriers is adopted as a standard for judging the quality of the semiconductor material for solar cells. Usually, the smaller the impurity in the silicon substrate, the longer the diffusion length or lifetime. However, solar cells need to be manufactured at low cost, and the silicon wafers used are not necessarily as pure as silicon wafers used in the manufacture of semiconductor devices such as integrated circuits. This is because the use of high purity and high accuracy inevitably increases the cost. However, when a silicon substrate contaminated with impurities is used or a silicon substrate is contaminated with impurities in the process of manufacturing solar cells, there is a problem that the energy conversion efficiency of the completed solar cell is limited as a result. . Therefore, many attempts have been made to improve the diffusion length of the silicon substrate in the process of manufacturing a solar cell using solar cell grade silicon.

  One such attempt is gettering. The basic concept of this method is to incorporate impurities from the entire wafer into a portion of the wafer that does not affect the device characteristics. In general, impurities to be gettered are metal elements such as Fe, Cu, Ti, and Cr. Specific gettering methods include intrinsic gettering and extrinsic gettering. The former is a standard method used in the manufacture of ICs (integrated circuits), etc., but the purpose is mainly to move metal impurities from the vicinity of the surface of the silicon substrate to the center of the silicon substrate. It is not appropriate when it is necessary to improve the quality of silicon over the entire substrate as in a battery. Therefore, exogenous gettering can be applied to the manufacturing process of the solar cell.

  Specific methods of extrinsic gettering are described in many documents (see Non-Patent Document 1). The most widely used method is a method in which P or Al or the like is diffused at a high temperature on the surface of the silicon substrate and metal impurities are taken into the surface. The reason why it is often used in the development of solar cells is that, in the production process of solar cells, the diffusion of P or Al serves as the step of forming a bonding layer and the step of forming electrodes, respectively. Therefore, in the manufacture of solar cells, it is very important to optimize the method of gettering using P or Al.

  Many theoretical and experimental analyzes have been performed on the mechanism of impurity removal by gettering (see Non-Patent Document 2 and Non-Patent Document 3). However, the assertions in these documents are all different and are often not clarified in practice. Among them, the theory that has been considered so far is that the dopant diffused on the surface of the silicon substrate at high temperature and the silicon substrate have different solid solubility of the metal impurity to be removed. There are two theories that the dopant is diffused into the surface and the dopant diffused on the surface induces a defect on the surface of the silicon substrate, and the metal impurity is taken in there. Probably, the mechanism of gettering cannot be clarified only by these two theories, but considering the combined effect of these two theories, the experimental results can be explained to some extent qualitatively.

Such substrate reforming technology has already been several decades old, and a recent novel method is the rapid thermal processing method. RTP is a method in which the heating and cooling rates are significantly increased compared to the belt furnace normally used in the heating process in solar cells, and such a heat treatment method has an effect different from that of normal heat treatment. Not only solar cells but also other semiconductor device fields have been reported for devices. In general, there is a literature claiming that rapid cooling from a high temperature state in RTP causes defects in a crystal (see Non-Patent Document 4). In the field of solar cells, a silicon nitride film is formed on the light-receiving surface side of a p-type silicon substrate. After depositing and printing the aluminum paste on the back surface, the temperature was raised from room temperature to 750 ° C. at about 100 ° C./s using an RTP furnace, maintained at that temperature for 1 second, and then dropped to 40 ° C./s or more. There is a report that the lifetime of the silicon substrate is improved by cooling at a speed, and the characteristics are improved when a solar cell is manufactured by introducing this process (see Non-Patent Document 5). The reason why the solar cell characteristics are improved in this method is that the authors of this document indicate that firing with RTP increases the amount of vacancies generated at the interface between aluminum and silicon, and as a result, the silicon substrate becomes a silicon nitride film. It is said that it will be able to take in more hydrogen contained in it and increase the deactivation effect of defects caused by hydrogen.
JP 2002-176186 A Appl. Phys. A, 1997, 64, p127-137 J. Appl. Phys., 2001, 90, p5388 J. Appl. Phys., 1989, 65 (8), p2974 IEEE TRANSACTIONS ON ELECTRON DEVICES, 1992, January, Vol. 39, NO. 1, p.96-104 Appl. Phys. Lett., 2003, 82, p.224

  However, this method is based on the premise that a silicon nitride film is deposited as an antireflection film, which limits the design of the solar cell and is not a versatile method. In other words, this method cannot be used for solar cells that employ titanium dioxide, magnesium fluoride, or the like as an antireflection film.

  The method for manufacturing a solar cell of the present invention is a method for manufacturing a solar cell having a p-type silicon substrate and a back surface electric field layer formed by firing aluminum on the surface opposite to the light receiving surface in the p-type silicon substrate, The temperature decreasing rate from the maximum temperature in the heat treatment step for forming the back surface electric field layer is 55 ° C./s or more. In addition, the solar cell of the present invention is manufactured by such a method.

  By the effective firing method according to the present invention, the diffusion length of minority carriers in a silicon substrate can be greatly improved, and a high-quality solar cell can be provided at low cost.

The method for producing a solar cell of the present invention is a method for producing a solar cell having a p-type silicon substrate and a back surface electric field layer formed by firing Al, and is from the highest temperature in the heat treatment step for forming the back surface electric field layer. The temperature decreasing rate is 55 ° C./s or more. In the heat treatment process of diffusing Al in the silicon substrate and forming the back surface electrolytic layer, by rapidly cooling at 55 ° C./s or more from the maximum temperature, defects higher in concentration than usual are formed near the surface of the silicon substrate, Metal impurities inside the silicon substrate can be effectively removed. As a result, in the gettering by Al, the diffusion length of minority carriers in the silicon substrate can be increased compared to the conventional method, and a high-quality solar cell can be provided. The vicinity of the front surface of the back surface of the substrate where defects are formed is a p ⁺ layer in which Al is diffused at a high concentration, and the influence of the occurrence of defects is small.

  Next, the effects of the present invention will be demonstrated by examples, but the present invention is not limited to these examples.

Example 1
In this example, a p-type polycrystalline silicon substrate (resistivity 0.5 Ωcm, thickness 300 μm) was used. Since the present invention provides an effective gettering method by Al diffusion, the diffusion length was measured for the purpose of grasping the quality of the substrate before processing.

First, for cleaning, the substrate was boiled with NH ₄ OH / H ₂ O ₂ / H ₂ O solution, the oxide film was removed with HF / H ₂ O solution, and then HCl / H ₂ O ₂ / Boiled with H ₂ O solution. Next, after sufficiently rinsing with pure water, etching was performed with a HNO ₃ / HF mixed solution in order to remove the surface damage layer. Thereafter, the diffusion length was measured by a surface photovoltage method using SPV-85XL manufactured by SEMILAB. The measured results are shown in Table 1 as the initial diffusion length.

In this way, after obtaining knowledge about the quality of the substrate in advance, the present invention was carried out as follows. First, the substrate was subjected to anisotropic etching with an aqueous NaOH solution for the purpose of removing the surface damage layer during slicing and texturing, and then spin-dried. Next, for the purpose of forming an n-type semiconductor layer on the surface of the substrate, P was diffused at 850 ° C. for 10 minutes in an atmosphere of N ₂ and O ₂ by vapor phase diffusion of POCl ₃ . Subsequently, in order to remove the diffusion layer formed on the back surface and the side surface of the substrate other than the light receiving surface of the substrate, the portion used as the light receiving surface is protected by applying an acid-resistant tape, and then HNO ₃ Etching was performed with a / HF mixed aqueous solution. Thereafter, for the purpose of forming an antireflection film, a silicon nitride film having a thickness of 700 mm was deposited on the n-type semiconductor layer on the light receiving surface side by plasma CVD. In order to form an electrode layer on the back side of the substrate, Al paste was screen-printed on almost the entire surface leaving 1 mm of the edge of the substrate and sufficiently dried at 150 ° C.

Subsequently, an Al and Al alloy layer was formed by firing an Al paste, and a p ⁺ layer and a back electrode having a back surface field layer (BSF) effect were formed. FIG. 2 schematically shows a cross section of a furnace for rapid heat treatment used for heating the substrate in this example. The reason why the apparatus shown in FIG. 2 is used is to accurately control the temperature rising rate and the temperature decreasing rate in the heat treatment process, but if the temperature can be sufficiently controlled, the apparatus is not limited to the furnace shown in FIG. The present invention can be implemented. First, a silicon substrate 21 to be fired was inserted into a chamber 22 made of quartz glass from a furnace door 23, and the substrate 21 was heated by tungsten / halogen lamps 24 installed at the top and bottom. The temperature of the substrate was observed by a thermocouple (not shown) on the dummy silicon substrate 25 located in the immediate vicinity of the substrate 21 in a low temperature region near room temperature. Further, in a high temperature region where Si and Al cause a reaction, observation was performed by a high-speed infrared radiation thermometer (pyrometer) 26 installed below the substrate 21.

The pyrometer 26 was temperature calibrated in advance using a thermocouple in order to accurately monitor the temperature. Moreover, the PID control method was employ | adopted for control, such as a temperature increase rate. In the heat treatment, the temperature is increased to 750 ° C. at a temperature increase rate of 10 ° C./s, and after reaching the maximum temperature of 750 ° C., this temperature is maintained for 2 minutes, and then the temperature is decreased to 550 ° C. at a temperature decrease rate of 85 ° C./s. Lowered. At this time, the temperature lowering rate was controlled by cooling water (not shown) flowing around the chamber. During the heat treatment, N ₂ and O ₂ were flowed into the furnace from the rear part 27 of the chamber. As for each flow rate, N ₂ was ₂ SLM (L / min in a standard state), and O ₂ was ₂ SLM. Next, a silver paste was printed by screen printing on the light receiving surface side, and baked in a belt furnace to form an electrode on the light receiving surface side. Finally, the short-circuit current, the open-circuit voltage, the curvature factor, and the conversion efficiency of the manufactured solar cell were measured with a solar simulator (conditions: AM1.5, 25 ° C.). The results are shown in Table 2.

In the manufactured solar cell, in order to investigate how the gettering by firing has an influence on the characteristics of the substrate, the light receiving surface side electrode, the n ⁺ diffusion layer, the back surface electrode and the back surface electrolytic layer are made of HNO ₃ / Etching and removing with HF mixed solution, and boiling with NH ₄ OH / H ₂ O ₂ / H ₂ O solution for cleaning, and then removing the oxide film with HF / H ₂ O solution, followed by HCl / H Boiled with H ₂ O ₂ / H ₂ O solution. Finally, after sufficiently rinsing with pure water, spin drying was performed, and the diffusion length was measured by the same method as described above. The results are also shown in Table 1. Then, in order to see the correlation between the change in diffusion length and the impurities in the silicon substrate, the concentration of Fe, which is the most well-known impurity that degrades the performance of solar cells, is measured by X-ray fluorescence analysis. ) Method. The measurement principle is as follows. When X-rays enter the sample and excite atoms, characteristic X-rays or Auger electrons are emitted in the relaxation process. That is, the electrons in the outer shell of the excited atom shift to the ground state by transitioning to the vacant level of the inner shell, and electromagnetic waves corresponding to the energy difference are emitted as characteristic X-rays or the like. The sample is excited using X-rays in this way, and the resulting X-rays are called fluorescent X-rays. Since the energy of the electron level that contributes to the transition is specific to the atom, the energy of the emitted fluorescent X-ray is specific to the atom. Therefore, qualitative analysis of elements can be performed by measuring fluorescent X-rays, and quantitative analysis can be performed by comparing intensities. Table 3 shows the results of the Fe concentration of the substrate after firing measured by this method. In addition, Shimadzu Corporation AXIS-ULTRA was used for the measurement.

Example 2
The diffusion length, Fe concentration, and solar cell characteristics were measured in the same manner as in Example 1 except that the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 70 ° C./s. The measurement results are shown in Table 1, Table 2 and Table 3.

Example 3
The diffusion length, Fe concentration, and solar cell characteristics were measured in the same manner as in Example 1 except that the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 55 ° C./s. The measurement results are shown in Table 1, Table 2 and Table 3.

Example 4
In order to show that the substrate modification effect according to the present invention is effective even if a substance other than the silicon nitride film is used as the antireflection film, titanium dioxide was used as the antireflection film in this example. By applying a solution containing titanium oxide (TiO _x ) to the surface and irradiating with ultraviolet rays, the coating film is dried, and the titanium oxide is oxidized or reduced, so that the thickness is mainly 10 to 100 nm. A titanium oxide antireflection film was formed. In this way, treatment and diffusion were performed in the same manner as in Example 1 except that the antireflection film was titanium dioxide and the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 70 ° C./s. The length, Fe concentration and solar cell characteristics were measured. The measurement results are shown in Table 1, Table 2 and Table 3.

  Next, a comparative example is presented in order to confirm the effectiveness of the present invention.

Comparative Example 1
The diffusion length, Fe concentration, and solar cell characteristics were measured in the same manner as in Example 1 except that the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 50 ° C./s. The measurement results are shown in Table 1, Table 2 and Table 3.

Comparative Example 2
The diffusion length, Fe concentration, and solar cell characteristics were measured in the same manner as in Example 1 except that the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 45 ° C./s. The measurement results are shown in Table 1, Table 2 and Table 3.

Comparative Example 3
The diffusion length, Fe concentration, and solar cell characteristics were measured in the same manner as in Example 1 except that the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 40 ° C./s. The measurement results are shown in Table 1, Table 2 and Table 3.

Comparative Example 4
In this comparative example, titanium dioxide was used as the antireflection film. By applying a solution containing titanium oxide (TiO _x ) to the surface and irradiating with ultraviolet rays, the coating film is dried, and the titanium oxide is oxidized or reduced, so that the thickness is mainly 10 to 100 nm. A titanium oxide antireflection film was formed. In this way, the treatment was performed in the same manner as in Example 1 except that the antireflection film was titanium dioxide, and the rate of temperature decrease from the maximum temperature in the heat treatment step for forming the back surface electrolytic layer was 45 ° C./s. The length, Fe concentration and solar cell characteristics were measured. The measurement results are shown in Table 1, Table 2 and Table 3.

As is clear from the results in Table 2, the energy conversion efficiency is 15.1 to 15.4% in the solar cell in the example, whereas 13.8 to 14.6% in the solar cell in the comparative example. Thus, it was found that the solar cell in the example has higher energy conversion efficiency. Next, among the results of Tables 1 and 3, Examples 1 to 3 and Comparative Examples 1 to 3 in which the antireflection film is a silicon nitride film are shown as graphs in FIGS. 3 and 4, respectively. From the results shown in FIG. 3, the diffusion length tends to improve as the rate of temperature decrease increases. In particular, when the rate of temperature decrease is 55 ° C./s or more, the improvement in diffusion length is significant. Further, from the results of FIG. 4, it was found that the improvement of the diffusion length correlates with the effect of removing impurities by gettering, and the diffusion length is improved as the number of impurities after baking is small. Looking again at Tables 1 and 3, it was found that this tendency was established even when the antireflection film was a titanium dioxide film, and therefore did not depend on the type of antireflection film. As described above, in general, there is a document that claims that rapid cooling from a high temperature state in RTP causes defects in crystals, but judging from the examples of the present invention and comparative examples, more defects are generated in the portion of the generated defects. It can be interpreted that the impurity element can be taken in. Furthermore, the vicinity of the interface between silicon and Al, in which impurities are incorporated, is sufficiently far from the pn junction portion of the silicon substrate, and not only the influence of the solar cell on the mechanism for generating current is small, but also the vicinity of the surface on the back side is This is probably because the p ⁺ layer doped with a large amount of Al is inherently low in diffusion length and relatively less affected by the occurrence of defects. For these reasons, it was determined that the superiority of the substrate modification effect during firing in this example was reflected in the solar cell characteristics shown in Table 2.

  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.

It is process drawing which shows the manufacturing method of the conventional silicon solar cell. It is a schematic diagram which shows the cross section of the baking furnace for rapid heat processing. It is a figure which shows the relationship between the temperature fall rate from the maximum temperature in a heat processing process, and diffusion length. It is a figure which shows the relationship between the temperature fall rate from the highest temperature in an annealing process, and impurity (Fe) density | concentration.

Explanation of symbols

1 p-type silicon substrate, 2 n-type semiconductor layer, 3 antireflection film, 4 aluminum paste, 5 p ⁺ layer, 6 back electrode, 7 light-receiving surface side electrode.

Claims (2)

  A solar cell manufacturing method having a p-type silicon substrate and a back surface field layer formed by firing aluminum on a surface opposite to the light receiving surface of the p type silicon substrate, the heat treatment step for forming the back surface field layer A method for producing a solar cell, wherein the temperature-decreasing rate from the maximum temperature is 55 ° C./s or more.   A solar cell having a p-type silicon substrate and a back surface electric field layer formed by firing aluminum on a surface opposite to the light receiving surface of the p type silicon substrate, the highest temperature in the heat treatment step for forming the back surface electric field layer The solar cell manufactured by the method characterized by the temperature-fall rate from 55 being 55 degree-C / s or more.

Images