JPH09148602A - Cis type solar battery and its manufacture - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable absorbing the difference of thermal expansion of glass and CIS alloy, with an electrode layer, by using a nickel phosphorus layer as an absorption layer side electrode. SOLUTION: After a soda lime glass board turning to a board 1 is dipped in potassium hydroxide aqueous solution (8wt.%), the glass board is washed by dipping it in mercleaner and a surface to be plated is roughened. Catalyst is stuck on the plated surface by dipping the glass board in conditioner. The catalyst is activated by dipping the glass board in activator. Chemical plating is performed by dipping the glass board in plating material, and a nickel phosphorus layer 2a composed of nickel (97wt.%) and phosphorus (3wt.%) is formed. An CIS layer 3 is formed on the nickel phosphorus layer 2a. By containing phosphorus in an absorption side electrode, the thermal expansion coefficient of an electrode layer is adjusted. The difference of thermal expansion coefficient of glass and CIS alloy at the time of high temperature heating in selenide treatment can be absorbed with the electrode layer. A chemical plating method is selected, and plating material is used.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a CIS type thin film solar cell and a method for manufacturing the same.

[0002]

2. Description of the Related Art A solar cell is a device for converting light energy into electric energy, and initially single crystal silicon was used. However, it is difficult to make a large area solar cell with single crystal silicon, and Because of high cost, amorphous silicon and compound semiconductors have been used recently. Among these compound semiconductors, copper - indium - selenium ternary alloy (CuInSe _2, hereinafter referred to as "CIS") is to have an excellent photoelectric conversion efficiency, the use of this as a p-type semiconductor absorber layer It is expected that a large area solar cell can be realized at low cost.

Such a solar cell will be described with reference to FIG. FIG. 2 is a cross-sectional view (model diagram) of a CIS solar cell. Reference numeral 1 in the figure is a substrate, and soda lime glass is usually used because of its thermal expansion coefficient being close to that of the CIS layer and its low price. Reference numeral 2 is a molybdenum layer which is an absorption layer side electrode. Reference numeral 3 is a p-type semiconductor CuInSe
A CIS layer which is an absorption layer composed of ₂ , 4 is a cadmium sulfide layer which is an n-type semiconductor, and 5 is a zinc oxide layer which is a window side electrode.

Molybdenum is selected as the material of the absorption layer side electrode for the following reasons. That is, the CIS layer as the absorption layer is formed by forming a layer made of copper and indium on the absorption layer side electrode (for example, by plating or sputtering), and using this as a precursor in an atmosphere containing hydrogen selenide or selenium vapor. At 350-6
It is obtained by heat treatment at 00 ° C. If the material of the back electrode is one that easily forms an alloy with selenium during the selenization treatment, expansion occurs as a result of the alloy formation, and the resulting CIS layer is cracked or peeled off, or the back electrode itself. The electrical conductivity of the device will be reduced and other problems will occur. Molybdenum is used as the back electrode because it has excellent selenium resistance during the selenization treatment.

However, in order to form the molybdenum layer, the method of applying a vacuum application technique such as sputtering is practically used. Here, the devices used for these vacuum application techniques are generally very expensive, and molybdenum itself is also expensive. In addition, it is difficult to increase the processing capacity per hour with these vacuum application techniques, so that the solar cell having the above molybdenum layer as the absorption layer side electrode must be expensive. Further, according to this vacuum application technique, it is very difficult to obtain an electrode layer having a large area and a uniform thickness.

Here, a semiconductor film forming substrate (Japanese Patent Laid-Open No. 4-304681) has been proposed in which a nickel metal layer is formed by using an electroless plating method instead of the molybdenum layer. However, although this method can be applied to an expensive alumina substrate, it is practically impossible to use a commonly used inexpensive glass substrate. That is, a nickel metal layer is formed on a glass substrate by electroless plating,
When a CIS layer is formed on this nickel metal layer, only those with many defects such as cracks and peeling are obtained, and as a result,
Yield is greatly reduced, or these CIs have defects.
When a solar cell is prepared by using one having an S layer, it causes a short circuit and a decrease in photoelectric conversion efficiency, which is a problem.

[0007]

SUMMARY OF THE INVENTION The present invention solves the above-mentioned problems of the prior art, that is, the number of steps by vacuum application technology that requires expensive equipment is reduced to a minimum, and the problems such as cracks and peeling are prevented. It is an object of the present invention to provide a method for obtaining an inexpensive, inexpensive CIS solar cell with high yield.

[0008]

The present invention has been made by paying attention to phosphorus in order to improve the drawbacks of the nickel metal layer which is the electrode layer in the above-mentioned prior art. That is, phosphorus is: ・ Adjusts the coefficient of thermal expansion of the electrode layer so that the electrode layer can absorb the difference in thermal expansion between the glass and the CIS alloy during high temperature heating during the selenization treatment. By incorporating phosphorus atoms into the crystal lattice, the lattice spacing is made suitable for joining with the CIS crystal. The covalent bond of phosphorus contributes to the binding with CIS (particularly indium and selenium in CIS), and It was selected to also contribute to the bonding with glass. That is, the CIS solar cell of the present invention has a structure having a nickel phosphorus layer as the absorption layer side electrode as described in claim 1.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION In the present invention, it is desirable that the nickel phosphorus layer has a phosphorus content of 3% by weight or more and less than 10% by weight, and the balance being substantially nickel. When the phosphorus content is less than 3% by weight, the effect of the present invention cannot be sufficiently obtained, and as a result, C formed on the electrode layer is not formed.
Problems such as cracks and peeling are likely to occur in the IS layer.
On the other hand, when the phosphorus content is 10% by weight or more, the selenium resistance in the selenization treatment step is deteriorated and the conductivity is deteriorated, resulting in a marked decrease in photoelectric conversion efficiency.

The selenization resistance of the nickel phosphorus layer according to the present invention in the selenization treatment is almost the same as that of the molybdenum layer, and the selenization value is less than 2%, which is currently considered as an upper limit standard. Further, the method for producing a CIS type solar cell of the present invention is a method suitable mainly for using glass as a substrate as described above, but can be applied to a case where alumina is used as a substrate.

The nickel phosphorus layer as described above can be obtained by an electroplating method using a nickel plating solution containing a phosphorus component, and can also be obtained by a chemical plating solution similarly containing a phosphorus component. it can. By adjusting the phosphorus component concentration in the plating solution or the pH of the plating solution in these, a nickel phosphorus layer having an appropriate phosphorus content can be obtained. In particular, the nickel-phosphorus layer obtained by the chemical plating method is more uniform and has less thickness unevenness as compared with the nickel layer formed by the sputtering method, and it is easy to manufacture a large-area solar cell. Further, in the nickel-phosphorus layer, it is desirable that the component other than phosphorus is only nickel, but some components essential in the plating step (such as a brightener) may be incorporated as impurities.

The CIS layer can be obtained, for example, by forming a copper layer and an indium layer on the above nickel phosphorus layer by, for example, an electroplating method, and then performing heat treatment in an atmosphere containing selenium. Here, a selenium colloid-dispersed indium plating bath is used to form a selenium colloid-dispersed indium layer on a copper layer to form a precursor, and the precursor is heat-treated in an atmosphere containing selenium to provide a spatial margin for expansion accompanying the formation of CIS alloy. Therefore, it is possible to obtain a very good CIS layer with particularly few defects.

As the atmosphere containing selenium, hydrogen selenide gas is generally used. This hydrogen selenide gas is extremely toxic, and special care and equipment are required for its handling. Here, selenium in the form of powder or the like is sealed together with the above-mentioned precursor in a closed container replaced with an inert gas and heat-treated to generate selenium vapor, so that it is safer than in the case of using hydrogen selenide. In addition, the selenization treatment can be performed at low cost.

A cadmium sulfide layer which is an n-type semiconductor is formed on the CIS layer thus formed. This can be formed by a solution growth method or a vapor deposition method. Especially in the case of the solution growth method, a special apparatus is not required as compared with the vapor deposition method, the productivity is good, and a uniform thin layer can be obtained even when manufacturing a large-area solar cell, which is preferable.

On the cadmium sulfide layer, a zinc oxide layer doped with aluminum for improving conductivity is formed as a window side electrode by sputtering or the like. The lamination is completed in this way, and then lead wires and the like are connected to the zinc oxide layer which is the window side electrode and the nickel phosphorus layer which is the absorption layer side electrode to complete the CIS solar cell.

[0016]

【Example】

<Example 1> A square solar cell having a light receiving portion with a side length of 15 mm (having a nickel phosphorus layer as an absorption layer side electrode) was produced. Here, in forming the nickel phosphorus layer, a chemical plating method was selected and a plating agent manufactured by Meltex was used. First, a soda lime glass plate used as a substrate is immersed in an 8 wt% potassium hydroxide aqueous solution at 75 ° C. for 5 minutes, and then immersed in Mel Cleaner 170 manufactured by Meltex Co. at 60 ° C. for 5 minutes to wash the glass plate. At the same time, the plated surface was roughened.

Thereafter, the catalyst is attached to the plating surface by immersing it in conditioner 1101 (room temperature) manufactured by the same Meltex Co., Ltd., and then activator 44 manufactured by the same Meltex company.
The catalyst was activated by immersing it in 0 (room temperature) for 1 minute, and finally immersed in a plating agent Melplate Ni866 (manufactured by Meltex Co., Ltd.) kept at 80 ° C. for 80 minutes for chemical plating.
In addition, in each of the above-mentioned steps, a portion which does not need to be treated was masked with an adhesive sheet to prevent the treatment.

When the composition of the plated layer thus obtained was examined by using a micro fluorescent X-ray analyzer (the same applies hereinafter), it was found that it consisted of 97% by weight of nickel and 3% by weight of phosphorus. . Also, the plating layer is good,
The thickness was measured with a microscopic fluorescent X-ray device to be 2 μm.
It was found that there were almost no plating spots on the entire plated surface.

Next, a CIS layer was formed on this nickel phosphorus layer. First, a copper layer was formed by electroplating. As a plating solution, copper (II) pyrophosphate trihydrate (Cu ₂ P ₂ O ₇・ 3) is used.
H ₂ O) (80 g) and potassium pyrophosphate (300 g) were dissolved in water to make 1 liter, and a pyrophosphate complex salt bath at 50 ° C. prepared by adding 5 ml of 28% by weight ammonia water was used, and electrolytic copper was used as an anode. Electroplating was performed so that the current density was ² Adm ^-2 . The thickness of this copper layer (the center of the plated surface) was 0.3 μm.

Then, a selenium colloid-dispersed indium plating solution was used as a plating solution to obtain a selenium colloid-dispersed indium layer. First, a selenium colloidal solution was prepared by the following procedure. After mixing 4 ml of gelatin solution (4 g / l), 10 ml of selenious acid (0.1 mol / l), 10 ml of hydrazinium sulfate (0.1 mol / l), and 30 ml of pure water, 40
After heating at 45 ° C. for 45 minutes, the pH was adjusted to 2 to stabilize the colloid to obtain a selenium colloid solution. here,
Gelatin is added to stabilize the colloid.

When the colloidal particles of the selenium colloidal solution thus prepared were examined by a scanning electron microscope,
The particle size of the selenium colloidal particles was 10 nm or more and 100 nm or less, and the selenium particles did not precipitate even when left standing for a long time and were very stable.

Using this selenium colloid solution, selenium colloid 10 mmol / l, indium sulfate 50 mmol / l,
Sodium sulfate 80 mmol / l, sodium citrate 50
The plating solution was adjusted to be mmol / l. The thickness of the plating layer (selenium-dispersed electrodeposited indium layer) at the center of the plating surface was adjusted on the copper layer using the plating solution at a voltage of 1.5 V with respect to a saturated mercuric sulfate electrode as a reference electrode. 0.7
Constant voltage plating was performed so that the thickness was 0 μm. At this time, a platinum plate was used as the counter electrode.

Since the plating solution was very stable and colloidal particles in the plating solution did not settle, stirring was not performed during plating. The abundance ratio of indium atoms and selenium atoms in the electrodeposited indium layer containing these selenium colloidal particles dispersed therein was 10: 2 according to composition analysis. The selenium-dispersed electrodeposited indium layer was very uniform.

The conductive substrate on which the copper layer and the selenium-dispersed electrodeposited indium layer were thus formed was placed together with selenium powder in a closed container which had been replaced with argon gas, and the pressure was 500 Torr.
Heat treatment was performed at 500 ° C. for 1 hour. The selenium atmosphere was used as described above in order to prevent selenium from becoming insufficient during the formation of the CIS layer. As a result of analysis, the obtained layers have atomic abundance ratios of copper, indium and selenium of 23, 24 and 53, respectively,
It was confirmed that the ratio was almost the same as that of the CuInSe ₂ alloy. Further, this CIS layer was excellent without cracking or peeling.

A cadmium sulfide layer was formed on the CIS layer by liquid phase epitaxy. That is, the soda lam glass having the above thin film layer formed thereon was dipped in a 0.01 mol / l aqueous solution of cadmium sulfate (CdSO ₄ ) at 70 ° C. (pH adjusted to 8.5 with ammonia water) for 5 minutes, and then, Then, a thiourea 0.01 mol / l aqueous solution was added and the mixture was kept for 5 minutes to grow a cadmium sulfide layer, thoroughly washed with water, and then dried under nitrogen. The thickness of the obtained cadmium sulfide layer is 0.0
It was 5 μm.

Thereafter, a 2 μm thick thin film of a zinc oxide layer doped with 2% by weight of aluminum was formed by a sputtering method to obtain a thin film solar cell according to the present invention. A cross-sectional view (model diagram) of this product is shown in FIG. In the figure, reference numeral 1 is a substrate made of soda lime glass, 2a is a nickel phosphorus layer, 3 is a CIS layer, 4 is a cadmium sulfide layer, and reference numeral 5 is a zinc oxide layer. In addition, in the same manner as above, four solar cells (solar cells 1 to 5: Example 1) were manufactured in addition to the other four solar cells. For all of them, a good thin film was obtained in each step, and finally, The product obtained in Example 2 was also excellent without defects such as peeling. The characteristics of a circular portion having a diameter of 2 mm at the center of each of the solar cells 1 to 5 were measured using a solar cell output characteristic device (manufactured by WACOM). Table 1 shows the results.

[0027]

[Table 1]

<Examples 2 and 3> As in Example 1 above, the solar cells 6 to 10 were prepared by adjusting the pH of the plating solution Melplate Ni868 using 1N-hydrochloric acid and 1N-sodium hydroxide aqueous solution. (Example 2) and solar cell 1
1 to 15 (Example 3), 5 solar cells each of 2 types, 10 solar cells in total were obtained.

In the solar cells 6 to 10, the phosphorus content of the nickel phosphorus layer was 8% by weight, and in the solar cells 11 to 15, the phosphorus content of the nickel phosphorus layer was 10% by weight. The performance of these solar cells 6 to 11 was measured in the same manner as the solar cell of Example 1. The results are shown in Tables 2 and 3.

[0030]

[Table 2]

[0031]

[Table 3]

Comparative Example As a comparative example, an attempt was made to manufacture a solar cell using an electrode made of nickel metal formed by electroless plating. The same 5 soda lime glasses as in Example 1 were used as substrates, and these were subjected to etching treatment with a solution containing hydrofluoric acid for 10 minutes, and then the substrate surface was subjected to activation treatment. Then, using a weakly acidic solution containing Ni, a Ni film was formed on each of the five substrates by electroless plating at a liquid temperature of 90 ° C. These Ni
A CIS layer was formed on the film as in Example 1.

When these CIS layers were observed in detail, two out of five samples were cracked and peeled,
In addition, another one had a small crack. About these three defects, it was expected that a failure such as a short circuit would occur, but a cadmium sulfide layer and a zinc oxide layer were formed as in Example 1.
After that, the performance as a solar cell was examined, but it was found that none of them could be actually used. On the other hand, with respect to the remaining two samples (those having no crack in the CIS layer and no peeling), a cadmium sulfide layer and a zinc oxide layer were formed in the same manner as in Example 1, and a solar cell with good performance was obtained. From the above Examples and Comparative Examples, it is understood that a good CIS solar cell can be manufactured with a good yield by forming a CIS layer on a nickel phosphorus layer having an appropriate content of phosphorus as an electrode.

<< Investigation of Selenium Resistance >> The nickel phosphorus layer and the molybdenum layer were investigated for resistance to selenium.
On each of the soda lime glass plates, a square nickel phosphorus layer having a thickness of 20 μm and a side length of 15 mm was formed in the same manner as in Example 1. For comparison, a soda lime glass plate having a molybdenum layer of the same size and the same thickness was formed by a sputtering method.

Each of these two types of glass plates was sealed in a closed container together with 10 mg of selenium powder, and the temperature was 500 ° C.
Time heat treatment was performed. When the composition of these layers was analyzed, the amount of selenium infiltrated into the nickel phosphorus layer was 1.8% by weight, and the amount of selenium infiltrated into the molybdenum layer was 1.2% by weight, which were almost the same level. Yes, there was no problem in practical use.

[0036]

According to the method for manufacturing a CIS solar cell of the present invention, an electrode layer can be formed without using an expensive device or expensive molybdenum. In addition, by producing a CIS solar cell using an electrode layer having a nickel phosphorus layer, it is possible to obtain the effect of improving the yield rate and productivity.
Therefore, it is possible to supply a CIS type solar cell having no defects at a lower cost.

[Brief description of the drawings]

FIG. 1 is a model diagram showing a cross section of a CIS solar cell according to the present invention.

FIG. 2 is a model diagram showing a cross section of a CIS solar cell according to a conventional technique.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Substrate 2a ... Nickel phosphorus layer 2 ... Molybdenum layer 3 ... CIS layer 4 ... Cadmium sulfide layer 5 ... Zinc oxide layer

Claims (4)

[Claims] 1. A CIS solar cell having a nickel phosphorus layer as an absorption layer side electrode. 2. The phosphorus content of the nickel phosphorus layer is 3% by weight.
The CIS solar cell according to claim 1, wherein the amount is less than 10% by weight. 3. A method for manufacturing a CIS solar cell, comprising forming a nickel phosphorus layer on a substrate and forming a CuInSe ₂ alloy layer on the nickel phosphorus layer. 4. The phosphorus content of the nickel phosphorus layer is 3% by weight.
The above is less than 10% by weight, and the method for producing a CIS solar cell according to claim 3, wherein