JP2006324519A - Manufacturing method of solar cell - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a manufacturing method of a solar cell, with which the highly efficient solar cell where contact resistance between an electrode and a substrate is largely reduced and the electrode is not peeled, can inexpensively and stably be obtained with high productivity. <P>SOLUTION: In the manufacturing method of the solar cell, the electrode is formed by printing and burning conductive paste after PN-junction is formed on the semiconductor substrate and immersing the semiconductor substrate in acid for once or more. The semiconductor substrate where conductive paste is printed and burnt is wetted by hydrophilic solvent and it is immersed in acid without drying it. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  In the present invention, at least after forming a PN junction on a semiconductor substrate, a conductive paste is printed and baked, and the electrode is formed by immersing the semiconductor substrate in an acid at least once. The present invention relates to a method for manufacturing a high-efficiency solar cell with reduced contact resistance between the two.

A solar cell includes at least a substrate portion on which a PN junction that converts sunlight light energy into electric energy is formed, and an electrode portion that is attached to the substrate portion and takes out electric energy to the outside. In the vicinity of the interface between the two, a large contact resistance occurs due to the discontinuity of the material.
Since the contact resistance between the electrode and the substrate decreases the electric energy that can be extracted, it greatly affects the efficiency of the solar cell. For this reason, many attempts have been made to optimize the solar cell structure and improve the electrode formation method, electrode material, and substrate material for the purpose of reducing the contact resistance.

One of the electrode materials currently in use is a conductive paste. Conductive pastes are widely used because screen printing can be applied to electrode forming methods.
However, there is a problem that the contact resistance in a solar cell using a conductive paste as an electrode material is higher than the contact resistance in a solar cell in which an electrode is produced by a PVD method such as vacuum deposition or sputtering.

As one method for reducing the contact resistance of the conductive paste, there is a method in which a conductive paste as shown in FIG. 2 is printed on a substrate and baked, and then immersed in an acid (see Patent Document 1).
The principle of contact resistance reduction in this method is not clear, but the following hypothesis is considered.

The conductive paste after firing is mainly composed of metal particles and glass frit, and is in a porous state having pores therein. This glass frit has a role as an adhesive for maintaining contact between metal particles or between a metal and a substrate.
However, on the other hand, this glass frit also causes an increase in contact resistance at the contact portion due to non-conductivity.
Therefore, it is said that by dissolving and removing the glass frit in the vicinity of the interface between the substrate and the electrode with an acid, the contact point between the substrate and the metal particles in the electrode is increased, so that the contact resistance is reduced.

However, this method for reducing contact resistance by acid immersion has problems that the variation in the contact resistance reduction range is large and that the electrodes are easily peeled off.
The reduction width of the contact resistance varies depending on the acid concentration and the immersion time. For example, when the acid concentration is lowered to shorten the acid dipping time, the acid dipping effect is low, and a sufficiently low contact resistance cannot be obtained. On the other hand, if the acid concentration is increased to increase the acid immersion time in order to enhance the effect of acid immersion, peeling of the electrode occurs.

  This is because, as can be seen from the above hypothesis, when a strong reaction between the glass frit and the acid occurs, the force of the glass frit as an adhesive decreases, and the electrode peels off. In particular, this peeling strongly occurs at the tip of the finger electrode. This is because the processed carrier volume at the electrode tip is larger than in other parts.

Furthermore, the range of reduction in contact resistance varies not only depending on the acid concentration and immersion time, but also greatly varies depending on individual differences among the manufactured solar cells.
This means that even if a semiconductor substrate printed and baked with conductive paste is immersed in the same acid at the same concentration for the same immersion time, it is difficult to control so that the portion where the acid penetrates and reaches is uniform. This is probably because the non-uniformity of the portion where the glass frit can be melted and removed is large. In order to sufficiently permeate the acid, it has been considered to increase the immersion time, but this facilitates peeling of the electrode and increases the process time.

JP-A-9-213979

  Therefore, the present invention has been made in view of the above problems, and an object of the present invention is to reduce the contact resistance between the electrode and the substrate, and to prevent the electrode from peeling off. It is an object of the present invention to provide a solar cell manufacturing method capable of stably obtaining a solar cell with high productivity and low cost.

  In order to achieve the above object, according to the present invention, at least after forming a PN junction on a semiconductor substrate, the conductive paste is printed and baked, and the semiconductor substrate is immersed in an acid at least once. In the manufacturing method of the solar cell which forms an electrode, after wetting the semiconductor substrate which printed and baked the said conductive paste with a hydrophilic solvent, it is immersed in an acid without drying, The manufacturing method of the solar cell characterized by the above-mentioned Is provided (claim 1).

Thus, after the semiconductor substrate printed and baked by wetting the conductive paste is wetted with a hydrophilic solvent, it is immersed in an acid without drying, whereby the outer periphery and the center of the baked conductive paste forming the electrode are formed. The glass frit can be uniformly exposed to the acid at the part, so that the glass frit can be uniformly dissolved and removed, and the in-plane distribution of the contact resistance can be made uniform, resulting in sufficiently low contact resistance. Can be obtained stably. In addition, since the number of locations that are locally exposed to acid for a long time decreases, peeling of the electrodes can be prevented.
This is because by immersing the substrate in the acid without drying, the acid diffuses and penetrates into the hydrophilic solvent already present in the pores in the fired conductive paste. This is because the uniformity of the reaction between the acid and the glass frit increases, and the hydrophilic solvent present in the vicinity of the reaction region dilutes the acid, so that the reaction proceeds slowly and the stability of the reaction also increases. .

In this case, it is preferable that the hydrophilic solvent is water or an aqueous solution.
Thus, by using a hydrophilic solvent as water or an aqueous solution, it is possible to provide a hydrophilic solvent that is an acid solvent that can be easily immersed in the pores in the baked conductive paste forming the electrode at low cost. it can.

The hydrophilic solvent is preferably pure water (Claim 3).
By using pure water as the hydrophilic solvent in this manner, it is possible to prevent contaminants from adhering to the substrate and the electrode and deteriorating the quality.

Furthermore, it is preferable that the method of wetting with the hydrophilic solvent is immersion in a liquid tank filled with the hydrophilic solvent.
Thus, by immersing in the liquid tank filled with the hydrophilic solvent, the entire conductive paste fired on the substrate can be sufficiently wetted with the hydrophilic solvent.

Further, the method of wetting with the hydrophilic solvent may be a spraying of the hydrophilic solvent (Claim 5).
Thus, it may be wetted by spraying the solvent. In this case, the liquid tank is not necessary, and therefore the process can be simplified.

Furthermore, it is preferable to apply ultrasonic waves when the semiconductor substrate is wetted with a hydrophilic solvent.
By applying ultrasonic waves when the substrate is wetted with the solvent in this way, the solvent can be quickly and surely penetrated into the conductive paste baked into the substrate, and the process time can be shortened. it can.

Moreover, it is preferable that the time of wetting with the hydrophilic solvent be 1 second or more and 10 minutes or less (claim 7).
Thus, by setting the time of wetting with the solvent to 1 second or more and 10 minutes or less, it is possible to surely wet the baked conductive paste.

Furthermore, it is preferable to use hydrofluoric acid or hydrobromic acid as the acid for immersing the semiconductor substrate (claim 8).
By using hydrofluoric acid or hydrobromic acid as the acid for immersing the substrate in this way, excellent electrical characteristics can be obtained and electrode corrosion does not occur.

Moreover, it is preferable to perform the acid immersion of the semiconductor substrate by immersing the semiconductor substrate in an aqueous solution containing 0.1 to 10% by volume of acid (claim 9).
Thus, the glass frit at the interface between the electrode and the substrate can be sufficiently dissolved and removed by immersing the substrate in an aqueous solution containing 0.1 to 10% by volume of acid.

Furthermore, it is preferable to use a p-type single crystal silicon substrate doped with gallium as the semiconductor substrate.
Thus, by using a p-type single crystal silicon substrate doped with gallium as the semiconductor substrate, the solar cell to be manufactured becomes practical with a very high photoelectric conversion efficiency without causing photodegradation.

  As described above, according to the method for manufacturing a solar cell of the present invention, a sufficiently low contact resistance can be stably obtained by one acid immersion, so that it is not always necessary to perform acid immersion multiple times, and further, electrode peeling. Therefore, a highly efficient solar cell can be manufactured with high productivity and low cost.

  Conventionally, in a solar cell manufacturing process, after a conductive paste is printed on a semiconductor substrate and baked, it is immersed in an acid in order to reduce contact resistance. However, the contact resistance reduction width obtained by such acid immersion varies greatly even when immersed in the same acid at the same concentration for the same immersion time, and if the immersion time is increased to sufficiently penetrate the acid. In addition to promoting the peeling of the electrodes, the process time is increased.

  Therefore, the present inventors have conducted extensive research, and at least after forming a PN junction on a semiconductor substrate, the conductive paste is printed and fired, and the semiconductor substrate is immersed in an acid one or more times to form an electrode. In the method of manufacturing a solar cell, the electrode is formed by dipping the semiconductor substrate printed and fired with the conductive paste with a hydrophilic solvent and then immersing it in an acid without drying. The time at which the glass frit is in contact with the acid can be made uniform at the periphery and center of the paste, so that the glass frit can be uniformly dissolved and removed, and the in-plane distribution of contact resistance can be made uniform. , Sufficiently low contact resistance can be stably obtained, and it is not necessary to lengthen the immersion time, so the number of places exposed to acid locally for a long time decreases. Found that peeling of the electrode can be prevented in order.

  Hereinafter, embodiments of the present invention will be described in detail, but the present invention is not limited thereto. FIG. 1 is a flowchart showing an embodiment of a method for manufacturing a solar cell of the present invention.

First, after forming a PN junction on a semiconductor substrate, a conductive paste is printed and baked.
The semiconductor substrate used here is preferably a p-type single crystal silicon substrate doped with gallium, which makes the solar cell to be manufactured practical with a very high photoelectric conversion efficiency that does not cause photodegradation. First, after removing the damaged layer from the semiconductor substrate by etching, it is preferable to form a PN junction on the semiconductor substrate on which the texture structure for preventing reflection is formed.
The PN junction is preferably formed by thermal diffusion of n-type impurities such as phosphorus on the light receiving surface side, but may be formed by coating diffusion or ion implantation. Here, in order to prevent sunlight reflection and protect the surface, it is preferable to form a nitride film on the light receiving surface by a plasma CVD method or a PVD method.

The conductive paste is printed with a conductive paste containing silver particles on the light-receiving surface side, and comb-like finger electrodes that extract power from the substrate and bus bar electrodes that extract power from the finger electrodes are printed on the back surface (light-receiving surface). It is preferable to print a conductive paste containing aluminum particles on the entire surface on the opposite side).
Firing of the conductive paste printed on the substrate can be performed by heating at 500 to 800 ° C. for 1 to 5 minutes.

Next, the substrate on which the conductive paste has been baked is wetted with a hydrophilic solvent. If the substrate is wetted with a hydrophilic solvent, the surface of the substrate can be quickly and uniformly wet when immersed in an aqueous solution such as hydrofluoric acid or hydrobromic acid, allowing the reaction to proceed evenly. . In addition, bubbles can be prevented from adhering to the substrate surface when immersed in an acid.
The hydrophilic solvent is not particularly limited as long as it functions sufficiently as a solvent for the acid to be immersed and can prevent contamination, but it is particularly preferable to use pure water. Moreover, as long as it becomes a solvent of the acid to immerse, the hydrophilic solvent of aqueous solution and alcohol (for example, methanol, ethanol, propanol, isopropanol etc.) may be sufficient.

  In the method of wetting the substrate with the hydrophilic solvent, it is preferable to immerse the substrate in a liquid tank filled with the hydrophilic solvent because the entire conductive paste fired on the substrate can be sufficiently wetted with the hydrophilic solvent. In this case, a liquid tank is not necessary, and the process can be simplified. For example, by spraying the hydrophilic solvent while the substrate is being transported to the acid bath, the substrate can be easily wetted without increasing the process time.

  In addition, it is preferable to apply ultrasonic waves when the substrate is wetted with a solvent. This allows water to quickly penetrate into the electrode, so that the process time can be shortened and water can penetrate into the pores inside the electrode. Uniform and reliable. The application of ultrasonic waves is preferably performed indirectly through a solvent in which the substrate is immersed, but may be performed directly on the substrate.

  The time of wetting with the solvent is preferably 1 second or more and 10 minutes or less, so that the entire conductive paste fired on the substrate can be reliably wetted.

Next, the semiconductor substrate wetted with the hydrophilic solvent is immersed in an acid without drying.
Conventionally, since a dry semiconductor substrate is directly immersed in an acid, bubbles are attached to the surface, and a part where the acid penetrates and reaches within the immersion time and a part where it does not occur are generated. The non-uniformity of the portion that can be removed by melting and the portion that cannot be removed is large, resulting in variations in contact resistance and peeling of the electrodes.

  However, by dipping the substrate in a solvent and then immersing it in an acid without drying it, the time for the glass frit to come into contact with the acid at the outer periphery and the center of the baked conductive paste forming the electrode is made uniform. Therefore, it is possible to uniformly dissolve and remove the glass frit. Therefore, as a result of uniforming the in-plane distribution of contact resistance, a sufficiently low contact resistance can be obtained stably, and the number of places exposed to acid locally for a long time is reduced, so that the electrode is peeled off. Can be prevented.

  This is because by immersing the substrate in the acid without drying, the acid diffuses and penetrates into the hydrophilic solvent already present in the pores in the fired conductive paste. This is because the uniformity of the reaction between the acid and the glass frit increases, and the hydrophilic solvent present in the vicinity of the reaction region dilutes the acid, so that the reaction proceeds slowly and the stability of the reaction also increases. .

  As the acid for immersing the substrate, hydrofluoric acid, hydrobromic acid, DL-malic acid, stearic acid, adipic acid, salicylic acid, citric acid, and lactic acid are excellent electrical materials without causing electrode corrosion problems. Since it is possible to obtain characteristics, hydrofluoric acid is particularly preferable, because very excellent electrical characteristics can be obtained. In addition, hydrobromic acid and DL-malic acid are preferable. It is preferable because high electrode strength can be obtained.

  It is preferable to immerse the substrate in an aqueous solution containing 0.1 to 10% by volume of an acid because a sufficient amount of acid can be surely immersed in the conductive paste. If the acid can be immersed in the solution, the acid may be dissolved in another hydrophilic solvent such as alcohol. The immersion time is preferably about 1 second to 5 minutes.

Next, the acid adhering to the semiconductor substrate is washed away with water. Here, it is preferable to perform washing with water to increase the cleaning effect or while applying ultrasonic waves.
Finally, the semiconductor substrate is dried to complete a highly efficient solar cell. As the drying method, standing, warm air, IPA drying, or the like may be used.

  According to the method for manufacturing a solar cell of the present invention as described above, a sufficiently low contact resistance can be stably obtained by one acid immersion, so there is no need to perform acid immersion multiple times, and electrode peeling is also possible. Therefore, a highly efficient solar cell can be manufactured with high productivity and low cost. However, acid dipping may be performed a plurality of times depending on the purpose.

EXAMPLES Hereinafter, although an Example and a comparative example are shown and this invention is demonstrated more concretely, this invention is not limited to these.
(Examples 1 and 2)
After the solar cell was manufactured according to the procedure shown in FIG. 1, the measurement result of the output characteristics and the evaluation of the adhesive strength of the electrode were performed.
First, a p-type single crystal solar cell silicon substrate (100 mm square, surface orientation {100}, substrate thickness 300 μm, resistivity 0.5 Ωcm) having a group III element gallium as an impurity element is etched with an aqueous potassium hydroxide solution. The damage layer was removed. Further, a texture structure as an antireflection structure was formed with an aqueous potassium hydroxide solution mixed with IPA.

Then, an n region having phosphorus as a group V element as an impurity was formed on the light receiving surface by thermal diffusion using a POCl ₃ liquid source on the light receiving surface side to form a PN junction. Here, a 70 nm-thick nitride film was formed on the light-receiving surface by plasma CVD to prevent sunlight reflection and protect the surface.

  Further, a conductive paste containing aluminum particles is printed on the entire surface of the back surface (the surface opposite to the light receiving surface) and baked, and the conductive paste containing silver particles is applied to the finger electrode and the bus bar on the light receiving surface. It printed on the shape of the electrode and baked. Firing was performed by holding the substrate at 700 ° C. for 3 minutes.

Next, the substrate was wetted by immersing it in a water tank filled with pure water for 60 seconds. During immersion, ultrasonic waves were applied to pure water.
Then, the substrate wetted with pure water was immersed in 1 vol% hydrofluoric acid (hydrofluoric acid) for 45 seconds (Example 1) or 60 seconds (Example 2) without drying.
Next, the substrate was immersed for 5 minutes in a water tank filled with pure water and washed, and then the substrate was dried with warm air to manufacture a solar cell.

Next, the output characteristic of the produced solar cell was measured using a solar simulator (light intensity: 1 kW / m ² , spectrum: AM1.5 global).
In addition, the evaluation of the adhesive strength of the electrode is (1) the measurement of the probability of electrode peeling by visual confirmation of the occurrence of peeling of the finger electrode, and (2) the adhesive tape is applied to the finger electrode and the electrode remains on the substrate after peeling. (3) When the ribbon is soldered on the bus bar and the ribbon is pulled with a force of 2N in the vertical direction of the board, the board and the bus bar are It was performed by three kinds of methods of measuring the occurrence probability of peeling by a bus bar peeling test by visually confirming whether or not the peeling occurs.

(Comparative Examples 1 and 2)
After manufacturing the solar cell according to the conventional procedure shown in FIG. 2, the output characteristics were measured and the adhesion strength of the electrodes was evaluated. Specifically, all procedures were performed in the same manner as in Examples 1 and 2 except that the step of wetting the substrate with pure water was not performed before the acid immersion. In Comparative Examples 1 and 2, as in Examples 1 and 2, they were immersed in 1% by volume of hydrofluoric acid (hydrofluoric acid) for 45 seconds (Comparative Example 1) and 60 seconds (Comparative Example 2).

  Table 1 shows the measurement results of the output characteristics and the evaluation results of the adhesion strength of the electrodes of the solar cells manufactured by performing the above hydrofluoric acid immersion.

(Examples 3 and 4)
Moreover, in order to confirm the effect of this invention about the case where acid immersion is performed with other acids, after performing acid immersion with 1 volume% hydrobromic acid and manufacturing a solar cell, measurement of output characteristics and electrodes Evaluation of the adhesive strength was performed. Specifically, all procedures were performed in the same manner as in Examples 1 and 2 except that the sample was immersed in 1% by volume of hydrobromic acid instead of being immersed in 1% by volume of hydrofluoric acid. In Examples 3 and 4, as in Examples 1 and 2, they were immersed in 1% by volume of hydrobromic acid for 45 seconds (Example 3) and 60 seconds (Example 4).

(Comparative Examples 3 and 4)
Further, for comparison in the case of acid immersion with 1% by volume of hydrobromic acid, after the solar cell was manufactured according to the conventional procedure shown in FIG. 2, the output characteristics were measured and the adhesion strength of the electrode was evaluated. went. Specifically, all steps were performed in the same manner as in Examples 3 and 4 except that the step of wetting the substrate with pure water was not performed before the acid immersion. In Comparative Examples 3 and 4, as in Examples 3 and 4, they were immersed in 1% by volume hydrobromic acid for 45 seconds (Comparative Example 3) and 60 seconds (Comparative Example 4).

  Table 2 shows the measurement results of the output characteristics of the solar cells manufactured by performing the above hydrobromic acid immersion and the evaluation results of the adhesive strength of the electrodes.

From the evaluation results summarized in Tables 1 and 2, first, when immersed in hydrofluoric acid, the solar cell series resistance decreased by 0.49 Ωcm ^{2 in} the immersion time of 45 seconds according to the solar cell manufacturing method of the present invention. The fill factor was improved by 2.7% and the conversion efficiency by 0.7% (see Example 1 and Comparative Example 1). When the immersion time was 60 seconds, the series resistance was reduced by 1.23 Ωcm ² , the fill factor was improved by 6.2%, and the conversion efficiency was improved by 1.7% (see Example 2 and Comparative Example 2).

When immersed in hydrobromic acid, when the immersion time was 45 seconds, the series resistance of the solar cell was reduced by 0.34 Ωcm ² , the fill factor was improved by 1.6%, and the conversion efficiency was improved by 0.4% (implementation) See Example 3 and Comparative Example 3). Even when the hydrobromic acid immersion time was 60 seconds, the series resistance was reduced by 1.19 Ωcm ² , the fill factor was improved by 5.8%, and the conversion efficiency was improved by 1.5% (see Example 4 and Comparative Example 4).

In addition, from the results of evaluation of the adhesive strength of the electrodes, the probability of electrode peeling was determined in each test method when both the hydrofluoric acid and hydrobromic acid were immersed in pure water for 45 seconds and 60 seconds. The value of the example was significantly lower than that of the comparative example which was not wetted. In particular, in Examples 1 and 3 where the acid immersion time was 45 seconds, no electrode peeling occurred.
Compared with hydrofluoric acid immersion and hydrobromic acid immersion, hydrobromic acid is less susceptible to electrode peeling than hydrofluoric acid, and hydrofluoric acid is superior to hydrobromic acid. It can be seen that the electrical characteristics tend to be obtained.

  As described above, the contact resistance between the electrode and the substrate is greatly reduced by the solar cell manufacturing method of the present invention, as shown by the decrease in series resistance, improvement in conversion efficiency, and reduction in the probability of electrode peeling. The obtained high-efficiency solar cell can be stably obtained with high productivity and low cost.

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

It is a flowchart which shows embodiment of the manufacturing method of the solar cell of this invention. It is a flowchart which shows embodiment of the manufacturing method of the conventional solar cell.

Claims (10)

  In the method for manufacturing a solar cell, in which at least a PN junction is formed on a semiconductor substrate, and then an electroconductive paste is printed and fired, and the electrode is formed by immersing the semiconductor substrate in an acid at least once. A method for producing a solar cell, wherein a semiconductor substrate obtained by printing and baking a paste is wetted with a hydrophilic solvent and then immersed in an acid without drying.   The method for manufacturing a solar cell according to claim 1, wherein the hydrophilic solvent is water or an aqueous solution.   The method for manufacturing a solar cell according to claim 1, wherein the hydrophilic solvent is pure water.   The method for manufacturing a solar cell according to any one of claims 1 to 3, wherein the method of wetting with the hydrophilic solvent is immersion in a liquid tank filled with the hydrophilic solvent.   The method for manufacturing a solar cell according to any one of claims 1 to 3, wherein the method of wetting with the hydrophilic solvent is spraying of the hydrophilic solvent.   The method for manufacturing a solar cell according to claim 1, wherein ultrasonic waves are applied when the semiconductor substrate is wetted with a hydrophilic solvent.   The method for producing a solar cell according to any one of claims 1 to 6, wherein a time of wetting with the hydrophilic solvent is 1 second or more and 10 minutes or less.   The method for manufacturing a solar cell according to any one of claims 1 to 7, wherein hydrofluoric acid or hydrobromic acid is used as an acid for immersing the semiconductor substrate.   The solar cell production according to any one of claims 1 to 8, wherein the acid immersion of the semiconductor substrate is performed by immersing the semiconductor substrate in an aqueous solution containing 0.1 to 10% by volume of an acid. Method.   The method for manufacturing a solar cell according to claim 1, wherein a p-type single crystal silicon substrate doped with gallium is used as the semiconductor substrate.

Images