JP2014017377A - Compound solar cell and method for manufacturing the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide: a compound solar cell having no damage within a transparent conductive layer, in a buffer layer serving as a substrate of the transparent conductive layer and in a light absorption layer, and having high conversion efficiency; and a method for manufacturing the compound solar cell.SOLUTION: A compound solar cell includes at least a group I-III-VI compound semiconductor layer, a buffer layer and a surface electrode layer comprising a transparent conductive layer, in the order, on a substrate 1. The steam permeability of the transparent conductive layer is set to 5.0×10g/mday or less. The compound solar cell can be obtained, for example, by forming a transparent conductive layer by a specific counter sputtering method where two cathode targets 6, 6' are disposed to face each other on both sides of a virtual central axis X perpendicularly extending to the layer formation surface of the substrate 1.

Description

The present invention has a high light conversion efficiency (hereinafter referred to as “conversion efficiency”), CuInSe ₂ (CIS) composed of elements of Ib group, IIIb group and VIb group, or Cu (In , Ga) Se ₂ (CIGS) compound semiconductor (I-III-VI group compound semiconductor) The present invention relates to a method for efficiently producing a compound solar cell using a light absorption layer.

  Among solar cells, a compound solar cell using a CIS or CIGS (hereinafter collectively referred to as “CIGS-based”) compound semiconductor for a light absorption layer has high conversion efficiency and can be formed into a thin film, and conversion efficiency by light irradiation or the like. It is known that it has the advantage of less degradation.

  As a transparent conductive layer used as a surface electrode layer of such a CIGS-based compound solar cell, aluminum (Al) -added ZnO (AZO) or gallium (Ga) formed by a sputtering method (also referred to as “sputtering method”) is generally used. ) Added ZnO (GZO) and boron (B) added ZnO formed by the MOVOV method are used (for example, refer to pages 16 to 17 of Non-Patent Document 1).

Tokio Nakata “CIGS Solar Cell Fundamental Technology” Nikkan Kogyo Shimbun, September 30, 2010, first edition issued

  However, such a transparent conductive layer has a problem that the resistivity is greatly increased when used at high temperatures.

  In addition, when a thin film layer is formed under vacuum using a magnetron sputtering apparatus that is widely used as a sputtering apparatus, negative ions and high energy electrons in plasma generated by sputtering damage the thin film layer and its underlying layer, Recently, it has been found that the characteristics of solar cells deteriorate, which is a problem. This is considered to be a cause of the increase in resistivity of the transparent conductive layer, and the improvement is strongly desired.

  The present invention was made in order to solve the above-mentioned problems, and a compound solar cell having a high conversion efficiency without damage in the transparent conductive layer or in the underlying buffer layer or light absorption layer, and its production method The purpose is to provide

In order to achieve the above object, the present invention comprises, on a substrate, at least an I-III-VI group compound semiconductor layer, a buffer layer, and a surface electrode layer composed of a transparent conductive layer in this order. A first aspect is a compound solar cell in which the water vapor permeability of the layer is set to 5.0 × 10 ⁻² g / m ² · day or less.

  In addition, the present invention provides a compound according to the first aspect of the present invention, which comprises at least an I-III-VI group compound semiconductor layer, a buffer layer, and a surface electrode layer composed of a transparent conductive layer in this order on a substrate. A method for manufacturing a solar cell, wherein two sputtering cathode targets are arranged opposite to each other on both sides of a virtual central axis X extending perpendicularly to the layer forming surface of the substrate, and a radio frequency (RF) The second gist of the present invention is a compound solar cell manufacturing method in which the surface electrode layer is formed by sputtering using a power source or a radio frequency (RF) power source superimposed with a direct current (DC) power source.

The compound solar battery of the present invention has a special physical property that the transparent conductive layer constituting the surface electrode layer has a water vapor permeability of 5.0 × 10 ⁻² g / m ² · day or less, and has a very high density. Highly dense crystal structure. For this reason, the resistance value of the said transparent conductive layer is stably maintained low also under high temperature, and can achieve high conversion efficiency stably as a solar cell.

  Moreover, according to the manufacturing method of the compound solar cell of this invention, the surface electrode layer which consists of the said transparent conductive layer is for two sheets of sputtering on both sides of the virtual central axis X extended perpendicularly | vertically with respect to the layer formation surface of a board | substrate. Generated by using a sputtering method in which a cathode target is placed facing each other and applied using a radio frequency (RF) power source or a radio frequency (RF) power source superimposed with a direct current (DC) power source. Etc. can be confined between the cathode targets, and the damage to the transparent conductive layer itself and the underlying buffer layer and compound semiconductor layer (light absorption layer) can be reduced. A transparent conductive layer having a very low and dense crystal structure can be easily obtained.

  Note that, in the method for producing the compound solar cell of the present invention, not only the formation of the surface electrode layer, but also the formation of the buffer layer below it, a special sputtering method similar to the case of forming the surface electrode layer is used. In this case, both the buffer layer and the surface electrode layer have a dense crystal structure, and damage is also reduced in layers below the buffer layer, so that the conversion efficiency as a compound solar cell can be further increased.

  When the two opposing cathode targets are arranged in a substantially V shape that expands toward the substrate side, it is preferable that the crystal structure of the formed layer becomes denser.

  Among them, in particular, when each of the cathode targets is set to 5 to 15 ° with respect to the virtual central axis X extending perpendicularly to the layer forming surface of the substrate, The crystal structure is particularly dense and suitable.

It is sectional drawing of the CIGS solar cell which is one embodiment of this invention. It is explanatory drawing of the positional relationship of the cathode target and a board | substrate in the sputtering device used for the manufacturing method of the said CIGS solar cell. It is explanatory drawing which shows the other example of the positional relationship of the cathode target and a board | substrate in the said sputtering device. It is explanatory drawing which shows the other example of the positional relationship of the cathode target and a board | substrate in the said sputtering device.

  Next, an embodiment for carrying out the present invention will be described.

  FIG. 1 is a cross-sectional view of a CIGS solar cell according to an embodiment of the present invention. This CIGS solar cell includes a substrate 1, a back electrode layer 2, a CIGS light absorption layer (compound semiconductor layer) 3, a buffer layer 4, and a surface electrode layer 5 in this order. 5 has a special configuration in which the water vapor permeability of the transparent conductive layer 5 is kept very low, as will be described later. Hereinafter, each layer will be described in detail.

  As the substrate 1, an appropriate substrate is selected from a glass substrate, a metal substrate, a resin substrate, and the like according to the purpose and design requirements. Examples of the glass substrate include low alkali glass (high saddle point glass) having a very low alkali metal element content, alkali-free glass not containing an alkali metal element, and blue plate glass. Among them, it is preferable to use low alkali glass or non-alkali glass that can accurately control the amount of alkali metal element added and can control the alkali metal element not to diffuse from the glass substrate to the light absorption layer.

  Moreover, when manufacturing this CIGS solar cell by a roll-to-roll system or a stepping roll system, it is suitable that the said board | substrate 1 is elongate and has flexibility. The “long shape” means that the length in the length direction is 10 times or more of the length in the width direction, and more preferably 30 times or more.

  And it is preferable that the thickness of the said board | substrate 1 exists in the range of 5-200 micrometers, More preferably, it is the range of 10-100 micrometers. That is, if the thickness is too thick, the flexibility of the CIGS solar cell is lost, the stress applied when the CIGS solar cell is bent may increase, and the laminated structure such as the CIGS light absorption layer 3 may be damaged. This is because if the substrate is too thin, the substrate 1 is buckled when the CIGS solar cell is manufactured, and the product defect rate of the CIGS solar cell tends to increase.

  Examples of the material for forming the back electrode layer 2 formed on the substrate 1 include molybdenum (Mo), tungsten (W), chromium (Cr), and titanium (Ti). These materials are used to form a single layer or multiple layers by a method such as sputtering, vapor deposition, or inkjet.

  The thickness of the back electrode layer 2 (in the case of multiple layers, the total thickness of each layer) is preferably in the range of 10 to 1000 μm. However, when the substrate 1 is conductive and has the function of the back electrode layer 2, the back electrode layer 2 may not be provided.

  Moreover, since the performance of the CIGS compound solar cell is adversely affected when impurities derived from the substrate 1 are thermally diffused, a barrier layer (not shown) is provided on the substrate 1 or the back electrode layer 2 for the purpose of preventing this. It may be provided. Examples of the material for forming such a barrier layer include Cr, nickel (Ni), NiCr, cobalt (Co), and the like.

  The CIGS light absorption layer 3 formed on the back electrode layer 2 is a compound semiconductor having a chalcopyrite crystal structure composed of four elements of copper (Cu), indium (In), gallium (Ga), and selenium (Se). It is formed with. And it is preferable that the thickness exists in the range of 1.0-3.0 micrometers, and it is more preferable that it exists in the range of 1.5-2.5 micrometers. If the thickness is too thin, the amount of light absorption when used as the light absorption layer is reduced, and the performance of the solar cell tends to be reduced. Conversely, if it is too thick, the time taken to form the CIGS light absorption layer 3 This is because there is a tendency for productivity to be inferior.

  Moreover, it is preferable that the composition ratio of Cu, In, and Ga in the CIGS light absorption layer 3 satisfies the formula 0.7 <Cu / (Ga + In) <0.95 (molar ratio). If this equation is satisfied, it is possible to further prevent Cu (2-x) Se from being excessively taken into the CIGS light absorption layer 3 and to make the entire layer slightly deficient in Cu. Because. Moreover, it is preferable that the ratio of Ga and In which are the same element is in the range of 0.10 <Ga / (Ga + In) <0.40 (molar ratio).

The buffer layer 4 formed on the CIGS light absorption layer 3 is formed of a mixed crystal of Group IIa metal and Zn oxide. And it is preferable that it is a high resistance n-type semiconductor so that it can make a pn junction with the said CIGS light absorption layer 3, and what laminated | stacked not only a single layer but a several layer may be sufficient. If the buffer layer 4 is formed by stacking a plurality of layers, the pn junction with the CIGS light absorption layer 3 can be improved. As a material for forming such a buffer layer 4, in addition to a mixed crystal of Mg and ZnO, CdS, ZnMgO, ZnCaO, ZnMgCaO, ZnMgSrO, ZnSrO, ZnO, ZnS, Zn (OH) 2, In ₂ O ₃ , In ₂ S is ₃ and mixed crystals of these Zn (O, S, OH) , Zn (O, S) and the like. Moreover, it is preferable that the thickness exists in the range of 50-200 nm.

  Next, as the surface electrode layer 5 formed on the buffer layer 4, a transparent conductive layer is used. As such a transparent conductive layer, a material that forms a thin film with high transmittance and low resistance is preferable, and besides GZO, ITO, IZO, zinc aluminum oxide (AlZnO), and the like can be given. Moreover, it is preferable that the thickness exists in the range of 50-300 nm. And it is preferable that the light transmittance of this transparent conductive layer exceeds 80%.

The transparent conductive layer of the surface electrode layer 5 preferably has a dense structure as much as possible, and its water vapor transmission rate (WVTR) is 5.0 × 10 ⁻² g / m ² · day or less. Is set to This low value of water vapor permeability means that there are almost no defects or voids in the transparent conductive layer, and moisture and oxygen enter the layer through these gaps, resulting in deterioration of the layer. Is unlikely to occur. Therefore, the transparent conductive layer has an excellent characteristic that the resistance value is not affected by the environment and is kept low for a long period of time.

  According to the CIGS solar cell having the above configuration, as described above, the water vapor permeability of the transparent conductive layer constituting the surface electrode layer 5 is set to a very low value, and there are almost no defects or voids in the layer. Since it has a dense crystal structure, it does not easily deteriorate and can maintain a low resistance value over a long period of time. Therefore, it is excellent in wet heat resistance and can maintain good conversion efficiency over a long period of time.

  The CIGS solar cell can be manufactured, for example, as follows. That is, first, a long substrate 1 is prepared, a back electrode layer 2 is formed on the surface by a roll-to-roll method, and a CIGS light absorbing layer 3 is formed thereon by the roll-to-roll method, After cutting this to a predetermined size, the buffer layer 4 and the surface electrode layer 5 can be laminated on the CIGS light absorption layer 3 in this order. Hereafter, this manufacturing method is demonstrated in detail for every formation process of each layer.

[Step of forming back electrode layer 2]
The back electrode layer 2 is formed on the surface of the long substrate 1 by a roll-to-roll method using a forming material such as Mo by a sputtering method, a vapor deposition method, an inkjet method, or the like.

[Formation process of CIGS light absorption layer 3]
Next, the CIGS light absorption layer 3 is formed on the back electrode layer 2 while the substrate 1 on which the back electrode layer 2 is formed is also run by the roll-to-roll method. Examples of the method for forming the CIGS light absorption layer 3 include vacuum deposition, selenization / sulfurization, and sputtering. In addition, when using the board | substrate 1 which does not contain an alkali metal element, an alkali metal element can be supplied by vapor deposition before the film formation of this CIGS light absorption layer 3, or during film formation.

[Formation process of buffer layer 4]
Next, the back electrode layer 2 and the CIGS light absorption layer 3 are formed by the roll-to-roll method, and the substrate 1 wound up in a roll shape is unwound again, using a predetermined cutting device. A predetermined length of the laminate (substrate 1 + back electrode layer 2 + CIGS light absorption layer 3 laminate) is obtained by cutting each predetermined length. Then, a buffer layer 4 is formed on the CIGS light absorption layer 3 of the cut laminate by using a forming material such as a mixed crystal of Mg and ZnO by a solution growth method, a vacuum evaporation method, a sputtering method, or the like.

  For the formation of the buffer layer 4, a sputtering method is particularly suitable. Among them, unlike a general magnetron sputtering method, two cathode targets are arranged facing each other and arranged perpendicular to the target surface. It is preferable to use a counter target sputtering method in which a magnetic field is applied from one target to the other target. That is, according to the facing target sputtering method, the plasma can be confined between the facing targets, and the plasma is not exposed to the substrate 1, so that damage caused by charged particles (electrons, ions) and recoil argon is small, and film formation is performed. This is because defects and voids are less likely to occur on the surface and the underlying surface. The application to the pair of targets is preferably performed by a high frequency (RF) power source or by superimposing a direct current (DC) power source on a high frequency (RF) power source.

  In the facing target sputtering method, as shown in FIG. 2, a virtual central axis X extending vertically from the layer forming surface of the substrate 1 is assumed, and these two targets are placed on both sides of the virtual central axis X. 6 and 6 ′ are opposed to each other, and when both 6 and 6 ′ are arranged so as to be substantially V-shaped spreading toward the layer forming surface side of the substrate 1 (hereinafter referred to as “substrate 1 side”), Since a film can be formed with a small amount of electric power, the film formation surface and the surface below the film formation surface are less susceptible to damage, which is preferable. In particular, it is particularly preferable that the angle θ of at least one of the targets 6 and 6 ′ with respect to the virtual center axis X is set in the range of 5 to 15 °. In FIG. 2, the back electrode layer 2 and the CIGS light absorption layer 3 formed on the substrate 1 are not shown.

[Formation process of surface electrode layer 5]
Next, a surface electrode layer is formed on the buffer layer 4 by using a forming material such as GZO by sputtering (DC, RF, RF superposition), vapor deposition, metal organic chemical vapor deposition (MOCVD), or the like. A transparent conductive layer to be 5 is formed.

  Also in the formation of the transparent conductive layer, the counter target sputtering method, particularly the counter target sputtering method in which the targets 6 and 6 ′ are set in a special arrangement as shown in FIG. It is preferable to use it.

That is, as already described, according to the facing target sputtering method, defects and voids are unlikely to occur on the film formation surface and the underlying surface, and as a result, the crystal structure of the layer is dense and an extra gap is formed. This is because it is difficult to achieve the characteristics of this layer with a water vapor transmission rate (WVTR) of 5.0 × 10 ⁻² g / m ² · day or less.

  The application to the target is also preferably performed by using a high frequency (RF) power source or by superimposing a DC power source on a high frequency (RF) power source, rather than a direct current (DC) power source, because the WVTR can be further reduced. is there. In addition, reducing the applied power or reducing the film forming pressure is also effective in reducing the WVTR.

  In this way, the CIGS solar cell of the present invention can be obtained.

  In the above manufacturing method, when the substrate 1 has the function of the back electrode layer 2 (when it has conductivity, etc.), the above [step of forming the back electrode layer 2] is unnecessary, and the substrate 1 is left as it is. It can be used as a back electrode layer.

  As described above, when the metal diffusion prevention layer is formed between the substrate 1 and the CIGS light absorption layer 3 (on the substrate 1 or the back electrode layer 2), the substrate 1 Or after forming CIGS light absorption layer 3, using metal diffusion prevention layer forming material such as Cr, metal diffusion by sputtering method, vapor deposition method, CVD method, sol-gel method, liquid phase precipitation method, etc. A prevention layer is formed.

And in said manufacturing method, although it is said that it is suitable to use a counter target sputtering method in the process of both formation of the buffer layer 4 and formation of the surface electrode layer 5, it is transparent conductive used as the surface electrode layer 5. In order for the water vapor permeability of the layer to be 5.0 × 10 ⁻² g / m ² · day or less, it is desirable to use the counter target sputtering method at least in the step of forming the surface electrode layer 5. That is, in the step of forming the surface electrode layer 5, if the water vapor permeability of the transparent conductive layer can be set sufficiently low, the counter target sputtering method is not necessarily used for the buffer layer 4. However, when the facing target sputtering method is used for both the layers 4 and 5, not only the damage to the both layers 4 and 5 is greatly reduced, but also the CIGS light absorption layer 3 underneath is protected from adverse effects. A higher quality CIGS solar cell can be obtained, which is preferable.

  Further, in the facing target sputtering method suitably used in the above manufacturing method, depending on the composition of the cathode targets 6 and 6 ′ and the sputtering conditions, the cathode targets 6 and 6 ′ are substantially V-shaped spreading toward the substrate 1 side. As shown in FIG. 3, both cathode targets 6 and 6 ′ may be arranged in parallel. Further, as shown in FIG. 4, only one of the cathode targets (in this example, the cathode target 6) may be arranged so as to be inclined with respect to the virtual central axis X by an angle θ.

  Further, in the above manufacturing method, the back electrode layer 2 and the CIGS light absorption layer 3 are formed by the roll-to-roll method, but it is not always necessary to adopt the roll-to-roll method. It is possible to use a method of preparing and arranging by type and sequentially forming each layer thereon.

  Next, examples will be described together with comparative examples. However, the present invention is not limited to the following examples.

[Example 1]
(Formation of back electrode layer)
First, a discharge gas is formed on the surface of a substrate made of degreased soda lime glass (thickness 0.55 mm, width 20 mm × length 20 mm in plan view) using a magnetron sputtering apparatus (SH-450, manufactured by ULVAC). In this case, argon (Ar) was used, a direct current (DC) power source was used so that the sputtering pressure was 1 Pa, and a back electrode layer made of Mo having a thickness of 0.8 μm was formed under a sputtering rate of 60 nm / min.

(Formation of CIGS light absorption layer)
Next, a CIGS light absorption layer was formed on the back electrode layer. That is, each of Ga, In, Cu, and Se is disposed as a deposition source in a chamber of a vacuum deposition apparatus, the degree of vacuum is set to 1 × 10 ⁻⁴ Pa, and the substrate is heated at a rate of 550 ° C./h. And heated to 550 ° C. At this time, the back electrode layer is heated by heating the deposition sources to Ga: 950 ° C., In: 780 ° C., Cu: 1100 ° C., and Se: 140 ° C. A CIGS light absorption layer was formed on the substrate. The composition (number of atoms) of the obtained CIGS light absorption layer was Cu / III group = 0.89, Ca / III group = 0.31, and its thickness was 2.1 μm.

(Formation of buffer layer)
Next, on the CIGS light absorption layer, using a counter target sputtering apparatus in which the pair of cathode targets shown in FIG. 2 is arranged in a substantially V shape (the angles θ with respect to the virtual central axis X are 10 °, respectively) A buffer layer was formed. A cathode target having a composition of Zn _0.85 Mg _0.15 O is used, Ar is used as a discharge gas during sputtering, a power density of 0.7 kW / cm ² and a sputtering pressure of 0 with a radio frequency (RF) power source. A buffer layer having a thickness of 70 nm was obtained by adjusting the power and the formation time under the condition of 3 Pa.

(Formation of surface electrode layer)
Next, similarly to the formation of the buffer layer, the above-described counter target sputtering apparatus in which the pair of cathode targets shown in FIG. 2 are arranged in a substantially V shape (the angle θ with respect to the virtual center axis X is 10 °) is used. A surface electrode layer was formed on the buffer layer. The cathode target is composed of GZO (Ga ₂ O ₃ content: 5.7% by weight), Ar is used as a discharge gas during sputtering, and power density is obtained by a radio frequency (RF) power source. A surface electrode layer (transparent conductive layer made of a GZO film) having a film thickness of 100 nm was obtained by adjusting the power and formation time under the conditions of 1.6 W / cm ² and sputtering pressure of 0.3 Pa.

[Examples 2 to 9]
The production conditions were changed as shown in Tables 1 to 3 below. Other than that was carried out similarly to Example 1, and obtained the CIGS solar cell. However, the buffer layer in Example 9 is formed using a general-purpose magnetron sputtering apparatus (manufactured by ULVAC, SH-450: one cathode target is arranged so as to be parallel to the layer forming surface of the substrate). went.

[Comparative Example 1]
The surface electrode layer was formed using the same magnetron sputtering apparatus as used in Example 9 above. The applied power source at this time was a radio frequency (RF) power source, and the power density was 2.0 W / cm ² . Other than that was carried out similarly to Example 1, and obtained the CIGS solar cell.

[Comparative Example 2]
In the formation of the surface electrode layer, the applied power source was a direct current (DC) power source. Other than that was carried out similarly to the comparative example 1, and obtained the CIGS solar cell.

The initial conversion efficiency (η ₀ ) of the example product and the comparative product thus obtained, the conversion efficiency (η ₁ ) after being left in an atmosphere of 85 ° C. × 85% RH for 100 hours, and the conversion efficiency The loss rate is measured and calculated as described below, and is shown in Tables 1 to 4 below.

<Initial conversion efficiency of solar cell>
Twenty CIGS solar cells were prepared, each was irradiated with simulated sunlight (AM1.5), and each conversion efficiency was measured using an IV measurement system (manufactured by Yamashita Denso Co., Ltd.). The average (η ₀ ) of the measured conversion efficiencies was defined as “initial conversion efficiency”.

<Loss rate of conversion efficiency of solar cells>
Each CIGS solar cell whose initial conversion efficiency was measured was put into a constant temperature and humidity chamber of 85 ° C. × 85% RH, the conversion efficiency after being left for 100 hours was measured, and the average (η ₁ ) was obtained. The ratio of loss after input to the conversion efficiency of: [(η ₀ −η ₁ ) / (η ₀ )] × 100 (%) was calculated.

  In addition, a transparent conductive layer (thickness 100 nm) corresponding to the surface electrode layer was formed on a polyethylene terephthalate film (PET film, thickness 125 μm) by the same method as the formation of the surface electrode layer in the example product and the comparative product. Formed. And the water-vapor permeability (WVTR) of the surface electrode layer (transparent conductive layer) on this laminated film and the rate of change of surface resistance were measured as described later. In addition, since the water vapor transmission rate of the said PET film is 5 digits or more larger than that of the said transparent conductive layer, the influence on the water vapor transmission rate of a transparent conductive layer can be disregarded. These results are also shown in Tables 1 to 4 below.

<Measurement of water vapor transmission rate>
The amount of water vapor (unit: g / m ² · day) permeating through each laminated film was measured. However, “day” means 24 hours.

<Change rate of surface resistance>
First, after measuring the surface resistance value (R0) of the surface electrode layer on each laminated film with a resistance meter (Loresta MCP-P600, manufactured by Mitsubishi Yuka Co., Ltd.), the temperature is controlled in a constant temperature and humidity chamber of 85 ° C. × 85% RH. Then, the resistance value (R1) after being left for 100 hours was measured with the same resistance meter as described above, and [R1 / R0] was calculated. The larger this value, the higher the resistance value at high temperature and high humidity, indicating that the conductivity is impaired.

From the above results, it can be seen that most of the examples have an initial conversion efficiency exceeding 6%, and they have excellent conversion efficiency. In addition, in most of the examples, the loss rate of conversion efficiency is 10% or less, and even when exposed to high temperature and high humidity, the decrease in conversion efficiency is small and good quality is generally maintained. . The product of Example 9 is a buffer layer formed by a magnetron sputtering method, and the performance is slightly inferior to that of other Example products in which the buffer layer is formed by a counter target sputtering method. It can be seen that the conversion efficiency drop is small and superior compared to the conventional products. This means that the surface electrode layer of each of the examples has a dense crystal structure that is very difficult to permeate water vapor with a water vapor permeability of 5.0 × 10 ⁻² g / m ² · day or less. This is probably due to the fact that

In particular, when the water vapor permeability of the surface electrode layer is 1.1 × 10 ⁻² g / m ² · day or less (Examples 1, 2, and 4), the conversion efficiency loss rate is about 2%. Therefore, it can be seen that it is more preferable to set the water vapor permeability of the surface electrode layer in the range of 1.1 × 10 ⁻² g / m ² · day or less.

  In addition, although the said Examples 1-9 goods are CIGS solar cells, when manufacturing the CIS solar cell by the same procedure and performing evaluation similar to the above, also about these, similarly to the said CIGS solar cell, It was found that high conversion efficiency and low loss rate can be obtained.

  Since the compound solar cell of the present invention has high conversion efficiency and low conversion efficiency loss under high temperature and high humidity, it can be used not only under general conditions but also under severe conditions. It can be used well.

1 Substrate 6, 6 'cathode target X Virtual central axis

Claims (5)

On the substrate, at least an I-III-VI group compound semiconductor layer, a buffer layer, and a surface electrode layer made of a transparent conductive layer are provided in this order, and the water vapor permeability of the transparent conductive layer is 5.0 × 10 5. A compound solar cell, which is set to ⁻² g / m ² · day or less.   2. The method for producing a compound solar cell according to claim 1, further comprising a surface electrode layer comprising at least an I-III-VI group compound semiconductor layer, a buffer layer, and a transparent conductive layer in this order on the substrate. Two cathode targets for sputtering are arranged on both sides of a virtual central axis X extending perpendicularly to the layer forming surface of the substrate, and a radio frequency (RF) power source or a direct current (DC) power source is provided. A method for producing a compound solar cell, wherein the surface electrode layer is formed by a sputtering method applied using a superimposed radio frequency (RF) power source.   In the same manner as the formation of the surface electrode layer, the buffer layer is formed by facing two sputtering cathode targets on both sides of the virtual central axis X extending perpendicular to the layer formation surface of the substrate. 3. The method for producing a compound solar cell according to claim 2, wherein the compound solar cell is formed by sputtering using a high frequency (RF) power source or a high frequency (RF) power source superimposed with a direct current (DC) power source.   The method for producing a compound solar cell according to claim 2 or 3, wherein the two opposing cathode targets are arranged in a substantially V-shape extending toward the substrate side.   The method for producing a compound solar cell according to claim 4, wherein an angle between each of the cathode targets and a virtual central axis X extending perpendicularly to the layer forming surface of the substrate is set to 5 to 15 °.

Images