JP4215694B2 - Photoelectric conversion device and manufacturing method thereof - Google Patents

Description

  The present invention relates to a photoelectric conversion device, and more particularly to a technology for improving the efficiency of a photoelectric conversion device.

  With the expansion of the solar cell market, thin-film silicon solar cells are attracting attention as next-generation solar cell technologies for reducing costs. A thin-film silicon solar cell is formed by depositing a silicon thin film having a film thickness of about several μm on a glass substrate, a stainless steel substrate, or the like using a plasma CVD method or the like. Therefore, not only the amount of silicon used can be reduced, but also a large area solar cell can be formed by a single film formation, so that the cost can be reduced. In addition, since the thin-film silicon semiconductor layer can be formed by a low-temperature process of 500 ° C. or lower, the energy input during manufacturing can be greatly reduced when compared with a bulk crystal solar cell that requires a temperature of 1500 ° C. or higher. .

  As a structure of a general thin film silicon solar cell, there are two types, a super straight type and a substrate type. The super straight type structure is a structure in which a transparent conductive layer, a photoelectric conversion layer, and an electrode layer are laminated in this order on a translucent substrate, and the translucent substrate side serves as a light incident surface. On the other hand, the substrate type structure is a structure in which an electrode layer, a photoelectric conversion layer, a transparent conductive layer, and a grid electrode are stacked in this order on a substrate, and the grid electrode side serves as a light incident surface. In any structure, the photoelectric conversion layer includes a semiconductor layer exhibiting p conductivity type (hereinafter referred to as a p layer), an intrinsic semiconductor layer (hereinafter referred to as i layer), and a semiconductor layer exhibiting an n conductivity type (hereinafter referred to as p layer). It is often provided with a pin junction composed of n layers).

  As described above, the thin-film silicon solar cell has the advantage of being able to reduce the cost due to the simultaneous formation of a large area and the effect of reducing the input energy, but it has actually reached the market expansion. Not. The main factor is that the photoelectric conversion efficiency is lower than that of bulk silicon crystal solar cells which are currently mainstream. Therefore, high efficiency is an important issue for full-scale spread of thin film silicon solar cells.

  As one of the effective means for solving the above-mentioned problem, there is a wide band gap of a semiconductor layer (hereinafter referred to as a window layer) on the light incident surface side of the solar cell. This is because, in the above thin film silicon solar cell, by widening the band gap of the window layer, the effect of increasing the short circuit current density by reducing the light absorption loss in the window layer and increasing the open circuit voltage by increasing the diffusion potential is obtained. This can improve the photoelectric conversion efficiency.

As a conventional technique related to the wide band gap of the window layer as described above, there is a thin film photoelectric conversion device described in Japanese Patent Laid-Open No. 2002-16271 (Patent Document 1). The technology includes a p-type conductive p-type semiconductor layer, a substantially intrinsic i-type semiconductor layer, and a photoelectric conversion layer including at least one pin junction structure formed by laminating an n-type conductive n-type semiconductor layer; In a thin film photoelectric conversion device having a conductive and light transmissive first electrode provided on the light incident side of the photoelectric conversion layer and a second electrode provided on a surface facing the first electrode, The i-type semiconductor layer constituting the junction is made of microcrystalline silicon or microcrystalline silicon germanium, and at least one of the p-type semiconductor layer and the n-type semiconductor layer in contact therewith is a mixed crystal of microcrystalline silicon carbide and microcrystalline silicon. Consists of. According to this thin film photoelectric conversion device, the light absorption of the p-type semiconductor layer and the n-type semiconductor layer is reduced, and the band gap of the interface is increased, so that interface recombination is reduced and high photoelectric conversion efficiency is obtained. It is said. Further, in Patent Document 1, it is desirable that the carbon content of the mixed crystal of the microcrystalline silicon carbide and the microcrystalline silicon is in the range of 10 to 30% in terms of atomic ratio, and the atomic ratio is 10%. If it is less than%, the light absorption rate is not sufficiently lowered and there is no effect of mixing microcrystalline silicon carbide. Conversely, if the amount of carbon exceeds 30%, the resistance increases and a sufficient saturation current density cannot be obtained. It is described as not practical.
JP 2002-16271 A

However, in the technique of Patent Document 1, when carbon is used as an additive element for wide band gap, a large amount of carbon of 10 atomic% or more is required to be contained in the silicon film. As the carbon concentration in the film increases, the activation efficiency when the conductivity determining element doped in the p-type semiconductor layer is activated as an acceptor in the film is known to decrease. There is a problem that the carrier concentration is lowered due to the addition of impurities. Furthermore, the smaller the impurity element concentration in the film, the easier it is to crystallize, and there is also a problem that the ratio of the crystal phase in the p-type semiconductor layer decreases under the high concentration conditions.
Therefore, when an additive element for wide band gap is added at a high concentration, if a sufficient carrier concentration and crystal phase ratio can be maintained, further increase in efficiency of the photoelectric conversion device can be expected.

  The present invention has been made in view of such circumstances, and realizes an appropriate carrier concentration and ratio of crystal phases even when an impurity element is contained at a high concentration in the p-type semiconductor of the window layer on the light incident side. Thus, an open circuit voltage and a short circuit current are increased in a wide impurity element concentration range, thereby providing a photoelectric conversion device with improved photoelectric conversion efficiency.

According to the present invention, a pin-type photoelectric conversion layer configured by stacking a p-type semiconductor layer, an i-type semiconductor layer, and an n-type semiconductor layer is provided, and the p-type semiconductor layer and the i-type semiconductor layer are made of a microcrystalline semiconductor. And the p-type semiconductor layer contains silicon atoms, carbon atoms, and boron atoms, and the carbon atom concentration A atom% and the boron atom concentration B atom% with respect to the silicon atoms are 0.25−0.001A + 0.015A ² ≦ B ≦ A photoelectric conversion device represented by 0.35−0.01A + 0.022A ² is provided.

  Here, the carbon atom concentration with respect to silicon atoms is more preferably 1 atomic% or more and 40 atomic% or less. Furthermore, it is more preferable that the carbon atom concentration with respect to silicon atoms is 20 atom% or more and 40 atom% or less.

  Here, the p-type semiconductor layer containing carbon atoms and boron atoms preferably has a crystal volume fraction of Ic / Ia ≧ 3.

  The present inventors have found that the p-type semiconductor layer in the pin-type photoelectric conversion layer contains carbon atoms and boron atoms, and the carbon atom concentration A atom% and boron atom concentration B atom% contained therein are represented by the above formula. It has been found that by having the correlation, a higher photoelectric conversion efficiency can be obtained than in the case of not having the correlation, and the present invention has been completed.

According to another aspect of the present invention, the method includes forming a photoelectric conversion layer including a p-type semiconductor layer, an i-type semiconductor layer, and an n-type semiconductor layer, and forming the p-type semiconductor layer of the photoelectric conversion layer. In the process, using the source gas containing silicon atoms, boron atoms and carbon atoms, the p-type semiconductor layer has a carbon atom concentration A atom% and a boron atom concentration B atom% of 0% relative to the silicon atoms contained therein. Provided is a method for manufacturing a photoelectric conversion device formed so as to be included at a ratio represented by .25−0.001A + 0.015A ² ≦ B ≦ 0.35−0.01A + 0.022A ² .

Here, the raw material gas for forming one p-type semiconductor layer containing carbon atoms further contains H ₂ , and the gas flow rate ratio of [H ₂ ] / [SiH ₄ ] is controlled in the range of 150 to 900. It is more preferable.

  According to the photoelectric conversion device of the present invention, the open circuit voltage and the short circuit current can be increased, the photoelectric conversion efficiency can be improved, and the high efficiency of the thin film silicon solar cell can be achieved.

  Moreover, according to the manufacturing method of the photoelectric conversion apparatus of this invention, the said highly efficient photoelectric conversion apparatus can be manufactured.

  One embodiment of a photoelectric conversion device of the present invention will be described in detail with reference to the drawings. However, the present invention is not limited thereby. Here, the photoelectric conversion device may have a super straight structure in which light enters from the substrate side, or may have a substrate structure in which light enters from the opposite side of the substrate, that is, the grid electrode side. Further, it may be a super straight type stacked photoelectric conversion device provided with two or more pin type photoelectric conversion layers, or may be a substrate type stacked photoelectric conversion device.

  As shown in FIG. 1, the photoelectric conversion device 100 is a super straight type, and is configured by laminating a photoelectric conversion layer 10, a transparent conductive layer 15, and an electrode 16 in this order on a substrate 11.

<Description of substrate>
The substrate 11 is manufactured by depositing a transparent conductive layer 11b on a translucent substrate 11a. As the translucent substrate 11a, a glass plate, a translucent resin plate having heat resistance such as polyimide and polyvinyl, and a laminate of them are preferably used. However, the entire photoelectric conversion device has high light transmissivity. If it can support structurally, it will not specifically limit. Further, the surface thereof may be a metal film, a transparent conductive film, an insulating film, or the like. However, when the photoelectric conversion device is applied to a substrate type structure, an opaque substrate such as stainless steel may be used instead of the transparent substrate 11a.

  The transparent conductive layer 11b is made of a transparent conductive material. For example, a single layer or a plurality of laminated layers of transparent conductive films such as ITO, tin oxide, and zinc oxide can be used. Since the transparent conductive layer 11b plays a role as an electrode, it is preferable that the electrical conductivity is high, and a layer whose electrical conductivity is improved by adding a small amount of impurities can also be used. Examples of the method for forming the transparent conductive layer 11b include known methods such as a sputtering method, a CVD method, an electron beam vapor deposition method, a sol-gel method, a spray method, and an electrodeposition method. Further, it is desirable that an uneven shape is formed on the surface of the transparent conductive layer 11b. The unevenness can scatter and refract incident light incident from the translucent substrate 11a side to extend the optical path length, so that the light confinement effect in the photoelectric conversion layer 10 is enhanced and an improvement in the short circuit current can be expected. As a method for forming irregularities on the surface of the transparent conductive layer 11b, after the transparent conductive layer 11b is once deposited on the translucent substrate 11a, the irregularities are formed by mechanical processing such as etching or sand blasting, Using a method using surface irregularities formed by crystal growth of a film material when forming a transparent conductive layer, a method utilizing the fact that regular surface irregularities are formed because the crystal growth surface is oriented, etc. Also good. In the present embodiment, as a substrate using the unevenness formed during the crystal growth of the film material, a substrate in which a tin oxide layer is deposited on a white plate glass by a CVD method (product name: Asahi-U, manufactured by Asahi Glass Co., Ltd.) Used as the substrate 11. Further, it is more preferable to deposit a zinc oxide layer on the substrate 11 by a sputtering method since the tin oxide layer can be prevented from being damaged by plasma when a photoelectric conversion layer is formed later.

<Description of photoelectric conversion layer>
In the photoelectric conversion layer 10, the main material among the constituent materials is usually silicon, and amorphous silicon, microcrystalline silicon, and the like are particularly preferably used. Here, in the present invention, the terms “amorphous silicon” and “microcrystalline silicon” include “hydrogenated amorphous silicon” and “hydrogenated microcrystalline silicon”, which are generally used in this field, respectively. In the photoelectric conversion layer 10 of the present embodiment, a p-type semiconductor layer 12, an i-type semiconductor layer 13, and an n-type semiconductor layer 14 are deposited in this order from the substrate 11 side to form a pin junction structure. The film thickness of each type of semiconductor layer is not particularly limited, but the p-type semiconductor layer 12 is 5 to 50 nm, the i-type semiconductor layer 13 is 100 to 5000 nm, and the n-type semiconductor layer 14 is 5 to 100 nm. Preferably, the p-type semiconductor layer 12 is 10 to 30 nm, and the i-type semiconductor layer 13 is 200 to 200 nm.
4000 nm, n-type semiconductor layer 14 is 10-30 nm.

The p-type semiconductor layer 12 is a layer in which carbon atoms are added to a microcrystalline semiconductor layer (for example, a microcrystalline silicon layer) doped with at least boron as a p-type conductivity determining element. Here, the microcrystalline semiconductor layer means a semiconductor layer composed of a mixed phase of a fine crystalline phase and an amorphous phase when formed at a low temperature using a non-equilibrium process such as a plasma CVD method. As the p-type conductivity determining element, impurity atoms such as boron, aluminum, and gallium can be used. In this embodiment, boron is used. The p-type semiconductor layer 12 has a carbon atom concentration A atom% and a boron atom concentration B atom% with respect to silicon atoms of 0.25−0.001A + 0.015A ² ≦ B ≦ 0.35−0.01A + 0.022A ² . The open circuit voltage is increased and a high photoelectric conversion efficiency is obtained as compared with a p-type semiconductor layer not having this correlation. Here, the carbon concentration and the boron concentration with respect to the silicon atom are values obtained by highly sensitive secondary ion mass spectrometry (SIMS) or Auger electron spectroscopy. The reason why the open circuit voltage increases by containing carbon atoms is as follows: (a) the band gap of the p-type semiconductor layer is widened to increase the diffusion potential; (b) interfacial passivation of the grain boundaries due to carbon addition and p The interface recombination may be reduced by the effect of / i layer interface passivation. The effect of the above (b) is the surface passivation effect by a silicon nitride film or an oxide film generally used in a crystalline silicon solar cell or the like, that is, the effect of suppressing the surface recombination of photogenerated carriers on the silicon substrate surface. It is considered similar.

However, when the carbon atom addition amount is increased, (1) the activation efficiency of the p-type conductivity determining element due to the carbon atom addition decreases, and (2) the ratio of the amorphous phase in the p-type semiconductor layer due to the carbon atom addition. May increase (the crystal volume fraction will decrease), and the photoelectric conversion efficiency may be reduced. Here, in the case of (1), the carbon atom concentration A atomic% and the boron atom concentration B atomic% are 0.25−0.001A + 0.015A ² ≦ B ≦ 0.35−0.01A + 0.022A ² By forming so as to have the correlation shown, sufficient carrier concentration can be realized even when the activation efficiency of the p-type conductivity determining element is low, and the characteristics of the photoelectric conversion device can be improved. it can. In the case of (2), the [H ₂ ] / [SiH ₄ ] gas flow rate ratio in the raw material gas is increased in the range of 150 to 900, and the production temperature is lowered (200 ° C. or lower). A crystal volume fraction can be realized and the characteristics of the photoelectric conversion device can be improved. The reason is that it is experimentally known that atomic hydrogen (hydrogen radical) plays an important role in forming a crystalline phase at a low temperature, but [H ₂ ] / [SiH ₄ ] gas flow rate. This is because increasing the ratio can increase the amount of hydrogen radicals on the semiconductor layer growth surface. Here, when the [H ₂ ] / [SiH ₄ ] gas flow rate ratio is 150 or less, a sufficiently high crystal volume fraction cannot be realized, and when it is 900 or more, the amount of raw material gas used is increased and safety is ensured. This increases the amount of dilution gas (N ₂ , Ar) (about 100 times that of hydrogen), which is not desirable because the cost increases.

As the p-type semiconductor layer 12, it is preferable that the crystallization rate is 3 or more because high conductivity can be obtained and the series resistance of the photoelectric conversion layer can be reduced. Here, the crystallization ratio is a peak height Ic of 520 cm ⁻¹ attributed to a silicon-silicon bond with respect to a peak height Ia of amorphous silicon of 480 cm ⁻¹ in a Raman scattering spectrum of a single p-type semiconductor layer. The ratio, that is, Ic / Ia. This is not a value representing the absolute value of the crystal volume fraction, but the value of Ic / Ia well reflects the crystal volume fraction. Therefore, in this field, it is generally used as an index indicating the ratio of crystallized components in the film. Used. If the crystallization rate of the p-type semiconductor layer 12 is 3 or more, the crystallization rate is sufficiently high as an underlayer for the i-type semiconductor layer 13 described later, and therefore the initial layer deposition is affected by the underlayer. Since a high-quality i-type semiconductor layer 13 having a high crystallization rate and a crystal component that is easy to grow can be obtained, the short-circuit current density is increased and high photoelectric conversion efficiency can be obtained, which is preferable. In addition, when the manufacturing temperature is low (200 ° C. or lower), the amorphous phase that is easily etched is selectively removed to improve the etching effect by hydrogen radicals, and as a result, a semiconductor layer having a large number of crystal phases is formed. This is preferable because the conversion characteristics can be improved.

Here, in order to realize a state in which the crystallization rate of the p-type semiconductor layer 12 is high, as described above, a high hydrogen dilution rate and low-temperature formation are effective. In this case, the p-type conductivity is determined by hydrogen. Since the inactivation of the elements (B, Ga, Al) is accompanied, the activation efficiency of the p-type conductivity determining element is lowered. The reason is that the p-type conductivity determining element acts as an acceptor by bonding with silicon in a 4-coordinate in crystalline silicon, but hydrogen is inserted between one silicon-p-type conductivity determining element bond. Then, it is known that the p-type conductivity determining element is tricoordinated and is electrically inactivated. Therefore, the photoelectric conversion characteristics can be further improved by reactivating the deactivated p-type conductivity determining element according to the manufacturing conditions used for increasing the crystallization rate of the p-type semiconductor layer 12. . The method will be described in detail below.
When the p-type conductivity-determining element is inactivated by hydrogen, it is sharp in the infrared absorption measurement spectrum in the wave number range of 1800-1950 cm ⁻¹ due to the crystalline silicon-p-type conductivity-determining element bond. It is known that an infrared absorption peak is observed. In particular, when the p-type conductivity determining element is boron, an infrared absorption peak is observed in the vicinity of about 1875 cm ⁻¹ . In such a case, when the heat treatment at 180 ° C. to 350 ° C. is performed after the p-type semiconductor layer is formed, hydrogen that has deactivated the p-type conductivity determining element is desorbed, so that the crystalline silicon-p-type conductivity is obtained. It was found that the infrared absorption peak due to the sex-determining element bond disappears and the carrier concentration increases by about one digit. Therefore, the p-type semiconductor layer 12 does not have an absorption peak in the wave number range of 1800 to 1950 cm ⁻¹ in the infrared absorption measurement spectrum (the p-type conductivity determining element is not inactivated by hydrogen). Since the concentration is high and the photoelectric conversion efficiency can be improved, heat treatment at 180 ° C. to 350 ° C. after the formation of the p-type semiconductor layer is effective. Here, sufficient desorption of hydrogen occurs at a temperature of 180 ° C. or higher, and diffusion of impurity elements at the bonding interface is suppressed by preventing the deterioration of characteristics of the photoelectric conversion device by setting the temperature to 350 ° C. or lower. Therefore, the above temperature range is desirable.

  The i-type semiconductor layer 13 is a semiconductor layer to which impurities are not added (such as microcrystal or amorphous), and is preferably a microcrystalline semiconductor layer (such as a microcrystalline silicon layer). However, a small amount of impurity elements may be included as long as it is a substantially intrinsic semiconductor. Hereinafter, silicon will be described as an example of semiconductor. As the i-type semiconductor layer 13, amorphous silicon may be used instead of microcrystalline silicon, but microcrystalline silicon is more preferable in that high photoelectric conversion efficiency can be obtained because no photodegradation occurs. In particular, in the photoelectric conversion device having the p-type semiconductor layer made of microcrystalline silicon as described above, the i-type is used from the viewpoint of relaxing the band gap discontinuity at the interface between the p-type semiconductor layer and the i-type semiconductor layer 13. The semiconductor layer 13 is preferably microcrystalline silicon. More specifically, a p-type semiconductor layer made of microcrystalline silicon and an i-type semiconductor layer 13 made of amorphous silicon form a heterojunction, whereas a p-type semiconductor layer made of microcrystalline silicon and i made of microcrystalline silicon. Since the type semiconductor layer 13 forms a homojunction, the latter can further enhance the p / i layer interface recombination suppressing effect of the second p-type semiconductor layer. Therefore, if the i-type semiconductor layer 13 is microcrystalline silicon, the recombination of photocarriers caused by the band gap discontinuity at the interface is reduced, so that the shape factor increases and high photoelectric conversion efficiency can be obtained. preferable.

  The n-type semiconductor layer 14 is a semiconductor layer doped with an n-type conductivity determining element (such as microcrystal or amorphous), and is preferably an amorphous semiconductor layer (such as an amorphous silicon layer). As the n-type conductivity determining element, impurity atoms such as phosphorus, nitrogen, and oxygen can be used.

A typical method for forming the photoelectric conversion layer 10 is a CVD method. Examples of the CVD method include atmospheric pressure CVD, low pressure CVD, plasma CVD, thermal CVD, hot wire CVD, and MOCVD method. In this embodiment, the plasma CVD method is used. The silicon-containing gas used when forming the photoelectric conversion layer 10 by the plasma CVD method is not particularly limited as long as it contains silicon atoms such as SiH ₄ and Si ₂ H _6, but generally SiH ₄ is used. There are many cases. As the diluent gas used together with the silicon-containing gas, H ₂ , Ar, He, or the like can be used, but H ₂ is particularly preferably used when amorphous silicon and microcrystalline silicon are formed. Further, when forming the p-type semiconductor layer and the n-type semiconductor layer, a doping gas is used together with the silicon-containing gas and the dilution gas, and the doping gas is particularly limited as long as it contains a target type conductivity determining element. In general, however, B ₂ H ₆ is often used when the p-type conductivity determining element is boron, and PH ₃ is often used when the n-type conductivity determining element is phosphorus.

When the photoelectric conversion layer 10 is formed by the plasma CVD method, by appropriately controlling the film formation parameters such as the substrate temperature, the pressure, the gas flow rate, and the power input to the plasma, the abundance ratio of the amorphous phase and the crystalline phase can be set. It is possible to control. Further, as the carbon-containing gas used during the formation of the p-type semiconductor layer 12 is not particularly limited as long as it contains carbon atoms, such as CH _4, using CH ₄ in this embodiment. Here, when the [CH ₄ ] / [SiH ₄ ] gas flow rate ratio is in the range of 0.125 to 10.0 and the [B ₂ H ₆ ] / [SiH ₄ ] gas flow rate ratio is in the range of 0.3 to 10.5. It is desirable because the carbon atom concentration and boron atom concentration in the film can be easily controlled.

  Further, as described above, the p-type conductivity determining element that has been deactivated by hydrogen can be reactivated by performing a heating process at 180 ° C. to 350 ° C. after the p-type semiconductor layer 12 forming process. Since the photoelectric conversion efficiency can be improved by increasing the carrier concentration of the p-type semiconductor layer 12, it is desirable to perform the heating step. If the heating step is performed after the p-type semiconductor layer 12 formation step, an effect can be obtained. However, when the i-type semiconductor layer or the n-type semiconductor layer is subsequently formed, there is a possibility of being exposed to hydrogen radicals again. More preferably, the heating step is performed after the photoelectric conversion layer forming step.

<Description of electrode and transparent conductive layer>
The electrode 16 only needs to have at least one conductive layer, and the higher the light reflectivity, the higher the conductivity. As materials that satisfy these requirements, metallic materials such as silver, aluminum, titanium, and palladium with high visible light reflectivity and alloys thereof are used. CVD, sputtering, vacuum deposition, electron beam deposition, spraying, screen printing It is formed on the photoelectric conversion layer 10 by a method or the like. Since the electrode 16 reflects the light that has not been absorbed by the photoelectric conversion layer 10 and returns it to the photoelectric conversion layer 10 again, it contributes to the improvement of the photoelectric conversion efficiency. Furthermore, when the transparent conductive layer 15 is formed between the photoelectric conversion layer 10 and the electrode 16, an effect of improving light confinement and light reflectance with respect to incident light can be obtained, and in addition, photoelectric conversion of elements contained in the electrode 16 can be achieved. Diffusion to the conversion layer 10 can be suppressed. The transparent conductive layer 15 can be formed by the same material and manufacturing method as the transparent conductive layer 11b. However, when the present invention is applied to a substrate type structure, it is desirable that the electrode 16 has a grid shape such as a comb shape that does not cover the surface uniformly.

  With the above configuration, a super straight type (or substrate type) photoelectric conversion device 100 having a large open circuit voltage, short circuit current density, and high form factor and high photoelectric conversion efficiency can be obtained.

  Examples of the present invention and comparative examples will be described below.

(Examples 1-15 and Comparative Examples 1-11)
In Examples 1 to 15 and Comparative Examples 1 to 11, the super straight photoelectric conversion device 100 shown in FIG. 1 was produced as follows.

  As the substrate 11, a white plate glass (Asahi Glass Co., Ltd., product name: Asahi-U) having a length of 127 mm × width of 127 mm × thickness of 1.8 mm in which a transparent conductive film 11 b is formed on the surface of a light-transmitting substrate 11 a was used. . A zinc oxide layer having a thickness of 50 nm is formed on the substrate 11 by a magnetron sputtering method, and then the photoelectric conversion layer 10 is formed by a plasma CVD method under the conditions described later, the p-type semiconductor layer 12, the i-type semiconductor layer 13, and the n-type semiconductor. Layer 14 was deposited in that order. The plasma CVD apparatus used for film formation of the photoelectric conversion layer 10 is an ultra-high vacuum apparatus, and a high-quality photoelectric conversion layer with little mixing of impurity elements can be manufactured. Subsequently, a 50 nm-thickness zinc oxide layer as the transparent conductive layer 15 and a 500 nm-thickness silver layer as the electrode 16 are sequentially deposited on the photoelectric conversion layer 10 by the magnetron sputtering method. 100 was obtained.

Each layer of the photoelectric conversion layer 10 was formed under the following conditions.
The p-type semiconductor layer 12 uses SiH ₄ , H ₂ , B ₂ H ₆ and CH ₄ as source gases. The H ₂ / SiH ₄ gas flow rate ratio, B ₂ H ₆ / SiH ₄ gas flow rate ratio and CH ₄ / SiH ₄ gas flow rate ratio were changed as shown in Table 1, and the carbon concentration and boron concentration in the film at that time were changed. This is also shown in Table 1. The carbon concentration and boron concentration with respect to silicon atoms in the film are values (atomic%) obtained as a result of performing highly sensitive secondary ion mass spectrometry and Auger electron spectroscopy on the p-type semiconductor layer 12. The p-type semiconductor layer 12 is desirably thin as long as it does not impair the function as the p-type layer in order to increase the amount of light incident on the i-type semiconductor layer, which is a photoactive layer. In this embodiment, the p-type semiconductor layer 12 has a thickness of 20 nm. did. The H ₂ / SiH ₄ gas flow rate ratio was adjusted so that the crystallization rate was expressed as Ic / Ia described above, and Ic / Ia ≧ 3.

The i-type semiconductor layer 13 was produced using SiH ₄ and H ₂ as source gases. The H ₂ / SiH ₄ gas flow rate ratio was set to 80 times so as to be microcrystalline silicon. Further, the film forming time was adjusted so that the film thickness was 2500 nm.

The n-type semiconductor layer 14 used SiH ₄ , H ₂ and PH ₃ as source gases. The H ₂ / SiH ₄ gas flow rate ratio was 20 times, and the PH ₃ / SiH ₄ gas flow rate ratio was adjusted so that the phosphorus concentration in the film was 0.01 atomic%. The film thickness of the n-type semiconductor layer 14 was 20 nm.

In addition, when forming each semiconductor layer 12, 13, and 14 by plasma CVD, the substrate temperature at the time of film-forming was 170 degreeC, 180 degreeC, and 160 degreeC, respectively.
Further, after the photoelectric conversion layer 10 was fabricated, heat treatment was performed at 200 ° C. for 1 hour in a nitrogen atmosphere.

With respect to the photoelectric conversion devices of Examples 1 to 15 and Comparative Examples 1 to 11 thus obtained, the current-voltage characteristics of a cell area of 1 cm ² under AM1.5 (100 mW / cm ² ) irradiation conditions were measured. . The conversion efficiencies obtained as a result are summarized in Table 1 and, in particular, the carbon concentration dependence of the appropriate boron concentration is shown in FIG.

(Consideration on Examples 1 to 15 and Comparative Examples 1 to 11)
Hereinafter, the comparison results (●) of Examples 1 to 15 and the comparison results (Δ) of Comparative Examples 1 to 11 will be considered based on Table 1 and FIG.

  Examples 1 to 15 are manufactured by adjusting the boron concentration so that the sheet resistance becomes 300 kΩ / □ or less at a film thickness of 50 nm when the carbon concentration and boron concentration in the p-type semiconductor layer are gradually increased. ing. However, the sheet resistance value after heat treatment at 200 ° C. for 1 hour in a nitrogen atmosphere. FIG. 2 shows the correlation between the carbon concentration and boron concentration in the p-type semiconductor layer at this time.

As a result of fitting to FIG. 2, assuming that the carbon atom concentration A (atomic%) and the boron atom concentration B (atomic%),
(Formula 1) B ≧ 0.25−0.001A + 0.015A ²
It has been clarified that a sheet resistance of 300 kΩ / □ or less can be realized when there is a correlation. As a result of applying the p-type semiconductor layer having such a correlation to the photoelectric conversion device, when the carbon concentration is gradually increased, the photoelectric conversion efficiency is improved to a carbon concentration of 30 atomic%, and thereafter, the carbon concentration is relatively high to 40 atomic%. It became clear that the photoelectric conversion efficiency was maintained. The reason is considered as follows. When the p-type semiconductor layer contains carbon atoms, (a) the band gap of the p-type semiconductor layer is widened to increase the diffusion potential, and (b) the interface passivation of the crystal grain boundary due to the addition of carbon and the p / i layer The open-circuit voltage is improved by the effect of reducing interface recombination due to interface passivation. Here, since the boron concentration is increased so as to compensate for the decrease in the activation efficiency of boron as the carbon atom concentration increases, the carrier concentration of the p-type semiconductor layer does not decrease even at a high carbon concentration. Therefore, it is considered that the open-circuit voltage and the photoelectric conversion efficiency are improved even when the carbon concentration is high. Next, considering a comparative example in which the boron concentration B is lower than Equation 1, it is considered that since the sheet resistance exceeds 300 kΩ / □, the diffusion potential decreases due to insufficient carrier concentration and the conversion efficiency also decreases. . Therefore, it can be said that the fact that the boron concentration B (atomic%) satisfies Equation 1 is one of the conditions for obtaining high conversion efficiency.

On the other hand, as a result of fitting to FIG.
(Formula 2) B ≦ 0.35−0.01A + 0.022A ²
It is also clear that high conversion efficiency can be obtained when there is a correlation.
That is, when the boron concentration is shifted to a higher side beyond Equation 2, the conversion efficiency is lowered although the sheet resistance is sufficiently low. The reason for this seems to be that defects increase at the p-layer or the p / i interface as the boron concentration increases. That is, it is known that a film with fewer defects can be formed as the hydrogen coverage on the growth surface is higher, but boron increases the defects because it chemically extracts hydrogen on the growth surface. In this example and the comparative example, in order to maintain the ratio of the crystal phase, when the boron concentration is increased, the [H ₂ ] / [SiH ₄ ] gas flow rate ratio is also increased at the same time. If it is higher than 2, it is considered that the defect density is not sufficiently reduced.

From the above, assuming that the carbon concentration and boron concentration in the p-type semiconductor layer are carbon atom concentration A (atomic%) and boron atomic concentration B (atomic%), 0.25-0.001A + 0.015A ² ≦ B ≦ 0.35 It has been clarified that high photoelectric conversion efficiency can be obtained in the case of having a correlation of −0.01A + 0.022A ² .

  Further, when considering the optimum value of the carbon concentration, the photoelectric conversion efficiency is maximized when the carbon concentration is 30 atomic%. Therefore, in order to obtain high conversion efficiency, the carbon atom concentration with respect to silicon atoms is 20 atomic%. More preferably, it is 40 atomic% or less.

  The photoelectric conversion device of the present invention is suitable for a super straight type or substrate type thin film solar cell.

It is a schematic sectional drawing of the super straight type photoelectric conversion apparatus of embodiment of this invention. It is a graph which shows the carbon concentration dependence of the appropriate boron concentration in an Example and a comparative example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Photoelectric converting layer 11 Substrate 11a Translucent substrate 11b Transparent conductive layer 12 P-type semiconductor layer 13 i-type semiconductor layer 14 n-type semiconductor layer 15 Transparent conductive layer 16 Electrode 100 Super straight type photoelectric conversion device

Claims (11)

A pin-type photoelectric conversion layer including a p-type semiconductor layer, an i-type semiconductor layer, and an n-type semiconductor layer is provided, the p-type semiconductor layer and the i-type semiconductor layer are made of a microcrystalline semiconductor, and the p-type semiconductor layer Contains silicon atom, carbon atom and boron atom, and the carbon atom concentration A atom% and boron atom concentration B atom% with respect to silicon atom are 0.25−0.001A + 0.015A ² ≦ B ≦ 0.35−0.00. A photoelectric conversion device represented by 01A + 0.022A ² .   2. The photoelectric conversion device according to claim 1, wherein the p-type semiconductor layer has a carbon atom concentration A of 1 atom% to 40 atom% with respect to silicon atoms.   2. The photoelectric conversion device according to claim 1, wherein the p-type semiconductor layer has a carbon atom concentration A with respect to silicon atoms of 20 atom% or more and 40 atom% or less.   The photoelectric conversion device according to claim 1, wherein the p-type semiconductor layer has a crystal volume fraction of Ic / Ia ≧ 3. 5. The photoelectric conversion device according to claim 1, wherein the p-type semiconductor layer does not have an absorption peak in a wave number range of 1800 to 1950 cm ^{−1 in} a lattice vibration spectroscopy measurement spectrum.   The photoelectric conversion device according to claim 1, wherein the i-type semiconductor layer has at least a crystalline silicon phase. a step of forming a photoelectric conversion layer including a p-type semiconductor layer, an i-type semiconductor layer, and an n-type semiconductor layer, and in the step of forming the p-type semiconductor layer of the photoelectric conversion layer, silicon atoms, boron atoms, and carbon Using the source gas containing atoms, the p-type semiconductor layer has a carbon atom concentration A atom% and a boron atom concentration B atom% of 0.25-0.001A + 0.015A ² ≦ A method of manufacturing a photoelectric conversion device, wherein the photoelectric conversion device is formed so as to be included at a ratio represented by B ≦ 0.35−0.01A + 0.022A ² . The source gas for forming the p-type semiconductor layer contains CH ₄ and B ₂ H ₆ , the [B ₂ H ₆ ] / [SiH ₄ ] gas flow ratio is in the range of 0.3 to 10.5, and The method for producing a photoelectric conversion device according to claim 7, wherein the [CH ₄ ] / [SiH ₄ ] gas flow rate ratio is controlled within a range of 0.125 to 10.0. raw material gas for forming a p-type semiconductor layer further comprises H _2, [H _2] / photoelectric conversion according to claim 7 or 8 for controlling the gas flow rate ratio of [SiH _4] in the range of 150 to 900 Device manufacturing method.   The manufacturing method of the photoelectric conversion apparatus as described in any one of Claims 7-9 which has a heating process of 180 to 350 degreeC after a p-type semiconductor layer formation process.   The manufacturing method of the photoelectric conversion apparatus as described in any one of Claims 7-10 which has a heating process of 180 to 350 degreeC after a photoelectric converting layer formation process.

Images