JP2012129569A - Multi-junction thin film solar cell and method of manufacturing the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an amorphous-crystalline junction thin film solar cell in which defect density is reduced in the crystalline semiconductor layer, and sufficient light confinement effect can be attained by a texture structure of protrusions and recesses.SOLUTION: In the thin film solar cell, the surface supporting a photoelectric conversion element structure forms a texture structure where the average value of the ratio R/L of the size R and the interval L of protrusions and recesses is in the range of 0.1-1.5. In the photoelectric conversion element structure formed on the surface, a p-type semiconductor layer 13 and an intrinsic semiconductor layer 14 in contact with the surface are crystalline layers mainly composed of columnarly grown crystal grains having a height of 100 nm or higher.

Description

  The present invention relates to a thin film solar cell having high photoelectric conversion efficiency.

  Solar cells are attracting attention as an alternative energy source for fossil fuels such as oil, where there is concern about future supply and demand, and there is a problem of carbon dioxide emissions that cause global warming.

  This solar cell uses a pn junction for a photoelectric conversion layer that converts light energy into electric power, and silicon is most commonly used as a semiconductor constituting the pn junction. From the viewpoint of photoelectric conversion efficiency, it is preferable to use single crystal silicon, but there are problems of raw material supply, large area, and low cost.

  On the other hand, thin-film solar cells using amorphous silicon as a photoelectric conversion layer as an advantageous material for realizing large area and low cost have been put into practical use, but the photoelectric conversion efficiency is lower than that of single crystal silicon solar cells. Inferior. Furthermore, since amorphous silicon has a phenomenon called the Staebler-Wronski effect in which the density of defects in the film increases as it is irradiated with light, the amorphous silicon solar cell inevitably suffers from the problem of deterioration in photoelectric conversion efficiency over time.

  Therefore, in recent years, in order to realize a solar cell that has both high and stable photoelectric conversion efficiency at the single crystal silicon solar cell level, large area and low cost at the amorphous silicon solar cell level, a crystalline silicon photoelectric conversion layer Use in is being considered. In particular, a thin film solar cell in which a crystalline silicon thin film is formed by using a thin film formation technique by chemical vapor deposition (hereinafter referred to as a CVD method) similar to the case of amorphous silicon has attracted attention.

  JP-A-1-289173 discloses a multi-junction thin film in which a photoelectric conversion element using amorphous silicon as an active layer and a photoelectric conversion element using polycrystalline silicon having an energy gap smaller than that of amorphous silicon as an active layer are stacked. A solar cell is disclosed. Such a solar cell can use solar energy more efficiently than a single-junction type by adopting a configuration in which sunlight is incident from the photoelectric conversion element side using amorphous silicon as an active layer. There are advantages. Furthermore, an amorphous silicon layer and a polycrystalline silicon layer that can obtain a high voltage because a plurality of photoelectric conversion elements are connected in series, and that the amorphous silicon layer as an active layer can be thinned, so that deterioration of photoelectric conversion efficiency over time can be suppressed. As a means to achieve both high efficiency and low cost, research and development has been actively conducted.

  However, despite vigorous research and development, the photoelectric conversion efficiency of the above-mentioned multi-junction thin-film solar cell is far from the expected value so far. A major factor is that the defect density in the crystalline silicon active layer is high. In order to reduce the defect density, it is effective to increase the crystal grain size. In solar cells that have a structure in which carriers flow in the film thickness direction, the existence of polycrystalline grains is a structure that does not have grain boundaries that cross the film thickness direction, that is, crystal grains grow in a columnar shape in the film thickness direction. Such a structure is desirable, and such a structure is easily obtained when the crystal orientation is aligned with respect to the film thickness direction.

  Various attempts have been made to obtain a polycrystalline silicon thin film having a large crystal grain size. For example, a so-called laser annealing method is known in which an amorphous silicon thin film formed on a substrate by a CVD method or the like is irradiated with a laser beam and melted, and then a polycrystalline silicon thin film is obtained by solidification. There is also known a so-called partial doping method in which crystallization by solid phase growth is selectively started by doping a part of an amorphous silicon film with phosphorus, boron, or the like. However, the above-described method has problems that are difficult to put into practical use, such as high equipment cost and heat treatment time of several tens of hours.

On the other hand, various attempts have been made to obtain a crystalline silicon thin film in which the crystal orientation is aligned with respect to the film thickness direction. For example, Japanese Patent Application Laid-Open No. 7-240531 discloses a method for manufacturing a solar cell and a multilayer thin film, in which an inert gas beam is irradiated from a plurality of directions during or after the semiconductor thin film is deposited. According to this method, a silicon thin film is obtained in which the <111> direction, which is the densest direction of crystalline silicon, is aligned with the irradiation direction of an inert gas beam serving as a channeling beam. Japanese Patent Laid-Open No. 11-278988 discloses a single crystal thin film manufacturing method in which a SiH ₃ beam is irradiated from a direction perpendicular to a substrate and a H beam is irradiated from a direction parallel to the substrate during silicon film deposition. The method is shown. According to this method, since the molecular weight is large, the <111> direction, which is the densest direction of crystalline silicon, is aligned in the direction perpendicular to the substrate surface, which is the irradiation direction of the SiH ₃ beam, which is a non-channeling beam, and the molecular weight is small. It is said that a single crystal silicon thin film can be obtained by aligning the <110> direction of crystalline silicon with the H beam irradiation direction.

  As mentioned above, multi-junction thin-film solar cells are attracting attention in addition to the expectation of higher efficiency and suppression of deterioration over time. In addition, amorphous silicon layers and crystalline silicon layers are continuously formed by CVD. This is because it can be formed. However, H.C. In Yamamoto et al, PVSEC-11, Sapporo, Japan, 1999, when microcrystalline silicon is formed by plasma CVD on a substrate having a textured structure with surface irregularities in which tin oxide is laminated on glass, as shown in FIG. In addition, it has been reported that silicon grains grow preferentially in the direction perpendicular to the individual uneven surfaces, and a large number of defects occur due to collision of crystal grains with different crystal orientations grown from different uneven surfaces. It was. Such a defect due to the crystalline semiconductor layer rather than the amorphous semiconductor layer becomes a carrier recombination center and significantly deteriorates the photoelectric conversion efficiency, and thus must be eliminated as much as possible. H. Yamamoto et al. Show that when the size of the unevenness is reduced by further laminating zinc oxide on the tin oxide having surface unevenness, the silicon crystal grains are perpendicular to the surface of the zinc oxide as in the case of tin oxide. It was also reported at the same time that crystal grains grown from different uneven surfaces collide with each other but their orientation difference is small, resulting in fewer defects. However, it is clear that the surface irregularities of the substrate should be as small as possible in order to reduce defects in the crystalline semiconductor layer.

  However, as shown in Japanese Patent No. 1681183, the transparent conductive film having surface irregularities causes irregular reflection of light and increases the optical path length in the silicon film, so that the current value generated by light absorption is reduced. There is an important function of increasing the so-called light confinement effect. Furthermore, since the light confinement effect can make the photoelectric conversion layer thinner by improving the photoelectric conversion efficiency, an advantage that the film forming process can be shortened is also produced. This results in a significant improvement in throughput in a multi-junction thin film solar cell having a crystalline silicon layer that requires a photoelectric conversion layer thickness several times that of an amorphous silicon layer due to the difference in light absorption characteristics. Therefore, it should be avoided to eliminate or reduce the surface roughness.

  However, at present, it is very difficult to achieve both the defect density reduction in the crystalline silicon thin film and the light confinement effect due to the uneven surface, and this problem has not been solved. For example, a silicon thin film having a low defect density in the film can be produced by using the method for producing a crystalline silicon thin film described in Japanese Patent Application Laid-Open No. 7-240531 or Japanese Patent Application Laid-Open No. 11-278888 described above. Since the substrate to be formed is substantially flat and cannot exhibit the light confinement effect, it is clear that there is a limit to the increase in efficiency.

  Japanese Patent Application Laid-Open No. 10-150209 discloses a photoelectric conversion characterized in that it has a relationship between the normal direction of the substrate and the longitudinal direction of the columnar crystal grains or the normal direction of the uneven surface. Although an element is disclosed, the above-described defect is inevitable as long as the normal direction of the substrate and the longitudinal direction of the columnar crystal grains have an inclination that is not small. Furthermore, Japanese Patent Application Laid-Open No. 10-117006, Japanese Patent Application Laid-Open No. 10-294481, Japanese Patent Application Laid-Open No. 11-214728, Japanese Patent Application Laid-Open No. 11-266027, and Japanese Patent Application Laid-Open No. 2000-58892 have a back surface with an uneven surface. A thin film solar cell in which a lower photoelectric conversion element having a photoelectric conversion layer made of a polycrystalline silicon layer is formed on an electrode, and the polycrystalline silicon layer has a (110) preferential crystal orientation plane parallel to the substrate surface is shown. However, in forming the polycrystalline silicon layer near the surface of the back electrode having a concavo-convex shape, the above-described growth behavior of the polycrystalline silicon is not considered at all, so it is clear that the occurrence of defects is inevitable. is there.

  An object of the present invention is to provide a highly efficient multi-junction thin film solar cell having a crystalline semiconductor layer in which an increase in defect density is suppressed while having a sufficient light confinement effect.

  According to the present invention, a multi-junction thin film solar cell is provided, and the solar cell includes a substrate and a plurality of photoelectric conversion element structures stacked on the substrate, wherein each of the plurality of photoelectric conversion element structures is provided. , The first conductive semiconductor layer, the intrinsic semiconductor layer, and the second conductive semiconductor layer are sequentially stacked. Among the stacked photoelectric conversion element structures, N (N is one or more arbitrary numbers) counted from the substrate side The surface of the second conductive semiconductor layer side of the (integer) th photoelectric conversion element structure has a texture structure with irregularities such that the ratio R / L of the irregularity size R to the irregularity interval L is in the range of 0.1 to 1.5. Furthermore, the first conductive semiconductor layer and the intrinsic semiconductor layer of the (N + 1) th photoelectric conversion element structure formed on the Nth photoelectric conversion element structure are the second of the Nth photoelectric conversion element structure. Perpendicular to the surface of the conductive semiconductor layer A crystalline semiconductor layer which is composed mainly by the crystal grains have grown in a columnar countercurrent.

  Another aspect of the present invention provides a multi-junction thin film solar cell, which includes a substrate and a plurality of photoelectric conversion element structures stacked on the substrate, wherein the plurality of photoelectric conversion elements are provided. In each of the structures, a first conductivity type semiconductor layer, an intrinsic semiconductor layer, and a second conductivity type semiconductor layer are sequentially stacked. Among the plurality of stacked photoelectric conversion element structures, N (N Is an arbitrary integer greater than or equal to 1) An intermediate layer is provided between the Nth photoelectric conversion element structure and the (N + 1) th photoelectric conversion element structure, and the surface of the intermediate layer that supports the (N + 1) th photoelectric conversion element structure Has a texture structure with unevenness such that the ratio R / L of the unevenness size R to the unevenness interval L is in the range of 0.1 to 1.5, and the first of the N + 1th photoelectric conversion element structure. Conductive semiconductor layer and intrinsic semiconductor Is a crystalline semiconductor layer which is composed mainly by the crystal grains have grown in a columnar direction perpendicular to the surface of the intermediate layer.

  In the present invention, the first conductive crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure preferably has a thickness of 1 nm or more and less than 200 nm. In addition, in the portion from the surface of the N + 1th photoelectric conversion element structure on the first conductive type semiconductor layer side to the height of 200 nm, it is perpendicular to the surface of the second conductive type semiconductor layer side of the Nth photoelectric conversion element structure or the surface of the intermediate layer. The crystal grains growing in a columnar shape in any direction occupy at least 50% or more of the entire crystal grains in the portion, and are on the surface of the second conductive semiconductor layer side or the surface of the intermediate layer of the Nth photoelectric conversion element structure. It is preferable to have a length of 100 nm or more in the vertical direction.

  In the present invention, the first conductive crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure preferably has a thickness of 1 nm to 50 nm.

  In the present invention, the first conductive crystalline semiconductor layer and the intrinsic crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure are preferably mainly made of silicon. Further, in the first conductive crystalline semiconductor layer and the intrinsic crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure, parallel to the surface of the Nth photoelectric conversion element structure on the second conductive type semiconductor layer side or the surface of the intermediate layer. It is preferable that the crystal plane to be oriented is mainly (110).

In the present invention, the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) plane X-ray diffraction peak and the integrated intensity I _{111 of the} (111) plane X-ray diffraction peak obtained for the intrinsic crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure are obtained. The ratio I ₂₂₀ / I ₁₁₁ is preferably 3 or more.

  In the present invention, it is particularly preferable that the first conductive crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure contains boron as an impurity.

  In the present invention, the intermediate layer is preferably a transparent conductor mainly made of zinc oxide.

  In this specification, unless otherwise specified, the term “crystalline semiconductor layer” includes all forms of semiconductor layers including a plurality of crystal grains having different orientations as constituent elements. Therefore, the term “crystalline semiconductor layer” means, for example, not only a semiconductor layer that is an assembly of a plurality of crystallites, but also a semiconductor layer in which a crystal component called a microcrystal or a microcrystal and an amorphous component are mixed. Including.

  Second conductive semiconductor layer side surface of Nth (N is an arbitrary integer of 1 or more) photoelectric conversion element structure or N + 1th photoelectric conversion element structure counting from the substrate side of the multi-junction thin film solar cell according to the present invention. On the surface of the intermediate layer that supports the surface, an uneven texture structure is provided. The size of the unevenness is typically in the range of 0.01 to 10 μm, preferably 0.05 to 2 μm, and the uneven shape cannot be determined visually. That is, these surfaces are recognized as smooth surfaces by visual inspection. Therefore, in this specification, “perpendicular to the surface of the Nth photoelectric conversion element structure on the second conductive semiconductor layer side or the surface of the intermediate layer” or “the surface of the Nth photoelectric conversion element structure on the second conductive type semiconductor layer side or When referred to as “parallel to the surface of the intermediate layer”, unless otherwise specified, the criteria of “perpendicular” and “parallel” are surfaces (macroscopic surfaces) that are regarded as smooth by this visual observation.

  On the other hand, when viewed microscopically, the surface supporting the photoelectric conversion element structure forms a textured structure with irregularities. This uneven shape can be measured by an atomic force microscope, and the unevenness size R and the unevenness interval L are defined as follows. That is, in an arbitrary region of the surface, line measurement of the surface irregularity shape is performed with an atomic force microscope over a length of 5 μm, and the surface roughness waveform obtained by the measurement is the surface irregularity defined in Japanese Industrial Standard JIS B0602-1994. The arithmetic average value Ra is defined as the unevenness size R, and the average interval Sm of the surface unevenness defined by the Japanese Industrial Standard JIS B0602-1994 is defined as the unevenness interval L.

  The present invention provides a multi-junction thin film solar cell having a first conductive crystalline semiconductor layer and an intrinsic crystalline semiconductor layer that are grown in a columnar shape in a direction perpendicular to the macroscopic surface even on an uneven surface. With this structure, it is possible to achieve both an increase in the amount of light absorption due to the light confinement effect and good carrier transport characteristics in the film thickness direction due to a reduction in defects in the crystalline semiconductor layer. Therefore, the multi-junction thin film solar cell according to the present invention can have high photoelectric conversion efficiency.

The schematic diagram which shows the silicon crystalline semiconductor layer grown in the column shape in the direction perpendicular | vertical to a macroscopic surface. The schematic diagram which shows the silicon crystalline semiconductor layer grown in the column shape in the direction perpendicular | vertical to each uneven | corrugated surface. The schematic diagram which shows an example of the thin film solar cell structure by this invention. 4 is a schematic diagram showing a transmission electron microscope observation image of a sample cross section in Example 1. FIG. The schematic diagram which shows the transmission electron microscope observation image of the sample cross section in the comparative example 1. FIG. The figure which shows the relationship between ratio _I220 / _I111 and R / L.

The present inventor performed plasma CVD on a substrate in which the ratio R / L of the unevenness size R and the unevenness interval L was appropriately changed by etching zinc oxide coated on a glass plate having a smooth surface. A study was made to form a silicon crystalline semiconductor layer. As a result, under the condition that the number of channeling particles is small, the orientation of the silicon crystalline semiconductor layer can be strengthened only when the average value of R / L is small, in other words, when the surface unevenness is small, as shown by the curve 61 in FIG. On the other hand, under the condition where there are many channeling particles, as shown by the curve 62 in FIG. 6, not only when the average value of R / L is small, that is, when the surface unevenness is small, R / L is 0.00. In the range of 1 to 1.5, the ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) plane X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) plane X-ray diffraction peak is It was discovered that it was possible to make it 3 or more.

  For the first time, this discovery makes it possible to achieve both high light absorption due to the light confinement effect and good carrier transport properties in the film thickness direction due to the absence of defects in the crystalline semiconductor layer, resulting in high photoelectric conversion efficiency. A multi-junction thin-film solar cell having the above structure was realized.

  As a substrate material used for the multi-junction thin film solar cell of the present invention, various materials such as glass, metal, resin having heat resistance of about 200 ° C. such as polyimide and polyvinyl, and those laminated thereon are used. it can. Moreover, the thing which coat | covered the metal film, the transparent conductive film, or the insulating film etc. on those surfaces is also contained. The thickness of the substrate is not particularly limited, but is, for example, about 0.1 to 30 mm so as to have an appropriate strength and weight so that the structure can be supported.

  The surface of the substrate may be uneven. As a means for providing unevenness, for example, a film that forms unevenness on the surface at the same time as deposition may be formed on a substrate having a smooth surface. The film on which the unevenness is formed on the surface may be the same material as the substrate or a different material. Further, the unevenness can be formed by performing mechanical processing such as sandblasting or chemical processing such as etching on the substrate surface.

  In the multi-junction thin film solar cell of the present invention, among the plurality of photoelectric conversion element structures stacked on the substrate, the second (N is an arbitrary integer greater than or equal to 1) photoelectric conversion element structure counted from the substrate side. The surface of the conductive semiconductor layer side has a textured structure with unevenness such that the ratio R / L of the unevenness size R to the unevenness interval L is in the range of 0.1 to 1.5. Further, the (N + 1) th photoelectric conversion element structure has a first conductive structure mainly composed of crystal grains growing in a columnar shape in a direction perpendicular to the macroscopic surface even on the surface having a texture as shown in FIG. A type crystalline semiconductor layer and an intrinsic crystalline semiconductor layer. With these characteristics, it is possible to obtain a multi-junction thin-film solar cell that exhibits high photoelectric conversion efficiency that exhibits a light confinement effect and suppresses an increase in defects in the photoelectric conversion element structure.

  In another embodiment of the multi-junction thin film solar cell according to the present invention, among a plurality of photoelectric conversion element structures stacked on a substrate, Nth (N is an arbitrary integer greater than or equal to 1) photoelectric conversion counted from the substrate side The surface of the intermediate layer provided between the element structure and the (N + 1) th photoelectric conversion element structure has a ratio R / L of the unevenness size R to the unevenness interval L in the range of 0.1 to 1.5. It has a textured structure with irregularities. Further, the (N + 1) th photoelectric conversion element structure has a first conductive structure mainly composed of crystal grains growing in a columnar shape in a direction perpendicular to the macroscopic surface even on the surface having a texture as shown in FIG. A type crystalline semiconductor layer and an intrinsic crystalline semiconductor layer. These characteristics cause a sufficient light confinement effect and suppress an increase in defects in the photoelectric conversion element structure. Therefore, the multi-junction thin film solar cell according to the present invention can provide high photoelectric conversion efficiency.

  In the present invention, the intermediate layer is a layer provided between layers having different conductivity types in adjacent photoelectric conversion element structures. The intermediate layer is provided in order to prevent junction formation in the reverse direction that occurs when different conductive type layers are directly bonded, or bonding failure due to mixing of impurities.

  As the intermediate layer, a transparent conductive film made of tin oxide, indium oxide, zinc oxide or the like is suitable. Further, the intermediate layer may be composed of a single material of these, or may be a laminate of a plurality of layers containing these materials. In particular, the surface on the side not facing the substrate is more preferably made of a transparent conductive film. In particular, a transparent conductive film mainly composed of zinc oxide is more preferable because it has an advantage that it is inexpensive and has high plasma resistance and is hardly deteriorated.

  These transparent conductive films can be produced by known methods such as sputtering, atmospheric pressure CVD, reduced pressure CVD, electron beam evaporation, sol-gel, and electrodeposition. Among them, the sputtering method is particularly preferable because it is easy to control the transmittance and resistivity of the transparent conductive film to those suitable for the thin film solar cell.

A small amount of impurities may be added to these transparent conductive films. For example, in the case of zinc oxide, the resistivity is reduced by containing a Group 3B element such as gallium or aluminum of about 5 × 10 ^{20 to} 5 × 10 ²¹ cm ⁻³ or a Group 1B element such as copper. , Preferred for use as an intermediate layer. In addition, if the thickness of these transparent conductive films (intermediate layers) is too thin, there will be a problem in the uniformity of characteristics, and if it is too thick, the decrease in transmittance and the increase in series resistance will cause a decrease in photoelectric conversion efficiency and an increase in cost. Since it causes, Preferably it is about 1-50 nm.

  As a means for providing unevenness on the surface of the intermediate layer, for example, the intermediate layer may be formed under conditions such that unevenness is formed on the surface simultaneously with the deposition. At this time, the formation condition of the intermediate layer may be determined considering that the uneven shape of the surface of the intermediate layer is affected by the uneven shape of the photoelectric conversion element as a base. It is also possible to form irregularities by performing mechanical processing such as sandblasting or chemical processing such as etching on the surface of the intermediate layer.

  When the surface of the intermediate layer is made of a transparent conductive film, the surface shape of the transparent conductive film can be easily controlled by appropriately changing the type, concentration, etching time, etc. of the etchant. Is particularly preferred. It is more preferable to use an acid or alkali solution as an etchant because it can be produced at a lower cost. In this case, the acid solution may be one or a mixture of two or more of hydrochloric acid, sulfuric acid, nitric acid, hydrofluoric acid, acetic acid, formic acid, perchloric acid and the like. Of these, hydrochloric acid and acetic acid are preferable. These acid solutions can be used at a concentration of, for example, about 0.05 to 5% by weight. In particular, in the case of a relatively weak acid such as acetic acid, it is preferably used at a concentration of about 0.1 to 5% by weight. Moreover, 1 type, or 2 or more types of mixtures, such as sodium hydroxide, ammonia, potassium hydroxide, calcium hydroxide, aluminum hydroxide, can be used for an alkaline solution. Of these, sodium hydroxide is preferable. These alkaline solutions are preferably used at a concentration of about 1 to 10% by weight.

  At least a part of the unevenness provided on the surface of the intermediate layer that does not face the substrate by etching can be a substantially spherical or substantially conical hole. When the diameter of the hole is in the range of 200 to 2000 nm, the unevenness in the preferred range described above for tan θ when the height mean square value RMS and the inclination angle are θ is obtained with good reproducibility, which is preferable. Furthermore, the diameter of the hole is more preferably in the range of 400 to 1200 nm.

  In order to improve the light confinement effect, a larger R / L value is desirable, and in order to suppress an increase in defects in the photoelectric conversion element, a smaller R / L value is desirable. Therefore, the uneven texture structure capable of satisfying both of these is preferably an R / L value in the range of 0.2 to 1.2, more preferably in the range of 0.25 to 0.8. preferable.

  In a particularly preferred embodiment of the present invention, the first conductive crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure has a thickness of 1 nm or more and less than 200 nm, and is perpendicular to the substrate in a portion from the surface of the substrate to a height of 200 nm. The crystal grains growing in a columnar shape in any direction occupy at least 50% or more of the entire crystal grains in the portion, and the second conductive type semiconductor layer side surface (macroscopic surface) of the Nth photoelectric conversion element structure or It has a length of 100 nm or more in a direction perpendicular to the surface (macroscopic surface) of the intermediate layer. In this case, there is a strong tendency that a junction interface between the first conductivity type crystalline semiconductor layer and the intrinsic crystalline semiconductor layer is formed in the same crystal grain, which is particularly preferable.

  Furthermore, by setting the film thickness of the first conductive type crystalline semiconductor layer to 1 nm or more and 50 nm or less, the effect of reducing the series resistance or the amount of light absorption in the conductive type layer can be obtained, so that the photoelectric conversion efficiency is improved. It is effective.

By making the first conductive crystalline semiconductor layer and the intrinsic crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure mainly composed of silicon, long wavelength light having a wavelength of 700 nm or more which cannot be used for photoelectric conversion with amorphous silicon is also photoelectrically generated. Can be used for conversion. Thereby, while being able to obtain high photoelectric conversion efficiency, it is possible to obtain a stable solar cell in which light degradation is suppressed. It should be noted that the semiconductor layer mainly composed of silicon is substantially composed only of silicon, or composed of a combination of silicon and other elements, for example, Si _x Sn _1-x in which tin is added to silicon and germanium are added. Also included are Si _x Ge _{1-x and the} like.

  Further, in the first conductive silicon crystalline semiconductor layer and the intrinsic silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure, on the surface of the Nth photoelectric conversion element structure on the second conductive semiconductor layer side or on the surface of the intermediate layer. By setting the crystal plane oriented in parallel to (110), the silicon layer tends to be etched during the formation of the silicon layer, as in the case where the orientation plane is (111) or (100), and film damage The first conductivity type silicon crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure can be formed without using conditions that may cause Thereby, the interface with the 2nd conductivity type layer or intermediate | middle layer of the Nth photoelectric conversion element structure located under a 1st conductivity type silicon crystalline semiconductor layer can be formed favorably.

  Further, in the intrinsic silicon crystalline semiconductor layer formed on the first conductive silicon crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure, the surface of the second conductive semiconductor layer side or the middle of the Nth photoelectric conversion element structure The crystal plane oriented parallel to the surface of the layer can be (110). Thereby, the interface between the first conductive silicon crystalline semiconductor layer and the intrinsic silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure can be satisfactorily formed.

In addition, since the intrinsic silicon crystalline semiconductor layer can be formed without using the condition that the formation speed is slowed by the (110) orientation described above, a highly efficient thin film solar cell can be stably manufactured in a short time. . In particular, the integrated intensity I ₂₂₀ and (111) X-ray diffraction of the (220) X-ray diffraction peak of the intrinsic silicon crystalline semiconductor layer formed on the first conductive silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure. When the ratio I ₂₂₀ / I _{111 of the} peak integrated intensity I ₁₁₁ is 3 or more, preferably 5 or more, good photoelectric conversion characteristics can be obtained.

  When boron is contained as an impurity in the first conductive silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure, the tendency of (110) orientation is particularly strong, which is desirable. The reason for this is not clear, but in References (1) to (4), the reactivity of silicon decreases when boron is present on the surface of single crystal silicon whose plane orientation is (100) or (111). There is a report. Therefore, it is considered that the orientation of (100) or (111) is suppressed by the same phenomenon. When the boron content is in the range of 0.01 to 10 atomic%, the effect of enhancing the (110) orientation is obtained. A preferable range of the boron content is 0.05 to 9 atomic%, and a more preferable range is 0.2 to 8 atomic%.

  Hereinafter, the present invention will be further described by examples.

Example 1
FIG. 3 shows the structure of the thin film solar cell produced. A substrate 31 in which a tin oxide 31b having a textured structure was formed on a glass plate 31a having a smooth surface and a zinc oxide layer 31c having a thickness of 50 nm was formed by an ordinary electron beam evaporation method was used. The average value of the ratio R / L of the irregularity size R and the irregularity interval L on the substrate surface measured by an atomic force microscope was 0.8.

  Subsequently, a first photoelectric conversion element structure 32 was formed by sequentially stacking a p-type amorphous silicon layer 32a, an i-type amorphous silicon layer 32b, and an n-type amorphous silicon layer 32c on the substrate 31 using a plasma CVD apparatus. The plasma CVD apparatus is provided with a forming chamber for each layer, and the substrate can be transferred between each forming chamber and the load lock chamber without breaking the vacuum. A parallel plate type electrode is provided in each forming chamber. The electrode includes a cathode electrode and an anode electrode facing the cathode electrode. High frequency power for plasma excitation is introduced into the cathode electrode. In each forming chamber, the substrate is placed on the anode electrode side having a temperature control function and the surface side having a texture structure is opposed to the cathode electrode.

A high frequency power of 13.56 MHz was input, and a p-type amorphous silicon layer 32a, an i-type amorphous silicon layer 32b, and an n-type amorphous silicon layer 32c were sequentially stacked on the substrate 31 to form an amorphous silicon photoelectric conversion layer 32. The p-type amorphous silicon layer 32a is composed of SiH ₄ gas 12 SCCM, H ₂ gas 30 SCCM, B ₂ H ₆ gas 1 SCCM adjusted to 5000 ppm with H ₂ gas, film forming chamber pressure 20 Pa, discharge power 25 W, and substrate temperature 180 ° C. To a thickness of 15 nm. The i-type amorphous silicon layer 32b was formed under the conditions of SiH ₄ gas 30 SCCM, H ₂ gas 70 SCCM, film forming chamber pressure 30 Pa, discharge power 30 W, substrate temperature 180 ° C., and had a thickness of 350 nm. The n-type amorphous silicon layer 32c is formed under the conditions of SiH ₄ gas 10 SCCM, PH ₃ gas 100 SCCM adjusted to 1000 ppm with H ₂ gas, film forming chamber pressure 27 Pa, discharge power 30 W, substrate temperature 180 ° C. Thickness.

  After forming the first photoelectric conversion element 32, a film made of zinc oxide doped with Ga was formed as an intermediate layer 33 with a thickness of 50 nm by a normal sputtering method. At this time, the average value of the ratio R / L of the irregularity size R and the irregularity interval L on the surface of the intermediate layer 33 measured by an atomic force microscope was 0.6.

After forming the intermediate layer 33, the p-type silicon crystalline semiconductor layer 34a, the i-type silicon crystalline semiconductor layer 34b, the n-type silicon are used by using the plasma CVD apparatus used to form the first photoelectric conversion element 32. The second photoelectric conversion element 34 was formed by stacking the semiconductor layers 34c in this order. The p-type silicon crystalline semiconductor layer 34a uses a mixed gas of SiH ₄ , H ₂ and B ₂ H ₆ , and the gas mixture ratio is such that the boron concentration in the p-type silicon crystalline semiconductor layer is 0.5 atomic%. It was adjusted. The frequency of the high frequency input was 40.68 MHz, the forming chamber pressure was 25 Pa, and the substrate temperature was 150 ° C. The film thickness was 20 nm.

The i-type silicon crystalline semiconductor layer 34b uses a mixed gas of SiH ₄ and H ₂ , and the gas mixing ratio is adjusted so that the i-type silicon layer is sufficiently crystallized. The frequency of the high frequency input was 40.68 MHz, the forming chamber pressure was 40 Pa, and the substrate temperature was 150 ° C. The film thickness was 2 μm. When an i-type silicon crystalline semiconductor layer is formed on a flat glass substrate using this condition, the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak are The ratio I ₂₂₀ / I ₁₁₁ is 10.0 as shown in FIG. 6, and the i-type silicon crystalline semiconductor layer 33 itself exhibits a strong (220) orientation in the normal direction of the flat surface. .

The n-type silicon crystalline semiconductor layer 34c uses a mixed gas of SiH ₄ , H ₂ and PH ₃ , and the gas mixture ratio is adjusted so that the phosphorus concentration in the n-type silicon semiconductor layer is 0.5 atomic%. The frequency of the high frequency input was 40.68 MHz, the forming chamber pressure was 25 Pa, and the substrate temperature was 150 ° C. The film thickness was 20 nm.

  Thereafter, the substrate was taken out from the plasma CVD apparatus, and zinc oxide was vapor-deposited by a normal electron beam vapor deposition method to form the back reflection layer 35. And it isolate | separated into the magnitude | size of 1 cm square by the laser scribing method. Thereafter, a back electrode 36 made of silver was formed by a normal electron beam evaporation method, and a super-straight structure multi-junction thin film solar cell in which light was incident from the substrate 31 side was obtained.

  FIG. 4 schematically shows an image obtained by observing a cross section in the vicinity of the intermediate layer of the multi-junction thin film solar cell with a transmission electron microscope. The p-type silicon crystalline semiconductor layer 42 of the second photoelectric conversion element formed on the uneven surface of the intermediate layer 41 is composed of a plurality of crystal grains, of which about 50% of the crystal grains are substantially perpendicular to the individual uneven surface. The remaining approximately 50% of the grains grew in a direction perpendicular to the surface of the intermediate layer (macroscopic surface). And most of the crystal grains growing in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) are columnar crystal grains having a length of 100 nm or more in the direction perpendicular to the surface of the intermediate layer (macroscopic surface). As the film thickness increased, the proportion in the portion increased while the particle size in the direction parallel to the surface of the intermediate layer (macroscopic surface) increased.

  Furthermore, in order to obtain information on the crystal orientation in a region substantially corresponding to the entire i-type layer, limited-field electron diffraction was performed with a limited-field aperture of 0.8 μm in diameter. From the obtained diffraction image, it was found that the i-type silicon crystalline semiconductor layer 43 had the (110) plane oriented parallel to the surface (macroscopic surface) of the intermediate layer. That is, in the crystal grains of the p-type silicon crystalline semiconductor layer 42 growing in a direction perpendicular to the surface of the intermediate layer (macroscopic surface), the (110) plane is oriented parallel to the surface of the intermediate layer (macroscopic surface). It has been found that the i-type silicon crystalline semiconductor layer 43 formed thereon has grown in such a manner that it inherits the crystal orientation of the p-type silicon crystalline semiconductor layer 42 serving as a base.

Further, in order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, the i-type silicon crystalline semiconductor layer of the second photoelectric conversion element is processed in the same process as this multi-junction thin film solar cell. When the X-ray diffraction was performed on the product formed up to the above, the ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak was 7.0. Yes, the value was close to that when the film was formed on a flat glass substrate.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.205V in the cell area of 1 cm ^2, a short-circuit current 12.8 mA / cm ^2, shape Values of a factor of 0.695 and a photoelectric conversion efficiency of 10.72% were obtained.

  In this embodiment, all layers p, i, and n of the first photoelectric conversion element are formed of amorphous silicon. However, for the i layer, crystalline silicon that becomes the i layer of the second photoelectric conversion element. Any material having a larger energy gap may be used. For example, a-SiC: H may be used. The p-type and n-type layers need not be amorphous silicon layers, and both may be formed of a silicon crystalline semiconductor layer or the like. Moreover, although an intermediate layer is provided in this embodiment, it is not necessarily required in that the first conductive silicon crystalline semiconductor layer of the second photoelectric conversion element grown in a columnar shape in a direction perpendicular to the substrate surface is formed. .

(Comparative Example 1)
In order to verify the effect of the irregular shape, after forming tin oxide having a texture structure on a glass plate having a smooth surface, etching is performed by immersing in a 0.5% hydrochloric acid aqueous solution at a liquid temperature of 25 ° C. for 30 seconds. Then, the surface irregularities were sharpened and a zinc oxide layer having a thickness of 50 nm formed thereon by a normal electron beam evaporation method was used as a substrate. The ratio R / L between the unevenness size R and the unevenness interval L on the surface of the substrate, measured with an atomic force microscope, was 1.65. A thin-film solar cell was produced in the same manner as in Example 1 except for this substrate production process.

  FIG. 5 schematically shows an image obtained by observing a cross section of the thin film solar cell with a transmission electron microscope. However, as in FIG. 4, the n-type silicon semiconductor layer, the back reflecting layer, and the back electrode are omitted. A plurality of crystal grains are formed on the uneven surface, and most of them grow in a direction perpendicular to the individual uneven surface. The crystal grains growing in the direction perpendicular to the uneven surface eventually collide with the crystal grains grown from the adjacent uneven surface, and then change the growth direction in the direction perpendicular to the surface of the intermediate layer (macroscopic surface). It was gradually changing. However, no crystal grains remarkably growing in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) were confirmed.

  Furthermore, in order to obtain information on the crystal orientation in the region corresponding to the entire i-type layer, limited-field electron diffraction was performed with a limited-field aperture of 0.8 μm in diameter. From the obtained diffraction image, it was found that this i-type silicon crystalline semiconductor layer did not clearly have an orientation plane parallel to the surface of the intermediate layer (macroscopic surface). Thus, when the uneven shape is not appropriate, the crystal grains of the p-type silicon crystalline semiconductor layer do not have an orientation plane parallel to the surface (macroscopic surface) of the intermediate layer, and thus are formed on the surface. The i-type silicon crystalline semiconductor layer has also started to grow without having an orientation plane parallel to the surface (macroscopic surface) of the intermediate layer.

Further, in order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, X-ray diffraction was performed on an element formed up to the i-type layer by the same process as this thin film solar cell. ) The ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak is 2.0, and the orientation is remarkably higher than in the case of Example 1. It was inferior.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.080V in the cell area of 1 cm ^2, a short-circuit current 10.6mA / cm ^2, shape Values of a factor of 0.670 and a photoelectric conversion efficiency of 7.67% were obtained. Compared with the results of Example 1, the characteristics were deteriorated in all of the open circuit voltage, the short circuit current, and the form factor. Further, from the results of spectral sensitivity measurement for the elements of Example 1 and Comparative Example 1, it was found that the element of Comparative Example 1 was inferior in sensitivity especially on the long wavelength side. This means that there is a problem in the structure of the photoelectric conversion element located away from the light incident side, and can be explained without contradiction by the observation result of the transmission electron microscope.

(Comparative Example 2)
In order to verify the effect of the uneven shape, a multi-junction thin film solar cell was produced in the same manner as in Example 1 except that the zinc oxide layer 31c on the glass plate was formed to have a thickness of 500 nm. Compared with Example 1, the unevenness | corrugation shape of the surface became gentler by laminating | stacking the zinc oxide layer 31c thickly. As a result of measuring the microscopic uneven structure on the surface of the intermediate layer 33 with an atomic force microscope, the average value of the ratio R / L of the unevenness size R and the unevenness interval L was 0.05.

In order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, X-ray diffraction was performed on an i-type layer formed by the same process as this thin film solar cell. The ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak is 8.5, which is the same value as when a film is formed on a flat glass substrate. Met.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.208V in the cell area of 1 cm ^2, a short-circuit current 10.6mA / cm ^2, shape Values of a factor of 0.690 and a photoelectric conversion efficiency of 8.84% were obtained. Compared with the results of Example 1, the open circuit voltage and the shape factor values were almost the same, but the short circuit current was reduced because the light confinement effect was small.

(Example 2)
In order to verify the effect of channeling particles, when the p-type silicon crystalline semiconductor layer 32 of the second photoelectric conversion element 34 is formed on the intermediate layer 33 having an uneven surface, SiH ₄ , H ₂ and B _{2 are used} as source gases. A mixture of H ₆ with Ar gas added was used. Other than that produced the thin film solar cell similarly to Example 1. FIG. At this time, the mixing ratio of SiH ₄ , H ₂ and B ₂ H _{6 was} the same as in Example 1, and the amount of Ar added was adjusted to be 5% of the total mixed gas.

  When the cross section of the obtained thin film solar cell was observed with a transmission electron microscope, a plurality of crystal grains were formed on the concavo-convex surface as in the case of Example 1. The grain density was reduced, and about 80% of the crystal grains grew in a direction perpendicular to the surface of the intermediate layer (macroscopic surface). And most of the crystal grains growing in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) are columnar crystal grains having a length of 100 nm or more in the direction perpendicular to the surface of the intermediate layer (macroscopic surface). As the film thickness increases, the particle size in the direction parallel to the surface of the intermediate layer (macroscopic surface) increases, and the tendency to increase the proportion in the portion is the same.

  Furthermore, in order to obtain information on the crystal orientation in a region substantially corresponding to the entire i-type layer, limited-field electron diffraction was performed with a limited-field aperture of 0.8 μm in diameter. From the obtained diffraction image, it was found that the (110) plane of this i-type silicon crystalline semiconductor layer was oriented parallel to the surface (macroscopic surface) of the intermediate layer. That is, the crystal grains of the p-type layer growing in a direction perpendicular to the surface of the intermediate layer (macroscopic surface) are (110) oriented parallel to the surface of the intermediate layer (macroscopic surface), and It was found that the formed i-type silicon crystalline semiconductor layer grew in a form that inherited the crystal orientation of the p-type layer serving as the base.

In addition, in order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, X-ray diffraction was performed on an i-type layer formed by the same process as this thin film solar cell. 220) The ratio I ₂₂₀ / I ₁₁₁ between the integrated intensity I ₂₂₀ of the X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak was 8.

The reason why <110> oriented crystal grains preferentially grow in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) by Ar addition is not yet clear, but is almost perpendicular to the macroscopic surface due to the sheath potential. Ar ^{+ in} the plasma accelerated in the direction promotes growth to <110>, which is the channeling direction of silicon, and crystal grains in which <110> is not oriented in a direction substantially perpendicular to the macroscopic surface This is thought to be due to the effect of sputtering.

  In this embodiment, Ar is used as a rare gas to be added, but He or Ne may be used.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.242V in the cell area of 1 cm ^2, a short-circuit current 13.1mA / cm ^2, shape Values of a factor of 0.714 and a photoelectric conversion efficiency of 11.62% were obtained. Compared with the device of Example 1, the open circuit voltage and the shape factor were further improved. This is because the structure in the vicinity of the interface between the p-type layer and the i-type layer has become more appropriate, and can be explained without contradiction from the observation result of the transmission electron microscope.

  In the above embodiment, a super-straight type multi-junction thin film solar cell using a substrate in which a tin oxide film having a texture structure is formed on a glass plate having a smooth surface and a zinc oxide film is further formed thereon is provided. explained. However, it will be apparent to those skilled in the art that there are many embodiments according to the present invention other than the above-described embodiments. For example, the structure of the multi-junction thin film solar cell may be a substrate type in which light is incident from the side of a silicon crystalline semiconductor layer formed as a photoelectric conversion layer on a glass substrate or a metal substrate.

11 ... Uneven surface 12 ... Virtual surface (macroscopic surface) by visual observation
13... The first conductive silicon crystalline semiconductor layer 14 of the (N + 1) th photoelectric conversion element 14... The intrinsic silicon crystalline semiconductor layers 15 and 24 of the (N + 1) th photoelectric conversion element.
16, 25 ... Unevenness interval L
31 ... substrate 21, 31a ... glass plate 22, 31b ... tin oxide layer 31c ... zinc oxide layer 32 ... first photoelectric conversion element 32a ... p-type amorphous silicon layer 32b ... i-type amorphous silicon layer 32c ... N-type amorphous silicon layer 33, 41, 51 ... intermediate layer 34 ... second photoelectric conversion elements 34a, 42, 52 ... p-type silicon crystalline semiconductor layer 34b, 43, 53 ... i-type silicon crystalline semiconductor layer 34c ... N-type silicon semiconductor layer 35 ... Back surface reflection layer 36... Back electrode 61... I ₂₂₀ / I ₁₁₁ curve 62 with respect to substrate unevenness index R / L under conditions with few channeling particles. I ₂₂₀ / I ₁₁₁ curve

The present invention relates to a thin film solar cell having high photoelectric conversion efficiency and a method for producing the same .

  Solar cells are attracting attention as an alternative energy source for fossil fuels such as oil, where there is concern about future supply and demand, and there is a problem of carbon dioxide emissions that cause global warming.

  This solar cell uses a pn junction for a photoelectric conversion layer that converts light energy into electric power, and silicon is most commonly used as a semiconductor constituting the pn junction. From the viewpoint of photoelectric conversion efficiency, it is preferable to use single crystal silicon, but there are problems of raw material supply, large area, and low cost.

  On the other hand, thin-film solar cells using amorphous silicon as a photoelectric conversion layer as an advantageous material for realizing large area and low cost have been put into practical use, but the photoelectric conversion efficiency is lower than that of single crystal silicon solar cells. Inferior. Furthermore, since amorphous silicon has a phenomenon called the Staebler-Wronski effect in which the density of defects in the film increases as it is irradiated with light, the amorphous silicon solar cell inevitably suffers from the problem of deterioration in photoelectric conversion efficiency over time.

  Therefore, in recent years, in order to realize a solar cell that has both high and stable photoelectric conversion efficiency at the single crystal silicon solar cell level, large area and low cost at the amorphous silicon solar cell level, a crystalline silicon photoelectric conversion layer Use in is being considered. In particular, a thin film solar cell in which a crystalline silicon thin film is formed by using a thin film formation technique by chemical vapor deposition (hereinafter referred to as a CVD method) similar to the case of amorphous silicon has attracted attention.

  JP-A-1-289173 discloses a multi-junction thin film in which a photoelectric conversion element using amorphous silicon as an active layer and a photoelectric conversion element using polycrystalline silicon having an energy gap smaller than that of amorphous silicon as an active layer are stacked. A solar cell is disclosed. Such a solar cell can use solar energy more efficiently than a single-junction type by adopting a configuration in which sunlight is incident from the photoelectric conversion element side using amorphous silicon as an active layer. There are advantages. Furthermore, an amorphous silicon layer and a polycrystalline silicon layer that can obtain a high voltage because a plurality of photoelectric conversion elements are connected in series, and that the amorphous silicon layer as an active layer can be thinned, so that deterioration of photoelectric conversion efficiency over time can be suppressed. As a means to achieve both high efficiency and low cost, research and development has been actively conducted.

  However, despite vigorous research and development, the photoelectric conversion efficiency of the above-mentioned multi-junction thin-film solar cell is far from the expected value so far. A major factor is that the defect density in the crystalline silicon active layer is high. In order to reduce the defect density, it is effective to increase the crystal grain size. In solar cells that have a structure in which carriers flow in the film thickness direction, the existence of polycrystalline grains is a structure that does not have grain boundaries that cross the film thickness direction, that is, crystal grains grow in a columnar shape in the film thickness direction. Such a structure is desirable, and such a structure is easily obtained when the crystal orientation is aligned with respect to the film thickness direction.

  Various attempts have been made to obtain a polycrystalline silicon thin film having a large crystal grain size. For example, a so-called laser annealing method is known in which an amorphous silicon thin film formed on a substrate by a CVD method or the like is irradiated with a laser beam and melted, and then a polycrystalline silicon thin film is obtained by solidification. There is also known a so-called partial doping method in which crystallization by solid phase growth is selectively started by doping a part of an amorphous silicon film with phosphorus, boron, or the like. However, the above-described method has problems that are difficult to put into practical use, such as high equipment cost and heat treatment time of several tens of hours.

On the other hand, various attempts have been made to obtain a crystalline silicon thin film in which the crystal orientation is aligned with respect to the film thickness direction. For example, Japanese Patent Application Laid-Open No. 7-240531 discloses a method for manufacturing a solar cell and a multilayer thin film, in which an inert gas beam is irradiated from a plurality of directions during or after the semiconductor thin film is deposited. According to this method, a silicon thin film is obtained in which the <111> direction, which is the densest direction of crystalline silicon, is aligned with the irradiation direction of an inert gas beam serving as a channeling beam. Japanese Patent Laid-Open No. 11-278988 discloses a single crystal thin film manufacturing method in which a SiH ₃ beam is irradiated from a direction perpendicular to a substrate and a H beam is irradiated from a direction parallel to the substrate during silicon film deposition. The method is shown. According to this method, since the molecular weight is large, the <111> direction, which is the densest direction of crystalline silicon, is aligned in the direction perpendicular to the substrate surface, which is the irradiation direction of the SiH ₃ beam, which is a non-channeling beam, and the molecular weight is small. It is said that a single crystal silicon thin film can be obtained by aligning the <110> direction of crystalline silicon with the H beam irradiation direction.

  As mentioned above, multi-junction thin-film solar cells are attracting attention in addition to the expectation of higher efficiency and suppression of deterioration over time. In addition, amorphous silicon layers and crystalline silicon layers are continuously formed by CVD. This is because it can be formed. However, H.C. In Yamamoto et al, PVSEC-11, Sapporo, Japan, 1999, when microcrystalline silicon is formed by plasma CVD on a substrate having a textured structure with surface irregularities in which tin oxide is laminated on glass, as shown in FIG. In addition, it has been reported that silicon grains grow preferentially in the direction perpendicular to the individual uneven surfaces, and a large number of defects occur due to collision of crystal grains with different crystal orientations grown from different uneven surfaces. It was. Such a defect due to the crystalline semiconductor layer rather than the amorphous semiconductor layer becomes a carrier recombination center and significantly deteriorates the photoelectric conversion efficiency, and thus must be eliminated as much as possible. H. Yamamoto et al. Show that when the size of the unevenness is reduced by further laminating zinc oxide on the tin oxide having surface unevenness, the silicon crystal grains are perpendicular to the surface of the zinc oxide as in the case of tin oxide. It was also reported at the same time that crystal grains grown from different uneven surfaces collide with each other but their orientation difference is small, resulting in fewer defects. However, it is clear that the surface irregularities of the substrate should be as small as possible in order to reduce defects in the crystalline semiconductor layer.

  However, as shown in Japanese Patent No. 1681183, the transparent conductive film having surface irregularities causes irregular reflection of light and increases the optical path length in the silicon film, so that the current value generated by light absorption is reduced. There is an important function of increasing the so-called light confinement effect. Furthermore, since the light confinement effect can make the photoelectric conversion layer thinner by improving the photoelectric conversion efficiency, an advantage that the film forming process can be shortened is also produced. This results in a significant improvement in throughput in a multi-junction thin film solar cell having a crystalline silicon layer that requires a photoelectric conversion layer thickness several times that of an amorphous silicon layer due to the difference in light absorption characteristics. Therefore, it should be avoided to eliminate or reduce the surface roughness.

  However, at present, it is very difficult to achieve both the defect density reduction in the crystalline silicon thin film and the light confinement effect due to the uneven surface, and this problem has not been solved. For example, a silicon thin film having a low defect density in the film can be produced by using the method for producing a crystalline silicon thin film described in Japanese Patent Application Laid-Open No. 7-240531 or Japanese Patent Application Laid-Open No. 11-278888 described above. Since the substrate to be formed is substantially flat and cannot exhibit the light confinement effect, it is clear that there is a limit to the increase in efficiency.

  Japanese Patent Application Laid-Open No. 10-150209 discloses a photoelectric conversion characterized in that it has a relationship between the normal direction of the substrate and the longitudinal direction of the columnar crystal grains or the normal direction of the uneven surface. Although an element is disclosed, the above-described defect is inevitable as long as the normal direction of the substrate and the longitudinal direction of the columnar crystal grains have an inclination that is not small. Furthermore, Japanese Patent Application Laid-Open No. 10-117006, Japanese Patent Application Laid-Open No. 10-294481, Japanese Patent Application Laid-Open No. 11-214728, Japanese Patent Application Laid-Open No. 11-266027, and Japanese Patent Application Laid-Open No. 2000-58892 have a back surface with an uneven surface. A thin film solar cell in which a lower photoelectric conversion element having a photoelectric conversion layer made of a polycrystalline silicon layer is formed on an electrode, and the polycrystalline silicon layer has a (110) preferential crystal orientation plane parallel to the substrate surface is shown. However, in forming the polycrystalline silicon layer near the surface of the back electrode having a concavo-convex shape, the above-described growth behavior of the polycrystalline silicon is not considered at all, so it is clear that the occurrence of defects is inevitable. is there.

An object of the present invention is to provide a highly efficient multi-junction thin film solar cell having a crystalline semiconductor layer in which an increase in defect density is suppressed while having a sufficient light confinement effect, and a method for manufacturing the same. is there.

  According to the present invention, there is provided a method of manufacturing a multi-junction thin film solar cell including a plurality of photoelectric conversion element structures stacked on a substrate, wherein the photoelectric conversion element structure includes a first conductive type semiconductor layer, an intrinsic semiconductor layer, and a second conductive type. Type semiconductor layers are sequentially stacked, a photoelectric conversion element structure stacking step of stacking two photoelectric conversion element structures on a substrate, an intermediate layer forming step of forming an intermediate layer after the photoelectric conversion element structure stacking step, After the layer formation step, a crystalline semiconductor layer formation step of forming a crystalline first conductive semiconductor layer and an intrinsic semiconductor layer is provided, and the surface of the intermediate layer has a ratio R / L of the unevenness size R to the unevenness interval L Has a textured structure with irregularities in the range of 0.1 to 1.5, and the crystalline semiconductor layer forming step is mainly composed of crystal grains growing in a columnar shape in a direction perpendicular to the surface of the intermediate layer First conductivity Semiconductor layer and a method of manufacturing a multi-junction thin-film solar cell which is a step of forming a crystalline semiconductor layer made of intrinsic semiconductor layer.

Is provided by Ri multijunction thin film solar cell in the present invention, the solar cell comprises a substrate and a plurality of photoelectric conversion element structure that is laminated on a substrate, in which, of a plurality of photoelectric conversion element structure each the first conductive type semiconductor layer, intrinsic semiconductor layer and the second conductive type semiconductor layer are stacked in this order, among the plurality of stacked photoelectric conversion element structure, a second photoelectric conversion element as counted from the substrate side only between the structure and the third photoelectric conversion element structure, the intermediate layer is provided, the third surface of the intermediate layer supporting the photoelectric conversion element structure, the ratio of the unevenness magnitude R for uneven spacing L R / L has a texture structure due to irregularities such as in the range of 0.1 to 1.5, and the third of the first conductivity type semiconductor layer and intrinsic semiconductor layer of the photoelectric conversion element structure, the third Table of the intermediate layer supporting the photoelectric conversion element structure In a crystalline semiconductor layer which is composed mainly by the crystal grains have grown in a columnar direction perpendicular.

In the present invention, the first conductive type crystalline semiconductor layer of the photoelectric conversion element structure of the third photoelectric conversion element structure preferably has a 200nm thickness of less than or 1 nm. Further, the third in the portion of the first conductivity type semiconductor layer side surface of the photoelectric conversion element structure to a height 200 nm, perpendicular to the second surface of the second conductive type semiconductor layer side surface or the intermediate layer of the photoelectric conversion element structure crystal grains are grown into columnar in a direction occupies at least 50% or more crystal grains across the at that portion, and the second surface of the second conductive type semiconductor layer side surface or the intermediate layer of the photoelectric conversion element structure It is preferable to have a length of 100 nm or more in the vertical direction.

In the present invention, it is preferable first conductivity type semi-conductor layer of the third photoelectric conversion element structure having a thickness of not less 50nm least 1 nm.

In the present invention, the first conductivity type semi-conductive layer and said true semi-conductor layer of the third photoelectric conversion element structure is preferably made of mainly silicon. Further, the third in the first conductivity type semi-conductor layer and a true semi-conductor layer of the photoelectric conversion element structure, oriented parallel to the second surface of the second conductive type semiconductor layer side surface or the intermediate layer of the photoelectric conversion element structure It is preferable that the crystal plane to be mainly (110).

In the present invention, the third obtained for an intrinsic crystalline semiconductor layer of the photoelectric conversion element structure, (220) integral of the X-ray diffraction peaks of the integrated intensity I ₂₂₀ (111) plane of the X-ray diffraction peaks of surface intensity I ₁₁₁ The ratio I ₂₂₀ / I ₁₁₁ is preferably 3 or more.

In the present invention, the first conductive type crystalline semiconductor layer of the third photoelectric conversion element structure that contains boron as an impurity, particularly preferred.

  In the present invention, the intermediate layer is preferably a transparent conductor mainly made of zinc oxide.

The present invention provides a method for producing a multi-junction thin film solar cell, the multi-junction thin film solar cell comprising a substrate and a plurality of photoelectric conversion element structures stacked on the substrate, and a plurality of photoelectric conversion element structures. The first conductive semiconductor layer, the intrinsic semiconductor layer, and the second conductive semiconductor layer are sequentially stacked, and the Nth (N is an arbitrary integer greater than or equal to 1) photoelectric conversion element structure counted from the substrate type The surface of the second conductivity type semiconductor layer side has a texture structure with unevenness such that the ratio R / L of the unevenness size R to the unevenness interval L is in the range of 0.1 to 1.5, and the Nth photoelectric layer When forming the first conductive type semiconductor layer of the (N + 1) th photoelectric conversion element structure formed on the conversion element structure, SiH ₄ , H ₂ , B ₂ H ₆ And a method of manufacturing a multi-junction thin film solar cell using a rare gas as a source gas.

The present invention provides a method for producing a multi-junction thin film solar cell, the multi-junction thin film solar cell comprising a substrate and a plurality of photoelectric conversion element structures stacked on the substrate, and a plurality of photoelectric conversion element structures. The first conductive semiconductor layer, the intrinsic semiconductor layer, and the second conductive semiconductor layer are sequentially stacked, and the Nth (N is an arbitrary integer greater than or equal to 1) photoelectric conversion element structure counted from the substrate type And the (N + 1) th photoelectric conversion element structure is provided with an intermediate layer, and the surface of the intermediate layer supporting the (N + 1) th photoelectric conversion element has an unevenness size R with respect to the unevenness interval L. A first conductive type semiconductor layer having a texture structure with irregularities such that the ratio R / L is in the range of 0.1 to 1.5 and having an N + 1th photoelectric conversion element structure formed on the intermediate layer; When forming a film, SiH ₄ , H ₂ , B ₂ H ₆ And a method of manufacturing a multi-junction thin film solar cell using a rare gas as a source gas.

  In this specification, unless otherwise specified, the term “crystalline semiconductor layer” includes all forms of semiconductor layers including a plurality of crystal grains having different orientations as constituent elements. Therefore, the term “crystalline semiconductor layer” means, for example, not only a semiconductor layer that is an assembly of a plurality of crystallites, but also a semiconductor layer in which a crystal component called a microcrystal or a microcrystal and an amorphous component are mixed. Including.

  Second conductive semiconductor layer side surface of Nth (N is an arbitrary integer of 1 or more) photoelectric conversion element structure or N + 1th photoelectric conversion element structure counting from the substrate side of the multi-junction thin film solar cell according to the present invention. On the surface of the intermediate layer that supports the surface, an uneven texture structure is provided. The size of the unevenness is typically in the range of 0.01 to 10 μm, preferably 0.05 to 2 μm, and the uneven shape cannot be determined visually. That is, these surfaces are recognized as smooth surfaces by visual inspection. Therefore, in this specification, “perpendicular to the surface of the Nth photoelectric conversion element structure on the second conductive semiconductor layer side or the surface of the intermediate layer” or “the surface of the Nth photoelectric conversion element structure on the second conductive type semiconductor layer side or When referred to as “parallel to the surface of the intermediate layer”, unless otherwise specified, the criteria of “perpendicular” and “parallel” are surfaces (macroscopic surfaces) that are regarded as smooth by this visual observation.

  On the other hand, when viewed microscopically, the surface supporting the photoelectric conversion element structure forms a textured structure with irregularities. This uneven shape can be measured by an atomic force microscope, and the unevenness size R and the unevenness interval L are defined as follows. That is, in an arbitrary region of the surface, line measurement of the surface irregularity shape is performed with an atomic force microscope over a length of 5 μm, and the surface roughness waveform obtained by the measurement is the surface irregularity defined in Japanese Industrial Standard JIS B0602-1994. The arithmetic average value Ra is defined as the unevenness size R, and the average interval Sm of the surface unevenness defined by the Japanese Industrial Standard JIS B0602-1994 is defined as the unevenness interval L.

  The present invention provides a multi-junction thin film solar cell having a first conductive crystalline semiconductor layer and an intrinsic crystalline semiconductor layer that are grown in a columnar shape in a direction perpendicular to the macroscopic surface even on an uneven surface. With this structure, it is possible to achieve both an increase in the amount of light absorption due to the light confinement effect and good carrier transport characteristics in the film thickness direction due to a reduction in defects in the crystalline semiconductor layer. Therefore, the multi-junction thin film solar cell according to the present invention can have high photoelectric conversion efficiency.

The schematic diagram which shows the silicon crystalline semiconductor layer grown in the column shape in the direction perpendicular | vertical to a macroscopic surface. The schematic diagram which shows the silicon crystalline semiconductor layer grown in the column shape in the direction perpendicular | vertical to each uneven | corrugated surface. The schematic diagram which shows an example of the thin film solar cell structure by this invention. 4 is a schematic diagram showing a transmission electron microscope observation image of a sample cross section in Example 1. FIG. The schematic diagram which shows the transmission electron microscope observation image of the sample cross section in the comparative example 1. FIG. The figure which shows the relationship between ratio _I220 / _I111 and R / L.

The present inventor performed plasma CVD on a substrate in which the ratio R / L of the unevenness size R and the unevenness interval L was appropriately changed by etching zinc oxide coated on a glass plate having a smooth surface. A study was made to form a silicon crystalline semiconductor layer. As a result, under the condition that the number of channeling particles is small, the orientation of the silicon crystalline semiconductor layer can be strengthened only when the average value of R / L is small, in other words, when the surface unevenness is small, as shown by the curve 61 in FIG. On the other hand, under the condition where there are many channeling particles, as shown by the curve 62 in FIG. 6, not only when the average value of R / L is small, that is, when the surface unevenness is small, R / L is 0.00. In the range of 1 to 1.5, the ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) plane X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) plane X-ray diffraction peak is It was discovered that it was possible to make it 3 or more.

  For the first time, this discovery makes it possible to achieve both high light absorption due to the light confinement effect and good carrier transport properties in the film thickness direction due to the absence of defects in the crystalline semiconductor layer, resulting in high photoelectric conversion efficiency. A multi-junction thin-film solar cell having the above structure was realized.

  As a substrate material used for the multi-junction thin film solar cell of the present invention, various materials such as glass, metal, resin having heat resistance of about 200 ° C. such as polyimide and polyvinyl, and those laminated thereon are used. it can. In addition, those whose surfaces are coated with a metal film, a transparent conductive film, an insulating film, or the like are also included. The thickness of the substrate is not particularly limited, but is, for example, about 0.1 to 30 mm so as to have an appropriate strength and weight so that the structure can be supported.

  The surface of the substrate may be uneven. As a means for providing unevenness, for example, a film that forms unevenness on the surface at the same time as deposition may be formed on a substrate having a smooth surface. The film on which the unevenness is formed on the surface may be the same material as the substrate or a different material. Further, the unevenness can be formed by performing mechanical processing such as sandblasting or chemical processing such as etching on the substrate surface.

  In the multi-junction thin film solar cell of the present invention, among the plurality of photoelectric conversion element structures stacked on the substrate, the second (N is an arbitrary integer greater than or equal to 1) photoelectric conversion element structure counted from the substrate side. The surface of the conductive semiconductor layer side has a textured structure with unevenness such that the ratio R / L of the unevenness size R to the unevenness interval L is in the range of 0.1 to 1.5. Further, the (N + 1) th photoelectric conversion element structure has a first conductive structure mainly composed of crystal grains growing in a columnar shape in a direction perpendicular to the macroscopic surface even on the surface having a texture as shown in FIG. A type crystalline semiconductor layer and an intrinsic crystalline semiconductor layer. With these characteristics, it is possible to obtain a multi-junction thin-film solar cell that exhibits high photoelectric conversion efficiency that exhibits a light confinement effect and suppresses an increase in defects in the photoelectric conversion element structure.

  In another embodiment of the multi-junction thin film solar cell according to the present invention, among a plurality of photoelectric conversion element structures stacked on a substrate, Nth (N is an arbitrary integer greater than or equal to 1) photoelectric conversion counted from the substrate side The surface of the intermediate layer provided between the element structure and the (N + 1) th photoelectric conversion element structure has a ratio R / L of the unevenness size R to the unevenness interval L in the range of 0.1 to 1.5. It has a textured structure with irregularities. Further, the (N + 1) th photoelectric conversion element structure has a first conductive structure mainly composed of crystal grains growing in a columnar shape in a direction perpendicular to the macroscopic surface even on the surface having a texture as shown in FIG. A type crystalline semiconductor layer and an intrinsic crystalline semiconductor layer. These characteristics cause a sufficient light confinement effect and suppress an increase in defects in the photoelectric conversion element structure. Therefore, the multi-junction thin film solar cell according to the present invention can provide high photoelectric conversion efficiency.

  In the present invention, the intermediate layer is a layer provided between layers having different conductivity types in adjacent photoelectric conversion element structures. The intermediate layer is provided in order to prevent junction formation in the reverse direction that occurs when different conductive type layers are directly bonded, or bonding failure due to mixing of impurities.

  As the intermediate layer, a transparent conductive film made of tin oxide, indium oxide, zinc oxide or the like is suitable. Further, the intermediate layer may be composed of a single material of these, or may be a laminate of a plurality of layers containing these materials. In particular, the surface on the side not facing the substrate is more preferably made of a transparent conductive film. In particular, a transparent conductive film mainly composed of zinc oxide is more preferable because it has an advantage that it is inexpensive and has high plasma resistance and is hardly deteriorated.

  These transparent conductive films can be produced by known methods such as sputtering, atmospheric pressure CVD, reduced pressure CVD, electron beam evaporation, sol-gel, and electrodeposition. Among them, the sputtering method is particularly preferable because it is easy to control the transmittance and resistivity of the transparent conductive film to those suitable for the thin film solar cell.

A small amount of impurities may be added to these transparent conductive films. For example, in the case of zinc oxide, the resistivity is reduced by containing a Group 3B element such as gallium or aluminum of about 5 × 10 ^{20 to} 5 × 10 ²¹ cm ⁻³ or a Group 1B element such as copper. , Preferred for use as an intermediate layer. In addition, if the thickness of these transparent conductive films (intermediate layers) is too thin, there will be a problem in the uniformity of characteristics, and if it is too thick, the decrease in transmittance and the increase in series resistance will cause a decrease in photoelectric conversion efficiency and an increase in cost. Since it causes, Preferably it is about 1-50 nm.

  As a means for providing unevenness on the surface of the intermediate layer, for example, the intermediate layer may be formed under conditions such that unevenness is formed on the surface simultaneously with the deposition. At this time, the formation condition of the intermediate layer may be determined considering that the uneven shape of the surface of the intermediate layer is affected by the uneven shape of the photoelectric conversion element as a base. It is also possible to form irregularities by performing mechanical processing such as sandblasting or chemical processing such as etching on the surface of the intermediate layer.

  When the surface of the intermediate layer is made of a transparent conductive film, the surface shape of the transparent conductive film can be easily controlled by appropriately changing the type, concentration, etching time, etc. of the etchant. Is particularly preferred. It is more preferable to use an acid or alkali solution as an etchant because it can be produced at a lower cost. In this case, the acid solution may be one or a mixture of two or more of hydrochloric acid, sulfuric acid, nitric acid, hydrofluoric acid, acetic acid, formic acid, perchloric acid and the like. Of these, hydrochloric acid and acetic acid are preferable. These acid solutions can be used at a concentration of, for example, about 0.05 to 5% by weight. In particular, in the case of a relatively weak acid such as acetic acid, it is preferably used at a concentration of about 0.1 to 5% by weight. Moreover, 1 type, or 2 or more types of mixtures, such as sodium hydroxide, ammonia, potassium hydroxide, calcium hydroxide, aluminum hydroxide, can be used for an alkaline solution. Of these, sodium hydroxide is preferable. These alkaline solutions are preferably used at a concentration of about 1 to 10% by weight.

  At least a part of the unevenness provided on the surface of the intermediate layer that does not face the substrate by etching can be a substantially spherical or substantially conical hole. When the diameter of the hole is in the range of 200 to 2000 nm, the unevenness in the preferred range described above for tan θ when the height mean square value RMS and the inclination angle are θ is obtained with good reproducibility, which is preferable. Furthermore, the diameter of the hole is more preferably in the range of 400 to 1200 nm.

  In order to improve the light confinement effect, a larger R / L value is desirable, and in order to suppress an increase in defects in the photoelectric conversion element, a smaller R / L value is desirable. Therefore, the uneven texture structure capable of satisfying both of these is preferably an R / L value in the range of 0.2 to 1.2, more preferably in the range of 0.25 to 0.8. preferable.

  In a particularly preferred embodiment of the present invention, the first conductive crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure has a thickness of 1 nm or more and less than 200 nm, and is perpendicular to the substrate in a portion from the surface of the substrate to a height of 200 nm. The crystal grains growing in a columnar shape in any direction occupy at least 50% or more of the entire crystal grains in the portion, and the second conductive type semiconductor layer side surface (macroscopic surface) of the Nth photoelectric conversion element structure or It has a length of 100 nm or more in a direction perpendicular to the surface (macroscopic surface) of the intermediate layer. In this case, there is a strong tendency that a junction interface between the first conductivity type crystalline semiconductor layer and the intrinsic crystalline semiconductor layer is formed in the same crystal grain, which is particularly preferable.

  Furthermore, by setting the film thickness of the first conductive type crystalline semiconductor layer to 1 nm or more and 50 nm or less, the effect of reducing the series resistance or the amount of light absorption in the conductive type layer can be obtained, so that the photoelectric conversion efficiency is improved. It is effective.

By making the first conductive crystalline semiconductor layer and the intrinsic crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure mainly composed of silicon, long wavelength light having a wavelength of 700 nm or more which cannot be used for photoelectric conversion with amorphous silicon is also photoelectrically generated. Can be used for conversion. Thereby, while being able to obtain high photoelectric conversion efficiency, it is possible to obtain a stable solar cell in which light degradation is suppressed. It should be noted that the semiconductor layer mainly composed of silicon is substantially composed only of silicon, or composed of a combination of silicon and other elements, for example, Si _x Sn _1-x in which tin is added to silicon and germanium are added. Also included are Si _x Ge _{1-x and the} like.

  Further, in the first conductive silicon crystalline semiconductor layer and the intrinsic silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure, on the surface of the Nth photoelectric conversion element structure on the second conductive semiconductor layer side or on the surface of the intermediate layer. By setting the crystal plane oriented in parallel to (110), the silicon layer tends to be etched during the formation of the silicon layer, as in the case where the orientation plane is (111) or (100), and film damage The first conductivity type silicon crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure can be formed without using conditions that may cause Thereby, the interface with the 2nd conductivity type layer or intermediate | middle layer of the Nth photoelectric conversion element structure located under a 1st conductivity type silicon crystalline semiconductor layer can be formed favorably.

  Further, in the intrinsic silicon crystalline semiconductor layer formed on the first conductive silicon crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure, the surface of the second conductive semiconductor layer side or the middle of the Nth photoelectric conversion element structure The crystal plane oriented parallel to the surface of the layer can be (110). Thereby, the interface between the first conductive silicon crystalline semiconductor layer and the intrinsic silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure can be satisfactorily formed.

In addition, since the intrinsic silicon crystalline semiconductor layer can be formed without using the condition that the formation speed is slowed by the (110) orientation described above, a highly efficient thin film solar cell can be stably manufactured in a short time. . In particular, the integrated intensity I ₂₂₀ and (111) X-ray diffraction of the (220) X-ray diffraction peak of the intrinsic silicon crystalline semiconductor layer formed on the first conductive silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure. When the ratio I ₂₂₀ / I _{111 of the} peak integrated intensity I ₁₁₁ is 3 or more, preferably 5 or more, good photoelectric conversion characteristics can be obtained.

  When boron is contained as an impurity in the first conductive silicon crystalline semiconductor layer in the (N + 1) th photoelectric conversion element structure, the tendency of (110) orientation is particularly strong, which is desirable. The reason for this is not clear, but in References (1) to (4), the reactivity of silicon decreases when boron is present on the surface of single crystal silicon whose plane orientation is (100) or (111). There is a report. Therefore, it is considered that the orientation of (100) or (111) is suppressed by the same phenomenon. When the boron content is in the range of 0.01 to 10 atomic%, the effect of enhancing the (110) orientation is obtained. A preferable range of the boron content is 0.05 to 9 atomic%, and a more preferable range is 0.2 to 8 atomic%.

  Hereinafter, the present invention will be further described by examples.

Example 1
FIG. 3 shows the structure of the thin film solar cell produced. A substrate 31 in which a tin oxide 31b having a textured structure was formed on a glass plate 31a having a smooth surface and a zinc oxide layer 31c having a thickness of 50 nm was formed by an ordinary electron beam evaporation method was used. The average value of the ratio R / L of the irregularity size R and the irregularity interval L on the substrate surface measured by an atomic force microscope was 0.8.

  Subsequently, a first photoelectric conversion element structure 32 was formed by sequentially stacking a p-type amorphous silicon layer 32a, an i-type amorphous silicon layer 32b, and an n-type amorphous silicon layer 32c on the substrate 31 using a plasma CVD apparatus. The plasma CVD apparatus is provided with a forming chamber for each layer, and the substrate can be transferred between each forming chamber and the load lock chamber without breaking the vacuum. A parallel plate type electrode is provided in each forming chamber. The electrode includes a cathode electrode and an anode electrode facing the cathode electrode. High frequency power for plasma excitation is introduced into the cathode electrode. In each forming chamber, the substrate is placed on the anode electrode side having a temperature control function and the surface side having a texture structure is opposed to the cathode electrode.

A high frequency power of 13.56 MHz was input, and a p-type amorphous silicon layer 32a, an i-type amorphous silicon layer 32b, and an n-type amorphous silicon layer 32c were sequentially stacked on the substrate 31 to form an amorphous silicon photoelectric conversion layer 32. The p-type amorphous silicon layer 32a is composed of SiH ₄ gas 12 SCCM, H ₂ gas 30 SCCM, B ₂ H ₆ gas 1 SCCM adjusted to 5000 ppm with H ₂ gas, film forming chamber pressure 20 Pa, discharge power 25 W, and substrate temperature 180 ° C. To a thickness of 15 nm. The i-type amorphous silicon layer 32b was formed under the conditions of SiH ₄ gas 30 SCCM, H ₂ gas 70 SCCM, film forming chamber pressure 30 Pa, discharge power 30 W, substrate temperature 180 ° C., and had a thickness of 350 nm. The n-type amorphous silicon layer 32c is formed under the conditions of SiH ₄ gas 10 SCCM, PH ₃ gas 100 SCCM adjusted to 1000 ppm with H ₂ gas, film forming chamber pressure 27 Pa, discharge power 30 W, substrate temperature 180 ° C. Thickness.

  After forming the first photoelectric conversion element 32, a film made of zinc oxide doped with Ga was formed as an intermediate layer 33 with a thickness of 50 nm by a normal sputtering method. At this time, the average value of the ratio R / L of the irregularity size R and the irregularity interval L on the surface of the intermediate layer 33 measured by an atomic force microscope was 0.6.

After forming the intermediate layer 33, the p-type silicon crystalline semiconductor layer 34a, the i-type silicon crystalline semiconductor layer 34b, the n-type silicon are used by using the plasma CVD apparatus used to form the first photoelectric conversion element 32. The second photoelectric conversion element 34 was formed by stacking the semiconductor layers 34c in this order. The p-type silicon crystalline semiconductor layer 34a uses a mixed gas of SiH ₄ , H ₂ and B ₂ H ₆ , and the gas mixture ratio is such that the boron concentration in the p-type silicon crystalline semiconductor layer is 0.5 atomic%. It was adjusted. The frequency of the high frequency input was 40.68 MHz, the forming chamber pressure was 25 Pa, and the substrate temperature was 150 ° C. The film thickness was 20 nm.

The i-type silicon crystalline semiconductor layer 34b uses a mixed gas of SiH ₄ and H ₂ , and the gas mixing ratio is adjusted so that the i-type silicon layer is sufficiently crystallized. The frequency of the high frequency input was 40.68 MHz, the forming chamber pressure was 40 Pa, and the substrate temperature was 150 ° C. The film thickness was 2 μm. When an i-type silicon crystalline semiconductor layer is formed on a flat glass substrate using this condition, the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak are The ratio I ₂₂₀ / I ₁₁₁ is 10.0 as shown in FIG. 6, and the i-type silicon crystalline semiconductor layer 33 itself exhibits a strong (220) orientation in the normal direction of the flat surface. .

The n-type silicon crystalline semiconductor layer 34c uses a mixed gas of SiH ₄ , H ₂ and PH ₃ , and the gas mixture ratio is adjusted so that the phosphorus concentration in the n-type silicon semiconductor layer is 0.5 atomic%. The frequency of the high frequency input was 40.68 MHz, the forming chamber pressure was 25 Pa, and the substrate temperature was 150 ° C. The film thickness was 20 nm.

  Thereafter, the substrate was taken out from the plasma CVD apparatus, and zinc oxide was vapor-deposited by a normal electron beam vapor deposition method to form the back reflection layer 35. And it isolate | separated into the magnitude | size of 1 cm square by the laser scribing method. Thereafter, a back electrode 36 made of silver was formed by a normal electron beam evaporation method, and a super-straight structure multi-junction thin film solar cell in which light was incident from the substrate 31 side was obtained.

  FIG. 4 schematically shows an image obtained by observing a cross section in the vicinity of the intermediate layer of the multi-junction thin film solar cell with a transmission electron microscope. The p-type silicon crystalline semiconductor layer 42 of the second photoelectric conversion element formed on the uneven surface of the intermediate layer 41 is composed of a plurality of crystal grains, of which about 50% of the crystal grains are substantially perpendicular to the individual uneven surface. The remaining approximately 50% of the grains grew in a direction perpendicular to the surface of the intermediate layer (macroscopic surface). And most of the crystal grains growing in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) are columnar crystal grains having a length of 100 nm or more in the direction perpendicular to the surface of the intermediate layer (macroscopic surface). As the film thickness increased, the proportion in the portion increased while the particle size in the direction parallel to the surface of the intermediate layer (macroscopic surface) increased.

  Furthermore, in order to obtain information on the crystal orientation in a region substantially corresponding to the entire i-type layer, limited-field electron diffraction was performed with a limited-field aperture of 0.8 μm in diameter. From the obtained diffraction image, it was found that the i-type silicon crystalline semiconductor layer 43 had the (110) plane oriented parallel to the surface (macroscopic surface) of the intermediate layer. That is, in the crystal grains of the p-type silicon crystalline semiconductor layer 42 growing in a direction perpendicular to the surface of the intermediate layer (macroscopic surface), the (110) plane is oriented parallel to the surface of the intermediate layer (macroscopic surface). It has been found that the i-type silicon crystalline semiconductor layer 43 formed thereon has grown in such a manner that it inherits the crystal orientation of the p-type silicon crystalline semiconductor layer 42 serving as a base.

Further, in order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, the i-type silicon crystalline semiconductor layer of the second photoelectric conversion element is processed in the same process as this multi-junction thin film solar cell. When the X-ray diffraction was performed on the product formed up to the above, the ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the ( ₂₂₀ ) X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak was 7.0. Yes, the value was close to that when the film was formed on a flat glass substrate.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.205V in the cell area of 1 cm ^2, a short-circuit current 12.8 mA / cm ^2, shape Values of a factor of 0.695 and a photoelectric conversion efficiency of 10.72% were obtained.

  In this embodiment, all layers p, i, and n of the first photoelectric conversion element are formed of amorphous silicon. However, for the i layer, crystalline silicon that becomes the i layer of the second photoelectric conversion element. Any material having a larger energy gap may be used. For example, a-SiC: H may be used. The p-type and n-type layers need not be amorphous silicon layers, and both may be formed of a silicon crystalline semiconductor layer or the like. Moreover, although an intermediate layer is provided in this embodiment, it is not necessarily required in that the first conductive silicon crystalline semiconductor layer of the second photoelectric conversion element grown in a columnar shape in a direction perpendicular to the substrate surface is formed. .

(Comparative Example 1)
In order to verify the effect of the irregular shape, after forming tin oxide having a texture structure on a glass plate having a smooth surface, etching is performed by immersing in a 0.5% hydrochloric acid aqueous solution at a liquid temperature of 25 ° C. for 30 seconds. Then, the surface irregularities were sharpened and a zinc oxide layer having a thickness of 50 nm formed thereon by a normal electron beam evaporation method was used as a substrate. The ratio R / L between the unevenness size R and the unevenness interval L on the surface of the substrate, measured with an atomic force microscope, was 1.65. A thin-film solar cell was produced in the same manner as in Example 1 except for this substrate production process.

  FIG. 5 schematically shows an image obtained by observing a cross section of the thin film solar cell with a transmission electron microscope. However, as in FIG. 4, the n-type silicon semiconductor layer, the back reflecting layer, and the back electrode are omitted. A plurality of crystal grains are formed on the uneven surface, and most of them grow in a direction perpendicular to the individual uneven surface. The crystal grains growing in the direction perpendicular to the uneven surface eventually collide with the crystal grains grown from the adjacent uneven surface, and then change the growth direction in the direction perpendicular to the surface of the intermediate layer (macroscopic surface). It was gradually changing. However, no crystal grains remarkably growing in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) were confirmed.

  Furthermore, in order to obtain information on the crystal orientation in the region corresponding to the entire i-type layer, limited-field electron diffraction was performed with a limited-field aperture of 0.8 μm in diameter. From the obtained diffraction image, it was found that this i-type silicon crystalline semiconductor layer did not clearly have an orientation plane parallel to the surface of the intermediate layer (macroscopic surface). Thus, when the uneven shape is not appropriate, the crystal grains of the p-type silicon crystalline semiconductor layer do not have an orientation plane parallel to the surface (macroscopic surface) of the intermediate layer, and thus are formed on the surface. The i-type silicon crystalline semiconductor layer has also started to grow without having an orientation plane parallel to the surface (macroscopic surface) of the intermediate layer.

Further, in order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, X-ray diffraction was performed on an element formed up to the i-type layer by the same process as this thin film solar cell. ) The ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak is 2.0, and the orientation is remarkably higher than in the case of Example 1. It was inferior.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.080V in the cell area of 1 cm ^2, a short-circuit current 10.6mA / cm ^2, shape Values of a factor of 0.670 and a photoelectric conversion efficiency of 7.67% were obtained. Compared with the results of Example 1, the characteristics were deteriorated in all of the open circuit voltage, the short circuit current, and the form factor. Further, from the results of spectral sensitivity measurement for the elements of Example 1 and Comparative Example 1, it was found that the element of Comparative Example 1 was inferior in sensitivity especially on the long wavelength side. This means that there is a problem in the structure of the photoelectric conversion element located away from the light incident side, and can be explained without contradiction by the observation result of the transmission electron microscope.

(Comparative Example 2)
In order to verify the effect of the uneven shape, a multi-junction thin film solar cell was produced in the same manner as in Example 1 except that the zinc oxide layer 31c on the glass plate was formed to have a thickness of 500 nm. Compared with Example 1, the unevenness | corrugation shape of the surface became gentler by laminating | stacking the zinc oxide layer 31c thickly. As a result of measuring the microscopic uneven structure on the surface of the intermediate layer 33 with an atomic force microscope, the average value of the ratio R / L of the unevenness size R and the unevenness interval L was 0.05.

In order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, X-ray diffraction was performed on an i-type layer formed by the same process as this thin film solar cell. The ratio I ₂₂₀ / I ₁₁₁ of the integrated intensity I ₂₂₀ of the X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak is 8.5, which is the same value as when a film is formed on a flat glass substrate. Met.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.208V in the cell area of 1 cm ^2, a short-circuit current 10.6mA / cm ^2, shape Values of a factor of 0.690 and a photoelectric conversion efficiency of 8.84% were obtained. Compared with the results of Example 1, the open circuit voltage and the shape factor values were almost the same, but the short circuit current was reduced because the light confinement effect was small.

(Example 2)
In order to verify the effect of channeling particles, when the p-type silicon crystalline semiconductor layer 32 of the second photoelectric conversion element 34 is formed on the intermediate layer 33 having an uneven surface, SiH ₄ , H ₂ and B _{2 are used} as source gases. A mixture of H ₆ with Ar gas added was used. Other than that produced the thin film solar cell similarly to Example 1. FIG. At this time, the mixing ratio of SiH ₄ , H ₂ and B ₂ H _{6 was} the same as in Example 1, and the amount of Ar added was adjusted to be 5% of the total mixed gas.

  When the cross section of the obtained thin film solar cell was observed with a transmission electron microscope, a plurality of crystal grains were formed on the concavo-convex surface as in the case of Example 1. The grain density was reduced, and about 80% of the crystal grains grew in a direction perpendicular to the surface of the intermediate layer (macroscopic surface). And most of the crystal grains growing in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) are columnar crystal grains having a length of 100 nm or more in the direction perpendicular to the surface of the intermediate layer (macroscopic surface). As the film thickness increases, the particle size in the direction parallel to the surface of the intermediate layer (macroscopic surface) increases, and the tendency to increase the proportion in the portion is the same.

  Furthermore, in order to obtain information on the crystal orientation in a region substantially corresponding to the entire i-type layer, limited-field electron diffraction was performed with a limited-field aperture of 0.8 μm in diameter. From the obtained diffraction image, it was found that the (110) plane of this i-type silicon crystalline semiconductor layer was oriented parallel to the surface (macroscopic surface) of the intermediate layer. That is, the crystal grains of the p-type layer growing in a direction perpendicular to the surface of the intermediate layer (macroscopic surface) are (110) oriented parallel to the surface of the intermediate layer (macroscopic surface), and It was found that the formed i-type silicon crystalline semiconductor layer grew in a form that inherited the crystal orientation of the p-type layer serving as the base.

In addition, in order to quantitatively examine the degree of (110) orientation of the i-type silicon crystalline semiconductor layer, X-ray diffraction was performed on an i-type layer formed by the same process as this thin film solar cell. 220) The ratio I ₂₂₀ / I ₁₁₁ between the integrated intensity I ₂₂₀ of the X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) X-ray diffraction peak was 8.

The reason why <110> oriented crystal grains preferentially grow in the direction perpendicular to the surface of the intermediate layer (macroscopic surface) by Ar addition is not yet clear, but is almost perpendicular to the macroscopic surface due to the sheath potential. Ar ^{+ in} the plasma accelerated in the direction promotes growth to <110>, which is the channeling direction of silicon, and crystal grains in which <110> is not oriented in a direction substantially perpendicular to the macroscopic surface This is thought to be due to the effect of sputtering.

  In this embodiment, Ar is used as a rare gas to be added, but He or Ne may be used.

Current at AM1.5 (100mW / cm ²⁾ under the irradiation condition of the thin film solar cell - was measured for the voltage characteristic, the open-circuit voltage 1.242V in the cell area of 1 cm ^2, a short-circuit current 13.1mA / cm ^2, shape Values of a factor of 0.714 and a photoelectric conversion efficiency of 11.62% were obtained. Compared with the device of Example 1, the open circuit voltage and the shape factor were further improved. This is because the structure in the vicinity of the interface between the p-type layer and the i-type layer has become more appropriate, and can be explained without contradiction from the observation result of the transmission electron microscope.

  In the above embodiment, a super-straight type multi-junction thin film solar cell using a substrate in which a tin oxide film having a texture structure is formed on a glass plate having a smooth surface and a zinc oxide film is further formed thereon is provided. explained. However, it will be apparent to those skilled in the art that there are many embodiments according to the present invention other than the above-described embodiments. For example, the structure of the multi-junction thin film solar cell may be a substrate type in which light is incident from the side of a silicon crystalline semiconductor layer formed as a photoelectric conversion layer on a glass substrate or a metal substrate.

11 ... Uneven surface 12 ... Virtual surface (macroscopic surface) by visual observation
13... The first conductive silicon crystalline semiconductor layer 14 of the (N + 1) th photoelectric conversion element 14... The intrinsic silicon crystalline semiconductor layers 15 and 24 of the (N + 1) th photoelectric conversion element.
16, 25 ... Unevenness interval L
31 ... substrate 21, 31a ... glass plate 22, 31b ... tin oxide layer 31c ... zinc oxide layer 32 ... first photoelectric conversion element 32a ... p-type amorphous silicon layer 32b ... i-type amorphous silicon layer 32c ... N-type amorphous silicon layer 33, 41, 51 ... intermediate layer 34 ... second photoelectric conversion elements 34a, 42, 52 ... p-type silicon crystalline semiconductor layer 34b, 43, 53 ... i-type silicon crystalline semiconductor layer 34c ... N-type silicon semiconductor layer 35 ... Back surface reflection layer 36... Back electrode 61... I ₂₂₀ / I ₁₁₁ curve 62 with respect to substrate unevenness index R / L under conditions with few channeling particles. I ₂₂₀ / I ₁₁₁ curve

Claims (8)

A substrate, and a plurality of photoelectric conversion element structures stacked on the substrate,
In each of the plurality of photoelectric conversion element structures, a first conductive type semiconductor layer, an intrinsic semiconductor layer, and a second conductive type semiconductor layer are sequentially stacked,
Among the plurality of stacked photoelectric conversion element structures, the surface of the second conductive type semiconductor layer side of the Nth (N is an arbitrary integer greater than or equal to 1) photoelectric conversion element structure counted from the substrate side is an unevenness interval L. Having a texture structure with irregularities such that the ratio R / L of the irregularity size R to the range is 0.1 to 1.5, and N + 1th formed on the Nth photoelectric conversion element structure The first conductive semiconductor layer and the intrinsic semiconductor layer of the photoelectric conversion element structure are crystal grains grown in a columnar shape in a direction perpendicular to the second conductive semiconductor layer side surface of the Nth photoelectric conversion element structure A multi-junction thin film solar cell, which is a crystalline semiconductor layer mainly composed of A substrate, and a plurality of photoelectric conversion element structures stacked on the substrate,
In each of the plurality of photoelectric conversion element structures, a first conductive type semiconductor layer, an intrinsic semiconductor layer, and a second conductive type semiconductor layer are sequentially stacked,
Among the plurality of stacked photoelectric conversion element structures, between the Nth (N is an arbitrary integer of 1 or more) photoelectric conversion element structure and the N + 1th photoelectric conversion element structure counted from the substrate side, An intermediate layer is provided,
The surface of the intermediate layer that supports the (N + 1) th photoelectric conversion element structure has a textured structure with irregularities such that the ratio R / L of the irregularity size R to the irregularity interval L is in the range of 0.1 to 1.5. And the first conductive semiconductor layer and the intrinsic semiconductor layer of the (N + 1) th photoelectric conversion element structure are perpendicular to the surface of the intermediate layer supporting the (N + 1) th photoelectric conversion element structure. A multi-junction thin-film solar cell, which is a crystalline semiconductor layer mainly composed of crystal grains growing in a columnar shape. The first conductive crystalline semiconductor layer of the N + 1th photoelectric conversion element structure has a thickness of 1 nm or more and less than 200 nm,
In the portion from the surface of the first conductive semiconductor layer side of the N + 1th photoelectric conversion element structure to a height of 200 nm, the surface of the second conductive semiconductor layer side of the Nth photoelectric conversion element structure or the surface of the intermediate layer The crystal grains growing in a columnar shape in the vertical direction occupy at least 50% or more of the entire crystal grains in the portion, and the surface of the Nth photoelectric conversion element structure on the second conductive semiconductor layer side or the intermediate layer The multi-junction thin-film solar cell according to claim 1, wherein the multi-junction thin-film solar cell has a length of 100 nm or more in a direction perpendicular to the surface.   4. The multi-junction thin film solar cell according to claim 3, wherein the first conductive crystalline semiconductor layer of the N + 1th photoelectric conversion element structure has a thickness of 1 nm to 50 nm. The first conductive crystalline semiconductor layer and the intrinsic crystalline semiconductor layer of the N + 1th photoelectric conversion element structure are mainly made of silicon,
In the first conductive crystalline semiconductor layer and the intrinsic crystalline semiconductor layer of the (N + 1) th photoelectric conversion element structure, the surface of the Nth photoelectric conversion element structure on the second conductive semiconductor layer side or the surface of the intermediate layer crystal plane oriented parallel to the plane is predominantly (110), thin-film solar cell according to any one of claims 1 to 4. The integrated intensity I ₂₂₀ of the ( ₂₂₀ ) plane X-ray diffraction peak and the integrated intensity I ₁₁₁ of the (111) plane X-ray diffraction peak obtained for the intrinsic crystalline semiconductor layer of the N + 1th photoelectric conversion element structure The multi-junction thin film solar cell according to claim 5, wherein the ratio I ₂₂₀ / I ₁₁₁ is 3 or more.   The multi-junction thin film solar cell according to claim 5 or 6, wherein the first conductive crystalline semiconductor layer of the N + 1th photoelectric conversion element structure contains boron as an impurity.   The multi-junction thin film solar cell according to any one of claims 2 to 7, wherein the intermediate layer is a transparent conductor mainly made of zinc oxide.