DE112014002196B4 - Photovoltaic device - Google Patents

Abstract

The photovoltaic device has the following configurations: (1) the photovoltaic device has a light-absorbing component including a light-absorbing material capable of generating a charge carrier by absorbing monochromatic light of a wavelength λ0; (2) the photovoltaic device is provided with a band pass filter on a surface thereof on which the incident light is incident at an incident angle θ, and the bandpass filter has a function of wavelength-selective transmission of light having the wavelength λ0; and (3) the light absorbing component has a diffuse reflection surface on a back side thereof opposite to the surface on which the band pass filter is formed.

Description

FIELD OF THE INVENTION

The present invention relates to a photovoltaic device, more precisely, relates to a highly efficient photovoltaic device suitable for irradiation of monochromatic light.

BACKGROUND OF THE INVENTION

A system of monochromatizing sunlight to irradiate a photovoltaic material has been studied. Further, a combination of a laser beam (monochromatic light) and a photovoltaic device has been used as a power supply technology for a moving body and the like or in a high electromagnetic noise environment (Non-Patent Literature 1, 2).

An ordinary solar cell (for white light) is also used for such use of monochromatic light irradiation. In this case, only a solar cell using a material having a band gap slightly smaller than the photon energy of monochromatic light is selected, and no particular improvement has been made.

In a case of irradiating a white light solar cell with monochromatic light, an efficiency of converting monochromatic light into electric power is low. Meanwhile, if improvement specializing in monochromatic light irradiation is made, it is possible to improve efficiency. However, there has been no example of proposing a photovoltaic device suitable for monochromatic light irradiation except for band gap selection in the technical background.

[List of quotations]

[Non-patent literature]

• Non-patent literature 1: N. Kawashima, The Journal of the Institute of Electrical Engeneers of Japan, 129, 422 (2009)
• Non-patent literature 2: Y. Tanaka et al., Opt. Rev. 16, 257 (2009)

SUMMARY OF THE INVENTION

A problem to be solved by the present invention is to provide a highly efficient photovoltaic device capable of irradiating monochromatic light.

To solve the problem described above, a photovoltaic device according to claim 1 is provided.

A band pass filter has been used for extracting light of a certain wavelength from white light in a prior art. When the band pass filter is provided over a light receiving surface of a photovoltaic device and monochromatic light is made incident on the light receiving surface, the band pass filter reflects light reflected from a rear surface back onto an inner portion of a light absorbing component without intercepting incident light. Thereby, even if the light absorbing component is made thinner, a high degree of light absorption (a high light absorption ratio) is ensured. Further, a material can be reduced, a tact time can be shortened, and a cost can be reduced.

Further, a high light absorption coefficient and a low internal resistance can be made compatible with each other by downsizing and thinning the light absorbing component. As a result, even in a case of a large incident light intensity, resistance loss is restrained and high conversion efficiency (high efficiency) is achieved.

Figure list

• 1 Fig. 13 is a diagram showing an equivalent circuit of a solar cell;
• 2A and 2 B are graphs showing a concentration ratio dependency of conversion efficiency (an efficiency) when a Si solar cell ( 2A , Reference literature 1) and a III-V group compound solar cell ( 2 B , Reference literature 2) irradiated with artificial sunlight;
• 3 Fig. 13 is a graph showing a result of calculation of conversion efficiency (Reference Literature 3) when white light is incident on a CIGS microsolar cell (18 µm in diameter);
• 4A , 4B and 4C are graphs showing a relationship between an incident light intensity and conversion efficiency ( 4A) , a relationship among a thickness of a Si wafer, conversion efficiency, and a light absorption degree ( 4B) and a relationship between a voltage V and a current density J ( 4C ) show when monochromatic light having a wavelength of 1064 nm is incident on an ordinary Si solar cell having a surface texture and a back surface diffuse reflection structure;
• 5A , 5C and 5D provide an SEM image of an Ag electrode with a periodic concavo-convex structure ( 5C ), an SEM image of a portion of an amorphous Si solar cell using the same ( 5D ), and external quantum efficiency spectra of the solar cell ( 5A , Reference literature 5);
• 6th FIG. 3 is a schematic view of a region of a solar cell of an exemplary embodiment according to the claim and a direction of propagation of light; FIG.
• 7A and 7B FIG. 13 shows schematic views of a coated structure of a band pass filter. A total layer number is 2 (m ₁ + m ₂ ) + 1 in a case of FIG 7A and 2 (m ₁ + m ₂ ) + 2 in a case of 7B ;
• 8A until 8E show a result of calculation of transmission spectra for normally incident light;
• 9A and 9B represent a graph showing a relationship between a thickness of Si and a light absorbance ( 9A) and a relationship between an advancing angle 0 and a surface reflectance R _B (θ) ( 9B) indicates;
• 10A , 10B and 10C represent graphs showing a relationship between an incident light intensity and conversion efficiency ( 10A) , a relationship among a thickness of Si, conversion efficiency, and a light absorbance ( 10B) and a relationship between a voltage V and a current density J ( 10C ) show when monochromatic light with a wavelength of 1064 nm is incident on a Si solar cell coated with a band pass filter having a total of 14 layers (m ₁ = m ₂ = 3) on a surface thereof and a diffuse reflection structure on a rear surface thereof is trained;
• 11A until 11G are graphs showing relationships among Δλ / λ ₀ , Δθ, Δn _AR / n _AR0 , Δd _AR / d _AR0 , Δd _L4 / d _L40 , Δd _H4 / d _H40 or Δd _L2 / d _L20 and conversion efficiency; and
• 12A and 12B represent a graph showing a relationship between a total reflectance and conversion efficiency ( 12A) and a relationship between a diffuse reflectance / total reflectance and conversion efficiency ( 12B) indicates.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

A detailed explanation of an embodiment of the present invention is given as follows.

[1. Photovoltaic device]

• (1) die Photovoltaikvorrichtung weist eine lichtabsorbierende Komponente mit einem lichtabsorbierenden Material auf, das im Stande ist, einen Ladungsträger durch Absorbieren monochromatischen Lichts einer Wellenlänge λ₀ zu erzeugen, und ein im Wesentlichen paralleler Strahl mit dem monochromatischen Licht wird als einfallendes Licht verwendet;
• (2) die Photovoltaikvorrichtung ist mit einem Bandpassfilter an einer Fläche davon ausgebildet, auf die das einfallende Licht unter einem Einfallswinkel 0 einfällt, und der Bandpassfilter weist eine Funktion wellenlängenselektiver Transmission von Licht mit der Wellenlänge λ₀ auf; und
• (3) die lichtabsorbierende Komponente weist eine diffuse Reflexionsfläche an einer Rückfläche davon auf.

A photovoltaic device according to the present invention has the following configurations:

• (1) the photovoltaic device comprises a light absorbing component including a light absorbing material _{capable of generating a charge carrier by absorbing monochromatic light of a wavelength λ 0} , and a substantially parallel beam with the monochromatic light is used as incident light;
• (2) the photovoltaic device is formed with a band pass filter on a surface thereof on which the incident light is incident at an incident angle 0, and the band pass filter has a function of wavelength selective transmission of light having the wavelength λ ₀ ; and
• (3) the light absorbing component has a diffuse reflection surface on a rear surface thereof.

[1.1. Incident light]

In the present invention, incident light is formed by monochromatic light having a wavelength λ ₀ . That is, a photovoltaic device according to the present invention is a photovoltaic device for monochromatic light irradiation.

"Monochromatic light with a wavelength λ ₀ " refers to light in which a central wavelength thereof is λ ₀ and a ratio of half a half width at half maximum (Δλ / 2) of a spectrum to the central wavelength λ ₀ (= Δλ / 2λ ₀ ) is equal to or less than 0.04.

In the present invention, incident light is formed by substantially parallel rays which are substantially parallel incident on a light receiving surface (a surface on which incident light is incident).

The “substantially parallel rays” denote light in which a distribution β of an angle indicating a direction of propagation of light is equal to or smaller than 23 °. The smaller the β, the more preferable it is. Parallel rays of β ≈ 0 are preferable as incident light to achieve high conversion efficiency.

In the present invention, incident light is incident on a light receiving surface at an angle of incidence of zero.

The “incident angle θ” denotes an angle formed by a normal straight line direction of the light receiving surface and an incident direction of incident light.

The "direction of incidence of incident light" describes an average direction of propagation of incident light. For example, in a case where incident light is formed by a parallel beam (β = 0), the incident direction denotes a direction parallel to the incident light. Further, for example, in a case where a traveling direction of incident light is within a shape of a circular cone having an apex angle: 2β (β ≠ 0), an incident direction denotes a direction parallel to a rotation axis of the circular cone.

[1.2. Light-absorbing material, light-absorbing component]

A photovoltaic device has a light absorbing component formed by a light absorbing material capable of generating _{a charge carrier by absorbing monochromatic light of a wavelength λ 0.}

• (1) kristallines Si, mikrokristallines Si, amorphes Si;
• (2) sulfitbasierenden oder selenidbasierenden Verbindungshalbleiter von CuIn_1-xGa_xSe₂ (CIGS), Cu₂ZnSnS₄ (CZTS), Cu₂ZnSnSe₄ (CZTSe) und dergleichen; und
• (3) III-V-Gruppenverbindungshalbleiter von Ga_1-xIn_xAs_1-yP_y-basiertem Verbindungshalbleiter (0≤x≤1, 0≤y≤1), In_1-x-yAl_xGa_yAs-basiertem Verbindungshalbleiter (0≤x≤1, 0≤y≤1) und dergleichen.

In the present invention, a light absorbing material is not particularly limited, and various materials can be used therefor. As light-absorbing materials there are, for example:

• (1) crystalline Si, microcrystalline Si, amorphous Si;
• (2) sulfite-based or selenide-based compound semiconductors of CuIn _1-x Ga _x Se ₂ (CIGS), Cu ₂ ZnSnS ₄ (CZTS), Cu ₂ ZnSnSe ₄ (CZTSe) and the like; and
• (3) III-V group _{compound semiconductor of Ga 1-x} In _x As _1-y P _y -based compound semiconductor (0≤x≤1, 0≤y≤1), In _1-xy Al _x Ga _y As-based compound semiconductor (0 x 1, 0 y 1) and the like.

A shape or a size of a light absorbing component is not particularly limited and can be arbitrarily selected in accordance with an objective.

For example, a light absorbing component may be formed by a disk or a self-standing thick film or a thin film formed on a substrate.

Further, a photovoltaic device in which a component for photovoltaic conversion is formed only by a light absorbing component (e.g., Si solar cell) is acceptable, or a photovoltaic device in which a component for photovoltaic conversion is formed by a coating device of a light absorbing component (light absorbing layer) in of a thin film shape and another layer (for example, a thin film solar cell in which a light absorbing layer is formed by CIGS or CZTS) is acceptable.

A light absorbing component, particularly a Si wafer, in which a thickness is equal to or smaller than 100 µm and a size in an in-plane direction (direction in a plane) is equal to or smaller than 100 µm is preferable.

The thickness of a Si wafer is more preferably equal to or smaller than 50 μm, further preferably equal to or smaller than 20 μm.

The size in the in-plane direction of the Si wafer is more preferably equal to or smaller than 50 μm, and further preferably equal to or smaller than 20 μm.

Here, the “size in an in-plane direction” denotes a diameter of a minimal circumference that is circumscribed around a light-absorbing surface of a light-absorbing component.

[1.3. Bandpass filter]

[1.3.1. Definition]

A photovoltaic device according to the present invention is formed with a band pass filter on a surface on which the monochromatic light is incident at an incident angle θ (light receiving surface).

The “bandpass filter” describes a filter with a function of wavelength-selective transmission of light with a wavelength λ ₀ . The band pass filter is commonly used to extract light of a certain wavelength from white light. In the present invention, the band pass filter is used for transmitting incident light (monochromatic light) as it is and re-reflecting light reflected from a rear surface onto an inner portion of a light absorbing component. The point differs from the prior art.

The band pass filter is provided on one side of a light receiving surface of a photovoltaic device.

In a case of a photovoltaic device in which a component for photovoltaic conversion is formed only by a light absorbing component (e.g., a case of a Si solar cell), the band pass filter is formed on a surface of the light absorbing component.

Meanwhile, in a case of a photovoltaic device in which a component for photovoltaic conversion is formed by a coating device of a light absorbing layer and another layer (e.g., a case of a thin film solar cell), the band pass filter is formed on the uppermost surface of the coating device.

[1.3.2. Central wavelength λ]

A wavelength of light that can pass through a band pass filter has a prescribed width (transmission band).

The “central wavelength λ” denotes a median of a wavelength of light that can pass the bandpass filter.

“Wavelength-selective transmission of light with a wavelength λ ₀ ” means that it is not necessary that the central wavelength λ of the bandpass filter completely coincides with the central wavelength λ ₀ incident light coincides. However, when a difference between the wavelength λ and the wavelength λ _{0 is} large, a number of photons passing through the band pass filter is decreased and conversion efficiency is decreased.

In order to achieve the high conversion efficiency, the central wavelength λ of the band pass filter is preferably equal to or greater than 0.97 × λ ₀ and equal to or less than 1.039 × λ ₀ . The central wavelength λ is more preferably equal to or greater than 0.98 × λ ₀ and equal to or less than 1.025 × λ ₀ , more preferably equal to or greater than 0.99 × λ ₀ and equal to or less than 1.01 × λ ₀ .

[1.3.3. Structure]

1. (a) einen Bandpassfilter mit einer be-/geschichteten Struktur von (L4/H4/)^m1L2(H4/L4/)^m2;
2. (b) einen Bandpassfilter mit einer be-/geschichteten Struktur von AR/(L4/H4/)^m1 L2(H4/L4/)^m2;
3. (c) einen Bandpassfilter mit einer be-/geschichteten Struktur von (H4/L4/)^m1H2(L4/H4/)^m2;
4. (d) einen Bandpassfilter mit einer be-/geschichteten Struktur von AR/(H4/L4/)^m1H2(L4/H4/)^m2 und dergleichen.

A structure of a band pass filter is not particularly limited as long as the above-described function is achieved. As a band pass filter there are for example:

1. (a) a band pass filter having a coated structure of (L4 / H4 /) ^m1 L2 (H4 / L4 /) ^m2 ;
2. (b) a band pass filter having a coated structure of AR / (L4 / H4 /) ^m1 L2 (H4 / L4 /) ^m2 ;
3. (c) a band pass filter having a coated structure of (H4 / L4 /) ^m1 H2 (L4 / H4 /) ^m2 ;
4. (d) a band pass filter having a coated structure of AR / (H4 / L4 /) ^m1 H2 (L4 / H4 /) ^m2 and the like.

Whereby
"AR" is an anti-reflective layer formed by a material with a refractive index of n _AR ,
"L4" is a low refractive index layer (A) formed by a low refractive index material (A) having a refractive index of n _L4 ,
"L2" is a low refractive index layer (B) formed by a low refractive index material (B) having a refractive index of n _L2 ,
"H4" is a high refractive index layer (A) which is formed by a high refractive index material (A) with a refractive index of n _H4 (> n _L4 ,> n _L2 ),
“H2” is a high refractive index layer (B) formed by a high refractive index material (B) having a refractive index of n _H2 (> n _L4 ,> n _L2 ), and
m ₁ , m ₂ are integers equal to or greater than 1, respectively.

"L4" and "L2" each represent layers represents in which optical thicknesses thereof (= refractive index n × actual thickness d) in the ratio / is proportional to λ _0/4 and λ _0/2. Θ in a case of an incident angle is a condition of optical thickness nd of respective layers that form a band-pass filter, nd = (1 - sin ² θ / n ^{2) 1/2} × (λ _0/4 or λ _0/2). The same applies to "H4" and "H2".
"(L4 / H4 /) ^m1 " means that a coating unit of "L4 / H4 /" is repeated mi times. A repetition number m _{1 of} the former half and a repetition number m _{2 of} the latter half may be the same or different from each other.
"(L4 / H4 /) ^m1 L2 (H4 / L4 /) ^m2 " means that an L2 layer is between a coated / coated structure of (L4 / H4 /) ^m1 and a coated / coated structure of (H4 / L4 /) ^{m2 is} inserted.
“AR /” means that an anti-reflective layer is formed on the uppermost surface of a band pass filter (incident side of light).

The low refractive index material (A) forming L4 and the low refractive index material (B) forming L2 may be the same material or different materials as long as a condition described later is satisfied. Further, individual L4s included in a coated structure can be the same material or different materials as long as the condition described later is met.
Similarly, the high refractive index material (A) forming H4 and the high refractive index material (B) forming H2 may be the same material or different materials as long as the condition described later is satisfied. In addition, individual H4s included in a coated structure can be the same material or different materials as long as the condition described later is met.

A band pass filter provided with the coated structure (a) or the coated structure (b) is particularly preferable. The band-pass filters with the coated structures achieve higher conversion efficiency than that of a band-pass filter provided with another coated structure.

An ideal thickness (actual thickness) do is a thickness (actual thickness) d of each layer that forms a band-pass filter. It is not necessary that the thickness d of each layer be completely the same as the ideal thickness d ₀ . However, if a difference between the thickness d and the ideal thickness do is increased, conversion efficiency is decreased. Therefore, it is preferable that the thickness d be close to the ideal thickness do.

For example, in a case of a band pass filter provided with the coated structure (a) or the coated structure (b) described above, it is preferable that a thickness d _{L4 of} the low refractive index layer (A), a thickness d _{L2 of} the low refractive index layer (B) and a thickness d _{H4 of} the high refractive index layer (A) satisfy the following relationships, respectively.

That is, a thickness d _{L4 of} the low refractive index layer (A) is preferably equal to or greater than 0.75 × d _L40 and equal to or less than 1.20 × d _L40 . The thickness d _L4 is more preferably equal to or greater than 0.85 × d _L40 and equal to or less than 1.10 × d _L40 , and more preferably equal to or greater than 0.92 × d _L40 and equal to or less than 1.06 × d _L40 . d _L40 is an ideal thickness (actual thickness) of the low refractive index layer (A) and is represented by d _L40 = (1 - sin ² θ / n _L4 ^{2) 1/2} × (λ _0/4) × (1 / n _L4) shown .

Further, a thickness d _{L2 of} the low refractive index layer (B) is preferably equal to or greater than 0.926 × d _L20 and equal to or less than 1.058 × d _L20 . The thickness d _L2 is more preferably equal to or greater than 0.95 × d _L20 and equal to or less than 1.03 × d _L20 , and more preferably equal to or greater than 0.97 × d _L20 and equal to or less than 1.02 × d _L20 .
d _L20 is an ideal thickness (actual thickness) of the low refractive index layer (B), and is represented by d _L20 = (1 - sin ² θ / n _L2 ^{2) 1/2} × (λ _0/2) × (1 / n _L2) shown.

Furthermore, a thickness d _{H4 of} the high refractive index layer (A) is preferably equal to or greater than 0.87 × d _H40 and equal to or less than 1.10 × d _H40 . The thickness d _H4 is more preferably equal to or greater than 0.92 × d _H40 and equal to or less than 1.05 × d _H40 , and more preferably equal to or greater than 0.96 × d _H40 and equal to or less than 1.02 × d _H40 .
d _H40 is an ideal thickness (actual thickness) of the high refractive index layer (A) and is represented by d _H40 = (1 - sin ² θ / n _H4 ^{2) 1/2} × (λ _0/4) × (1 / n _H4) shown .

In the bandpass filter provided with the coated structure (b) or the coated structure (d), a refractive index n _AR and a thickness (actual thickness) d _{AR of} the anti-reflective layer are not specifically limited. That is, the antireflection layer can be a layer formed by a material that transmits at least light.

Further, in the band pass filter provided with the coated structure (b), in order to achieve higher conversion efficiency, it is preferable that the refractive index n _AR and the thickness d _{AR of} the anti-reflective layer satisfy the following relationships, respectively.

That is, the refractive index n _{AR of} the anti-reflective layer is preferably equal to or greater than 0.60 × n _pv ^1/2 . The refractive index n _AR is more preferably equal to or greater than 0.7 × n _pv ^1/2 and equal to or less than 1.8 × n _pv ^1/2 , and more preferably equal to or greater than 0.8 × n _pv ^1/2 and equal to or less than 1.5 × n _pv ^1/2 .
n _pv is a refractive index of the light absorbing material.

Furthermore, the thickness d _{AR of} the anti-reflective layer is preferably equal to or less than 1.70 × d _AR0 . The thickness d _AR is more preferably equal to or greater than 0.4 × d _AR0 and equal to or less than 1.4 × d _AR0 , and more preferably equal to or greater than 0.7 × d _AR0 and equal to or less than 1.2 × d _AR0 .
d _AR0 is the ideal thickness (actual thickness) of the anti-reflection layer and d is _AR0 = (1 - sin ² θ / n _AR ^{2) 1/2} × (λ _0/4) × (1 / n _AR) shown.

[1.3.4. Material]

Materials of respective layers that form a band pass filter are not specifically limited, and various materials can be used in accordance with a purpose.

As the low refractive index materials, there are, for example, MgF ₂ (refractive index at wavelength of 1064 nm: about 1.36), SiO ₂ (refractive index at wavelength of 1064 nm: about 1.43) and the like.

As the high refractive index materials, there are, for example, ZnS (refractive index at wavelength of 1064 nm: about 2.29), TiO ₂ (refractive index at wavelength of 1064 nm: about 2.25) and the like.

As materials of anti-reflective layers, there are, for example, Y ₂ O ₃ (refractive index at a wavelength of 1064 nm: approximately 1.73), a composite oxide of TiO ₂ -SiO _2, and the like.

[1.4. Diffuse reflection surface]

A light absorbing component is provided with a diffuse reflective surface on its rear surface. The diffuse reflective surface is for reflecting incident light that cannot be absorbed by a light-absorbing component. Although a structure of the diffuse reflection surface is not specifically limited, a fine concavo-convex structure formed on the rear surface of the light absorbing component is preferable. When a shape or a period of the concavo-convex structure is optimized, a reflectance is improved.

In order to achieve high conversion efficiency, a total reflectance of the diffuse reflecting surface is preferably equal to or greater than 93%. A total reflectance is more preferably equal to or greater than 95%, more preferably equal to or greater than 96% and even more preferably equal to or greater than 97%.

Here the "total reflectance (%)" denotes the sum of a specular reflectance and a diffuse reflectance.

The “specular reflectance (%)” refers to a ratio (a proportion) of light that is reflected as specular reflection when incident light is brought to 100.

The “diffuse reflectance (%)” refers to a ratio (a proportion) of reflected light excluding specular reflection when the incident light is made 100.

Furthermore, the diffuse reflecting surface with an index, which is represented by “diffuse reflectance × 100 / total reflectance”, equal to or greater than 49% is preferred.

The index represents a relative ratio (a relative proportion) of the diffuse reflectance. For example, in a case where an incident angle θ is 0 °, (normal incidence) passes light specularly reflected from the diffuse reflecting surface and from the light absorbing surface Component cannot be absorbed, put a band pass filter the way it is. Therefore, in a case where the diffuse reflectance is relatively low, high conversion efficiency is not obtained.

In contrast, when a shape of the diffuse reflection surface is optimized so that the index is equal to or greater than 49%, a possibility that light reflected by the diffuse reflection surface is reflected again by the band pass filter is high. As a result, a high conversion efficiency is achieved. The index is more preferably equal to or greater than 60% and further preferably equal to or greater than 70%.

[2. Effect]

The band pass filter has been used for extracting light of a certain wavelength from white light in a prior art. When the band pass filter is provided over a light receiving surface of a photovoltaic device and monochromatic light is irradiated on the light receiving surface, the band pass filter reflects light reflected from a rear surface back onto an inner portion of the light absorbing component without intercepting the incident light. Thereby, even if the light absorbing component is made thinner, a high degree of light absorption is ensured. The material can also be reduced, the cycle time can be shortened and costs can be reduced.

Further, a high light absorption coefficient and a low internal resistance can be made compatible with each other by downsizing and thinning the light absorbing component. As a result, even in a case where an incident light intensity is large, resistance loss is restrained and high conversion efficiency is achieved.

• (1) während für gewöhnlich ein Wafer einer Dicke von in etwa 200 µm verwendet wird, wird die Dicke auf in etwa 20 µm dünner gemacht und eine Größe in einer Inebenenrichtung wird ebenfalls auf mehrere 10 µm oder weniger miniaturisiert, wodurch Reihenwiderstand an dem inneren Abschnitt der Solarzelle verringert wird;
• (2) durch Verwenden normalen Einfalls monochromatischen Lichts und durch Ausbilden eines Bandpassfilters an einer Oberfläche davon, kann einfallendes Licht selbst durch einen dünnen Wafer von in etwa 20 µm ausreichend absorbiert werden; und
• (3) infolgedessen wird hohe Umwandlungseffizienz erzielt.
• Ferner wird eine Kombination eines Laserstrahls und einer Solarzelle als eine Leistungsversorgungstechnologie für einen sich bewegenden Körper oder in einer hohen elektromagnetischen Rauschumgebung verwendet. Das folgende Ergebnis wird auch für die Solarzelle, die dafür verwendet wird, angewendet.

A specific structure of a Si solar cell, which is combined with a solar-pumped laser, is designed as follows. It will be shown:

• (1) While a wafer having a thickness of about 200 µm is usually used, the thickness is made thinner to about 20 µm, and a size in an in-plane direction is also miniaturized to several tens of µm or less, thereby increasing series resistance at the inner portion the solar cell is decreased;
• (2) by using normal incidence monochromatic light and forming a band pass filter on a surface thereof, incident light can be sufficiently absorbed even by a thin wafer of about 20 µm; and
• (3) As a result, high conversion efficiency is achieved.
• Further, a combination of a laser beam and a solar cell is used as a power supply technology for a moving body or in a high electromagnetic noise environment. The following result is also applied to the solar cell used for it.

[2.1. Influence of incident light intensity]

1 shows an equivalent circuit of a solar cell. In 1 J, J _ph and V denote an output current density, a light-induced current density and an output voltage, respectively. Designations R _s and R _sh denote series resistance and parallel resistance, respectively. A relationship between the current density J and the voltage V is represented by formula (1).
[Numerical formula 1] $$J=J_{pH}-J_0\left\{\text{exp}\left(\frac{q\left(V+{\mathrm{R.}}_{s}J\right)}{n{k}_{\mathrm{B.}}{T}_{\mathrm{R.}T}}\right)-1\right\}-\frac{V+{\mathrm{R.}}_{s}J}{{\mathrm{R.}}_{sH}}$$

Notations Jph, Jo and n denote a light-induced current density, a reverse saturation current density and an ideality factor, respectively. Designations k _B and T _RT each denote the Boltzmann constant and a room temperature. Jo can be seen roughly as a constant value that does not depend on the incident light intensity or V. Therefore, as the incident light density is increased, influence of reverse current on Jph is relatively decreased, and thereby conversion efficiency is improved. However, a loss due to the series resistance _{R s is also increased at the same time.} Therefore, after the conversion efficiency is maximized, the conversion efficiency is in turn lowered.

2A and 2 B give data on concentration ratio dependency of conversion efficiency when artificial sunlight is applied to a Si solar cell ( 2A , Reference literature 1) and a III-V group compound solar cell ( 2 B , Reference literature 2) is irradiated. In a case of Si, the conversion efficiency is maximized most at 100-fold concentration and then gradually decreased. Also in a case of a III-V group compound, although the conversion efficiency is increased up to a concentration several hundred times, the conversion efficiency is decreased thereafter.

[Reference Literature 1] M. Castro et al., Solar Energy Mater. Solar Cells 92, 1697 (2008)

[Reference Literature 2] R. R. King et al., Prog. Photovolt .: Res. Appl. 20, 801 (2012)

A Si solar cell is usually formed by forming an n-type layer with a doping concentration of about 10 ²⁰ cm ^-3 and a thickness of about 0.3 μm on a surface of a p-type wafer with a doping concentration of about 10 ¹⁶ manufactured up to 10 ¹⁷ cm ^-3 . Since a thickness of the p-type wafer is 200 µm and a specific resistance is about 1 to 0.1 Ωcm, resistance in a thickness direction of the wafer poses a problem. A thickness of the n-type layer is as thin as about zero .3 µm, and therefore resistance in a thickness direction poses no problem. However, since a sheet resistance is 10 to 100 Ω, there is a concern that resistance in an in-plane direction affects efficiency depending on a quantity.

With regard to the resistance in the in-plane direction of the n-type layer, a microsolar cell has been proposed in which a diameter of the solar cell is limited to several tens of μm. It has been found that the conversion efficiency increases with an increase in the incident light intensity up to to the incident light intensity of 1 to 10 kW / cm ^{2 is} improved without being influenced by it ( 3 ) (Reference literature 3, 4). In 3 R denotes a sheet resistance of a solar cell.

In a case where sunlight is directly irradiated, it is not realistic to focus sunlight on a solar cell of about several tens of µm by using a lens and the like. However, in a case of a solar pumped laser using a light guide, connection therebetween is facilitated.

[Reference Literature 3] M. Paire et al., J. Appl. Phys. 108, 034907 (2010)

[Reference Literature 4] M. Paire et al., Appl. Pys. Lett. 98, 264102 (2011)

Meanwhile, in terms of resistance in a thickness direction of a p-type wafer, a resistance value is decreased by making the wafer thinner. However, since an incident light wavelength (1064 nm) is close to an absorption edge of Si, there is a concern that light cannot be sufficiently absorbed. Therefore, a relationship among a wafer thickness, a light absorption degree, and conversion efficiency and influence of resistance in a thickness direction have been studied using a solar cell simulator PC1D.

A texture (transmittance: 1) with a depth of 3 µm is set on a surface, and a diffuse reflection structure (reflectance: 1) is set on a rear surface similar to an ordinary solar cell. Although a p-type wafer having a specific resistance of about 1 Ωcm is used in a case where it is used by non-concentrated irradiation, high conversion efficiency can be achieved by having a smaller specific resistance of 0.1 to 0.2 Ωcm in one In the case of large incident light intensity can be achieved. Therefore a value of 0.1 Ωcm is used in the calculation. A sheet resistance of an n-type layer is set to 50 Ω, which is a common value. A volume recombination lifetime is set to 1000 microseconds, which is a typical value.

In a case of a thin wafer, a surface recombination rate significantly affects the conversion efficiency. A surface recombination rate in a case of passivation / neutralization by a prior art thermally oxidized film is 10 ⁵ to 10 ⁶ cm / s. In contrast, in recent years a value below 10 cm / s has been achieved by passivation / neutralization using Al ₂ O ₃ or i-type / intrinsic amorphous Si. Therefore a value of 10 cm / s is used in the calculation this time.

Contact resistance with an electrode is set to a value that becomes substantially zero. Parallel resistance is also set to 0. The other parameters use a default value of PC1D. Conversion efficiency is 22.5% when AM 1.5 G 1 sunlight irradiates a solar cell with a thickness of 200 µm under the condition.

4A , 4B and 4C FIG. 13 shows a calculation result of characteristics when an ordinary Si solar cell provided with a surface texture and a back surface diffuse reflection structure is irradiated with monochromatic light having a wavelength of 1064 nm. 4A Fig. 13 is a graph showing a relationship between incident light intensity and conversion efficiency. 4B Fig. 13 is a graph showing a relationship among a thickness of a Si wafer, conversion efficiency, and a light absorption degree. 4C FIG. 13 is a graph showing a relationship between a voltage V and a current density J. FIG. In 4C the current density J is normalized by a converted value of an incident photon number into a current density (when an incident light intensity is 10 mW / cm ² , 8.582 mA / cm ² ).

4A compares incident light intensity dependence on conversion efficiency in terms of second types of thicknesses. In a case of a thickness of 200 µm, the conversion ^{efficiency is maximized at an incident light intensity of 1 W / cm 2} and is decreased thereafter. Meanwhile, in a case of a thickness of 50 µm, since R _{s is} small, the conversion efficiency is increased to a higher intensity. However, light is not sufficiently absorbed and hence the efficiency remains a low value.

4B Fig. 10 shows a wafer thickness dependency of a light absorbance along with conversion efficiency. In a case of small incident light intensity, the conversion efficiency is a monotonically increasing function of a thickness similar to the light absorption coefficient.

Meanwhile, in a case of an incident light intensity of 1 kW / cm ^2, the conversion efficiency by the influence of series resistance is brought into a saturated tendency from about a thickness of 30 µm, maximized at 75 µm (30.3%) and thereafter decreased.

4C shows JV characteristics of solar cells with thicknesses of 200 µm and 50 µm. J is normalized by a converted value of an incident photon number into a current density (when incident light intensity is 10 mW / cm ² , 8.582 mA / cm ² ). From this, it is found that influence of series resistance is significant in a case of the incident light intensity of 1 kW / cm ² even if a thickness is 50 µm.

Therefore, in order to obtain high conversion efficiency even in a case of a large incident light intensity, it is first indispensable _{to decrease R s} by making a wafer thinner. However, in a case of a surface texture and a diffuse reflection structure on a back surface of a related art, light cannot be sufficiently absorbed, and therefore a more effective light trapping structure is needed.

[2.2. Light trapping effect of the band pass filter]

Recently, a number of light trapping structures have been proposed using plasmon resonance or photonic tape in addition to surface or back surface texture used in an ordinary solar cell. 5A , 5C and 5D show an example thereof (Reference Literature 5). In 5A 500 nm and 700 nm denote periods of a concavo-convex Ag structure. Also shows 5A Results of cases of a flat plate electrode (flat) and a transparent electrode having a texture made by Asahi Glass Corporation (Asahi) for comparison.

[Reference literature 5] V. E. Ferry et al., Opt. Express 18, A237 (2010)

In the case of an amorphous substrate type Si solar cell, an Ag electrode with periodic concavo-convex asperities is used. Although external quantum efficiency thereof is higher than that of a flat plate structure, the efficiency is not very different from that of a case of a textured transparent electrode manufactured by Asahi Glass Co., Ltd. which is used as a standard in a thin film Si solar cell. In this way, light trapping is not easy for sunlight, which is white light and whose direction of incidence is changed occasionally. Therefore, consider a method of light trapping using normal (perpendicular) monochromatic incidence of light, which is a hallmark of current use.

In a case where both a surface and a rear surface of a solar cell are smooth and correspondingly formed with ideal anti-reflection (surface) and reflection (rear surface) structures, light that is perpendicularly incident on a solar cell is reflected by a rear surface, in turn passes a surface and spreads outwards. Therefore, a light absorbance AA = T _F (1 - exp (-2αd)). Here, a transmittance T _{F of} the surface is 1 because an ideal anti-reflection structure of the surface is assumed. In a case where a thickness d of the solar cell is close to or smaller than the reciprocal of a light absorption coefficient α (in a case of Si 9.6528 cm ^-1 at a wavelength of 1064 nm), incident light is not sufficiently absorbed and part of it escapes to the outside.

wobei
ao eine relative Intensität bezeichnet, wenn das normal einfallende Licht die Oberfläche passiert und danach die Rückfläche erreicht,
a_ex eine relative Intensität einer Komponente von Licht bezeichnet, das durch die Rückfläche reflektiert wird und die Oberfläche passiert, und
ain eine relative Intensität einer Komponente von Licht bezeichnet, das von der Rückfläche reflektiert wird, von der Oberfläche reflektiert wird, und die Rückfläche wieder erreicht. Diese Parameter werden durch Formeln (3) bis (5) ausgedrückt.
[Numerische Formel 3] $$a_0=\text{exp}\left(-\alpha d\right)$$

$$a_{\text{ex}}=\frac{{\displaystyle {\int }_0^{{\theta }_c}d\theta \text{ sin}\theta \text{ cos}\theta \text{ exp}\left(-\frac{\alpha d}{\text{cos}\theta }\right)}}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{ sin}\theta \text{ cos}\theta }}$$

$$a_{\text{in}}=\frac{{\displaystyle {\int }_{{\theta }_c}^{\mathrm{\pi }/2}d\theta \text{ sin}\theta \text{ cos}\theta }\left(-\frac{2\alpha d}{\text{cos}\theta }\right)}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{ sin}\theta \text{ cos}\theta }}$$

Next, consider a case where a back surface is a diffuse reflection surface. In this case, a refractive index n _{pv of} a solar cell material (in a case of Si 3.55 at a wavelength of 1064 nm) is larger than that of air. Therefore, light that is reflected at an angle smaller than a critical angle of total reflection θ _c = arcsin (1 / n _pv ) and propagates to a surface (propagation angle 0) passes the surface again and propagates outwards. However, in a case of an angle greater than the critical angle, light is totally reflected on the surface and is directed again onto the rear surface (see 6th which shows an exemplary embodiment according to the claims). Therefore, A can be represented as in formula (2).
[Numerical formula 2] $$\begin{array}{l}\mathrm{A.}=T_{\mathrm{F.}}\left(1-a_0\left(a_{\text{ex}}+a_{\text{in}}\left(a_{\text{ex}}+a_{\text{in}}\left(a_{\text{ex}}+\cdots \right)\right)\right)\right)\\ =T_{\mathrm{F.}}\left(1+\frac{a_0{a}_{\text{ex}}}{1-a_{\text{in}}}\right)\end{array}$$

whereby
ao denotes a relative intensity when the normally incident light passes the surface and then reaches the rear surface,
a _ex denotes a relative intensity of a component of light reflected by the rear surface and passing through the surface, and
ain denotes a relative intensity of a component of light that is reflected from the rear surface, is reflected from the surface, and reaches the rear surface again. These parameters are expressed by formulas (3) to (5).
[Numerical formula 3] $$a_0=\text{exp}\left(-\alpha d\right)$$

$$a_{\text{ex}}=\frac{{\displaystyle {\int }_0^{{\theta }_c}d\theta \text{sin}\theta \text{cos}\theta \text{exp}\left(-\frac{\alpha d}{\text{cos}\theta }\right)}}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{sin}\theta \text{cos}\theta }}$$

$$a_{\text{in}}=\frac{{\displaystyle {\int }_{{\theta }_c}^{\mathrm{\pi }/2}d\theta \text{sin}\theta \text{cos}\theta }\left(-\frac{2\alpha d}{\text{cos}\theta }\right)}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{sin}\theta \text{cos}\theta }}$$

For extreme values αd << 1, n _pv >> 1, A becomes formula (6), and the Yablonovitch limit, at which an effective optical path length is increased by 4n _pv ² , is derived.
[Numerical formula 4] $$\mathrm{A.}\cong T_{\mathrm{F.}}\left(1-\mathrm{<figure-callout\; id="2"\; label="4th\; n\; PV"\; filenames="DE112014002196B4\_0013.png,DE112014002196B4\_0015.png"\; state="\{\{state\}\}">4th</figure-callout>}{n}_{\text{PV}}^{}{}_2^{}\alpha d\right)$$

$$a_{\text{in}}=\frac{{\displaystyle {\int }_0^{{\theta }_c}d\theta \text{ sin}\theta \text{ cos}\theta \text{ exp}\left(-\frac{2\alpha d}{\text{cos}\theta }\right)+{\displaystyle {\int }_{{\theta }_c}^{\mathrm{\pi }/2}d\theta \text{ sin}\theta \text{ cos}\theta }\left(-\frac{\alpha d}{\text{cos}\theta }\right)}}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{ sin}\theta \text{ cos}\theta }}$$

Now the direction of incidence of light is limited to normal (perpendicular). Therefore, if the surface is subjected to improvements so that among the light reflected by the rear surface and directed back onto the surface, a component thereof which is normally propagating is transmitted and a component thereof which is in a oblique direction propagates, is reflected, propagation of light from inside the solar cell to the outside is restricted, and thereby A is enlarged. If a reflectance of light directed to the surface from inside the solar cell is defined as R _B (θ) (R _B (θ> θ _c ) = 1), a _ex , a _{in become} as in formula (7) and formula (8).
[Numerical formula 5] $$a_{\text{ex}}=\frac{{\displaystyle {\int }_0^{{\theta }_c}d\theta \text{sin}\theta \text{cos}\theta \left(1-{\mathrm{R.}}_{\mathrm{B.}}\left(\theta \right)\right)\text{exp}\left(-\frac{\alpha d}{\text{cos}\theta }\right)}}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{sin}\theta \text{cos}\theta }}$$

$$a_{\text{in}}=\frac{{\displaystyle {\int }_0^{{\theta }_c}d\theta \text{sin}\theta \text{cos}\theta \text{exp}\left(-\frac{2\alpha d}{\text{cos}\theta }\right)+{\displaystyle {\int }_{{\theta }_c}^{\mathrm{\pi }/2}d\theta \text{sin}\theta \text{cos}\theta }\left(-\frac{\alpha d}{\text{cos}\theta }\right)}}{{\displaystyle {\int }_0^{\mathrm{\pi }/2}d\theta \text{sin}\theta \text{cos}\theta }}$$

Since a value close to 1 is _{required for the transmittance T F of} the normally incident light, R _B (θ = 0) ≈ 1 is indispensable. That is, it is inevitable that a component of light which is reflected from the back surface of the solar cell and which is propagated in a direction close to the perpendicular is propagated outward. Meanwhile, a phase shift of a direction perpendicular to the surface of light traveling in an oblique direction is equivalent to that of light having a longer wavelength. Therefore, a filter which transmits normal incident light having a wavelength λ ₀ and reflects normal incident light having a longer wavelength, that is, a band pass filter or a long wavelength cut filter, can be formed on the surface of the solar cell.

In a case of using the technology for incidence of sunlight (white light), T _F ≈ 1 is required as described above. Hence, a wavelength that can be cut off is only part of it longer than the absorption edge of a solar cell material. Therefore, since the light trapping effect by the long wavelength cut filter is limited to light in a vicinity of the absorption edge, its effect is insignificant.

[2.3. Design of the bandpass filter]

An optical filter having various functions can be formed by alternately coating thin films of two kinds of materials having different refractive indexes. Assume that ZnS (refractive index: 2.29) and MgF ₂ (refractive index: 1.36) are used for a high refractive index material and a low refractive index material, respectively. A refractive index of Si, which is a solar cell material, is n _pv = 3.55 as described above. Consider optimal structures of a band pass filter and a long wavelength cut filter under these conditions.

A basic structure of a bandpass filter by inserting a film having an optical thickness (a value of a thickness having a refractive index of the material multiplied) formed of λ _0/2 between alternately coated films of high / low refractive index materials, a thickness of λ _0/4 . In a case where the arrangement is formed over a glass substrate having a refractive index of 1.5, it is known that a transmission window is formed in a lamination order (coating order) of (H4 / L4 /) ^m1 H2 / (L4 / H4 /) ^m2 / glass substrate, as in 7A shown is narrower than in a coating order of (L4 / H4 /) ^m1 L2 / (H4 / L4 /) ^m2 / glass substrate. Press Here L4 and H4 corresponding optical layer thicknesses λ _0/4 from Niedrigbrechungsindex- and high refractive index materials. The notation (L4 / H4 /) ^m expresses that L4 / H4 / s are repeatedly layered m times.

8A Fig. 13 shows a result of calculation of a transmission spectrum for normally incident light of the layered structure using a transfer matrix method. Here, wavelength dependencies of refractive indices of respective materials are ignored, and values at a wavelength of 1064 nm are used as they are.

In a case of forming them on a surface of a solar cell, since the refractive index of the solar cell material is large although a function of a band pass filter is obtained, a peak value of a transmittance becomes low ( 8B) . Therefore, it is preferable to form an anti-reflection (AR) layer on the top surface. A reflectance can be made zero if a layer with a refractive index n _AR = n _s ^1/2 and a thickness of λ ₀ / n _AR / 4 is formed over a substrate with a refractive index n _s .

In the present case, the anti-reflective layer is formed over the band pass layer, also in this case it is confirmed that a reflectance can be made zero when a layer having a refractive index n _AR = n _pv ^1/2 = 1.884 and a thickness of λ ₀ / n _AR / 4 is used. The value of the refractive index can be realized _{by using a mixed oxide of TiO 2 -SiO 2} , ZrO ₂ -SiO ₂ , Nb ₂ O ₃ -SiO _2, and the like. In this case, as in 8C and 8D is shown, a bandwidth in a layering order AR / (L4 / H4 /) ^m1 L2 / (H4 / L4 /) ^m2 / solar cell narrower ( 7B) .

Also, a basic structure of a long wavelength cut filter is repeatedly layered layers of (H4 / L4 /). However, when only (H4 / L4 /) is used, ripple that has occurred on a transmission belt is significant. Therefore, layers of optical thickness of λ _0/8 usually formed on the surface and an interface between the laminated layers and the substrate. It is confirmed in this case also by performing similar calculation that a structure of L8 / (H4 / L4 /) ^m H8 / solar cell included in 8E shown has the highest performance. However, it is found that its light trapping effect is smaller than that of a bandpass filter, since a change in transmission / reflection is more gradual / slower than that of a bandpass filter ( 8D ) with the same number of layers.

[Examples]

[Example 1]

A characteristic value if a bandpass filter with a structure of 7B provided on a surface of a Si solar cell has been studied. 9A Fig. 13 shows a result of calculation of a light absorbance A as a function of a thickness of Si. A result on a surface texture structure and an ideal anti-reflection structure which is commonly used (see [2.1.]) Is also shown for comparison.

Even in a case where a total number of layers 6th is, a significant light trapping effect occurs, and a light absorption rate is increased as compared with an ordinary solar cell. Furthermore, the effect is further intensified with an increase in the total number of layers. As in 9B as shown, when a total number of layers is increased, a surface reflectance R _B (θ) of light reflected from a back surface of a solar cell becomes a value close to 1 as θ is smaller. That is, it is synonymous with the fact that a critical angle θ _c of total internal reflection is decreased.

As described in [2.1.], In order to achieve high conversion efficiency even in a case of a large light intensity, it is indispensable to make a Si wafer thinner. It is found that when a band pass filter with a total number of layers of 14th is formed on a surface, a light absorption rate of 92% is achieved even through a thickness of 20 µm. However, even if the number of layers is further increased, their effect is insignificant.

10A , 10B and 10C FIG. 13 shows a result of calculation of a characteristic of a solar cell using PC1D similar to [2.1.] using a value of the light absorption coefficient shown in FIG 9A is shown. In a case of a thickness of 50 µm, conversion efficiency becomes higher than that of an ordinary Si solar cell ( 4th ) described above is achieved by a light trapping effect.

The conversion efficiency is increased with an increase in the incident light intensity up to the incident light intensity of 100 W / cm ² similarly to the case of the ordinary Si solar cell described above. In a case of the thickness of 20 µm, since a series resistance R _{s of} the solar cell is small, the conversion efficiency is maximized with a larger incident light intensity (1 kW / cm ^2). Further, when the incident light intensity is equal to or greater than 1 kW / cm ² , a value higher than that of 50 µm is obtained.

How out 10B is known, in a case of weak incident light intensity, conversion efficiency is dominated by a light absorbance and is increased with an increase in Si thickness. Meanwhile, when the incident light intensity is 1 kW / cm ² , the light absorbance becomes high with an increase in Si thickness. However, since influence of R _{s is} increased, conversion efficiency is maximized at a thickness of 20 µm. A value of 59.5% on this occasion is a value substantially twice as much as the highest value 30.3% (75 µm) of an ordinary Si solar cell. The influence of R _s is also shown in a JV characteristic curve, which is shown in 10C is shown, seen.

Such a thin solar cell can be made by smart cut technology in which the thin solar cell is peeled off after hydrogen ions are implanted to embrittle a Si wafer or sliver technology in which a thin piece in a strip shape after forming p ⁺ / n ⁺ layers and an electrode each cut out on both surfaces of a thin Si plate.

[Example 2]

Influences of fluctuations in various parameters that affect the conversion efficiency have been studied. The incident light intensity is made 1 kW / cm ² and the Si thickness is made 20 µm, at which the conversion efficiency is maximized. A total number of layers is made 6 (m ₁ = m ₂ = 1).

11A until 11G show a result. In 11A until 11G thin lines (conversion efficiency: 30.3%) show a result of an Si thickness of 75 µm in which the conversion efficiency is maximized in a case of forming an ordinary surface texture. The following are made from 11A until 11G found:

• (1) when Δλ / λ ₀ -3.0% to + 3.9%, the conversion efficiency is equal to or greater than 30% ( 11A) . It will be out 11A found that even if a wavelength λ _{0 of} incident light (or a central wavelength λ of light which can pass a band pass filter) deviates slightly from an ideal value, as long as the deviation is within a prescribed range, high conversion efficiency is obtained;
• (2) when Δθ is equal to or less than 23 °, the conversion efficiency is equal to or greater than 30% ( 11B) . It will be out 11B found that even in a case where incident light is not a normal incident parallel beam, as long as a deviation of an incident angle is 0 or a deviation from the parallel beam is within a prescribed range, high conversion efficiency is obtained;
• (3) when Δn _AR / n _AR0 = 0, the conversion efficiency is maximized ( 11C ). It will be out 11C found that even if n _AR deviates from an ideal value (= n _pv ^1/2 ), influence of the deviation, which affects the conversion efficiency, is insignificant. Further, it is found that, in order to achieve high conversion efficiency, Δn _AR / n AR0 preferably falls within a range of -30% to _+80%;
• (4) when Δd _AR / d _AP0 = 0, the conversion efficiency is maximized ( 11D ). It will be out 11D found that even if d _AR deviates from an ideal value, influence of the deviation, which affects the conversion efficiency, is insignificant. Further, it is found that, in order to achieve high conversion _{efficiency, Ad AR} / d AR0 preferably falls within a range of -60% to +40%; and
• (5) when Δd _L4 / d _L40 falls within a range of -25% to +20%, the conversion efficiency is equal to or greater than 30% ( 11E) . When Δd _H4 / d _{H40 falls} within a range of -13% to +10%, the conversion efficiency is equal to or greater than 30% ( 11F) . Further, when Δd _L2 / d _{LZ0 falls} within a range of -7.4 to +5.8%, the conversion efficiency is equal to or greater than 30% ( 11G) . It will be out 11E until 11G found that even if thicknesses of respective layers constituting a band-pass filter deviate slightly from ideal values, the high conversion efficiency is achieved as long as deviations fall within prescribed ranges.

[Example 3]

Influence of a function of a diffuse reflection layer affecting conversion efficiency has been studied. Incident light intensity is made 1 kW / cm ² , and a Si thickness is made 20 µm at which the conversion efficiency is maximized. A total number of layers is made 14 (m ₁ = m ₂ = 3).

12A and 12B show a result of it. In 12A and 12B thin lines (conversion efficiency: 30.3%) show a result of an Si thickness of 75 µm in which the conversion efficiency is maximized in a case of forming an ordinary surface texture. The following are off 12A and 12B known:

• (1) when the total reflectance is equal to or greater than 0.93, the conversion efficiency is equal to or greater than 30%; and
• (2) when an index represented by diffuse reflectance / total reflectance is equal to or greater than 0.49, the conversion efficiency is equal to or greater than 30%.

Although the detailed explanation has been given of the embodiments of the present invention as described above, the present invention is in no way limited by the above-described embodiments.

The photovoltaic device according to the present invention can be used for a solar cell, a photoresistor, a photodiode, a phototransistor, and the like.

Claims (10)

A photovoltaic device having the following configurations: (1) the photovoltaic device has a light-absorbing component comprising a light-absorbing material capable of generating _{a charge carrier by absorbing monochromatic light of a wavelength λ 0;} (2) the photovoltaic device is formed with a band pass filter on a surface thereof on which the incident light is incident at an incident angle θ, and the band pass filter has a function of wavelength selective transmission of light having the wavelength λ ₀ ; and (3) the light absorbing component has a diffuse reflection surface on a back side thereof opposite to the surface on which the band pass filter is formed. Photovoltaic device according to Claim 1 at which a central wavelength λ of light capable of passing the band pass filter is equal to or greater than 0.97 × λ ₀ and equal to or less than 1.039 × λ ₀ . Photovoltaic device according to Claim 1 or 2 , also with the following configurations: (4) the bandpass filter has a layered structure of (L4 / H4 /) ^m1 L2 (H4 / L4 /) ^m2 , in which, “L4” is a first low refractive index layer with a low refractive index material with a refractive index of n _L4 , “L2” is a second low refractive index _{layer with a low refractive index material with a refractive index of n L2} , “H4” is a high refractive index layer with a high refractive index material with a refractive index of n _H4 (> n is _L4 ,> n _L2 ), and m ₁ and m _{2 are} each integers equal to or greater than 1; and (5) a thickness d _{L4 of} the first low refractive index layer is equal to or greater than 0.75 × d _L40 and equal to or less than 1.20 × d _L40 , a thickness d _{L2 of} the second low refractive index layer is equal to or greater than 0.926 × d _L20 and equal to or less than 1.058 × d _L20 , and a thickness d _{H4 of} the high refractive index layer is equal to or greater than 0.87 × d _H40 and equal to or less than 1.10 × d _H40 , in which $${\text{d}}_{\text{L40}}={\left(1-{\text{sin}}^2\mathrm{\theta }/{\text{n}}_{\text{L4}}{}^2\right)}^{1/2}\times \left({\mathrm{\lambda }}_0/\mathrm{4th}\right)\times \left(1/{\text{n}}_{\text{L4}}\right)\text{,}$$

$${\text{d}}_{\text{L20}}={\left(1-{\text{sin}}^2\mathrm{\theta }/{\text{n}}_{\text{L2}}{}^2\right)}^{1/2}\times \left({\mathrm{\lambda }}_0/2\right)\times \left(1/{\text{n}}_{\text{L2}}\right)\text{,}$$

and $${\text{d}}_{\text{H40}}={\left(1-{\text{sin}}^2\mathrm{\theta }/{\text{n}}_{\text{H4}}{}^2\right)}^{1/2}\times \left({\mathrm{\lambda }}_0/\mathrm{4th}\right)\times \left(1/{\text{n}}_{\text{H4}}\right)\text{,}$$

Photovoltaic device according to Claim 3 wherein the band pass filter further has an anti-reflective (AR) layer on the top surface thereof. Photovoltaic device according to Claim 4 , in which a refractive index n _{AR of} the anti-reflective layer is equal to or greater than 0.60 × n _pv ^1/2 , and a thickness d _{AR of} the anti-reflective layer is equal to or smaller than 1.70 × d _AR0 , in which n _{pv is} a refractive index des light absorbing material, and d _AR0 = (1 - sin ² θ / n _AR ^{2) 1/2} × (λ _0/4) × (1 / n _AR). Photovoltaic device according to one of the Claims 1 until 5 wherein the light absorbing component comprises a Si wafer with a thickness equal to or less than 100 μm and a size in a direction in the plane of the Si wafer equal to or less than 100 μm. Photovoltaic device according to one of the Claims 1 until 6th , for which a total reflectance of the diffuse reflecting surface is equal to or greater than 93%. Photovoltaic device according to one of the Claims 1 until 7th where the diffuse reflecting surface has an index represented by “diffuse reflectance × 100 / total reflectance” that is equal to or greater than 49%. Use of the photovoltaic device according to one of the preceding claims, in which the monochromatic light for converting the monochromatic light into electric power in the light absorbing component is irradiated toward the band pass filter, a substantially parallel beam with the monochromatic light is used as the incident light, and the substantially parallel beam is light in which a distribution β of an angle indicating a traveling direction of light is equal to or smaller than 23 °. Use after Claim 9 at which the angle of incidence θ is 0 ° ± 23 °.

Images