JPH11214726A - Stacked solar cell - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve conversion efficiency by stacking a solar cell constituted of a semiconductor having, a band gap larger than that of a nitrogen mixed crystal compound semiconductor material on the solar cell constituted of the nitrogen mixed crystal compound semiconductor material containing the specified rate of nitrogen by V group mixed crystal ratio. SOLUTION: A GaInNAs solar cell 2 is formed of a base layer 4 made of n-type GaInNAs and an emitter layer 5 made of p-type GaInNAs on an n-type GaAs substrate 1. A GaAs solar cell 3 is formed of a base layer 7 made of n-type GaAs, an emitter layer 8 made of p-type GaAs, a window layer 9 made of p-type GaAs and a p-type GaAs contact layer 10 via a tunnel junction layer 6 constituted of p<++> -type GaAs and n<++> -type GaAs so as to form a two junction stacked solar battery. When a band gap is set to be not more than 0.7-0.8 eV, so that a V group mixed crystal ratio is set to be not less than 0.1% and GaInNAs composition is matched with GaAs, conversion efficiency can be improved.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a low-cost and high-conversion stacked solar cell using a nitrogen mixed crystal compound semiconductor having sensitivity in a long wavelength region.

[0002]

2. Description of the Related Art The conversion efficiency of a solar cell currently in practical use is as follows: AM 1.5, 1 sun (non-light-collecting).
The i-junction solar cell accounts for about 18%, and the GaAs single-junction solar cell accounts for about 23%. In order to convert dilute solar energy into electric power, a solar cell having a large area is required. Therefore, improving the conversion efficiency of the solar cell is one of the most important issues.

Therefore, as means for increasing the efficiency of a solar cell, a stacked solar cell of a type in which solar cells having different band gaps are stacked in order to efficiently convert sunlight as continuous light has been proposed.

[0004] Among them, for example, M. Ohmori et
al. ： Technical Digest of
the International PVSEC-
9,525 pp, gallium phosphide, indium shown in such 1996 _{(Ga z In 1-z P} : 0.4 <z <0.6)
It has been clarified that a two-junction stacked solar cell obtained by stacking GaAs and gallium arsenide (GaAs) can achieve a conversion efficiency of about 30%, and is attracting attention.

This structure has a structure in which GaAs is formed on a GaAs substrate.
A GaAs solar cell consisting of a pn junction is formed, and a GaInP solar cell consisting of a pn junction of GaInP lattice-matched to GaAs is formed thereon, and power is taken out by connecting the two solar cells with a tunnel junction. High conversion efficiency can be achieved with a simple configuration having only two terminals. However, the theoretical conversion efficiency limit of this two-junction type stacked structure is about 32%, and it has been found that it is difficult to realize a higher conversion efficiency than that.

[0006]

The above-mentioned GaInP / G
In the aAs stacked solar cell, the wavelength range of light that can be photoelectrically converted is limited to the range of 300 nm to 870 nm.
There is a disadvantage that light in the far infrared region in the range of 00 nm cannot be used effectively. Due to this drawback, there is a limit on the theoretical conversion efficiency, and an ultra-high efficiency of 32% or more cannot be expected.

In order to compensate for this drawback, a solar cell using a semiconductor smaller than the band gap of GaAs is formed, and a GaAs solar cell is provided thereon.
Preferably, light in the range of wavelengths greater than 70 nm can be converted.

However, in the prior art, since there is no material having a small band gap that lattice-matches with the GaAs substrate, Si or InGaAs having poor matching with the GaAs substrate does not exist.
After a solar cell using a material with a small band gap such as s was separately formed on a Si substrate or an InP substrate,
I had to take the configuration of mechanical stacking.

Although this mechanical cell stacking technique is important in terms of increasing the conversion efficiency, it not only causes an increase in the number of power extraction terminals but also complicates the system, and also increases the cost of producing solar cells. Therefore, it has been difficult to realize an ultrahigh-efficiency solar cell suitable for industrial use.

[0010]

SUMMARY OF THE INVENTION The present invention provides a method for manufacturing a semiconductor device comprising a nitrogen mixed crystal compound semiconductor material containing nitrogen in a group V mixed crystal ratio of 0.1% or more. By stacking solar cells made of a semiconductor material having a band gap larger than that of the material, the above-mentioned problems are solved at once, and an ultra-high-efficiency solar cell of 32% or more is realized at low cost.

Here, the present invention relates to gallium indium arsenide nitride (Ga _x In ₁ -xN _y As _{1 -y} : 0 ≦ x ≦ 1, 0.00
It has a structure in which a solar cell made of gallium arsenide (GaAs) is stacked on a solar cell made of 1 ≦ y <0.2).

In addition, the present invention is characterized in that a structure in which a solar cell made of gallium arsenide is stacked on a solar cell made of gallium indium arsenide nitride is formed on a GaAs substrate.

In addition, the present invention is characterized in that a structure in which a solar cell made of gallium arsenide is laminated on a solar cell made of indium gallium arsenide is formed on a Ge substrate.

Further, the present invention provides a solar cell comprising a nitrogen mixed crystal compound semiconductor material having a band gap in a range of 0.7 eV to 0.8 eV on a GaAs substrate or a Ge substrate. And a stacked solar cell having a structure in which a solar cell made of a semiconductor material having a wider band gap than the nitrogen mixed crystal compound semiconductor material is stacked thereon.

[0015] The present invention arsenide gallium indium nitride _{(Ga x In 1-x N} y As 1-y: 0 ≦ x ≦ 1,0.001
A solar cell made of gallium arsenide (GaAs) is laminated on a solar cell made of ≦ y <0.2, and further a gallium indium phosphide (Ga _z
In _1-z P: 0.4 <z <0.6.

[0016] The present invention arsenide gallium indium nitride _{(Ga x In 1-x N} y As 1-y: 0 ≦ x ≦ 1,0.001
≦ y <0.2), a solar cell made of gallium arsenide (GaAs) is stacked, and further thereon gallium aluminum arsenide (Al _u)
It has a structure in which solar cells composed of Ga _1-u As: 0.2 <u <0.4) are stacked.

In addition, the present invention relates to gallium indium arsenide arsenide (Ga _x In _1-x N _y As _1-y : 0 ≦ x ≦ 1, 0.00
A solar cell made of gallium arsenide (GaAs) is stacked on a solar cell made of 1 ≦ y <0.2, and further thereon, gallium indium phosphide (G) is formed.
_{a z In 1-z P:} 0.4 <z <0.6), or gallium aluminum arsenide _{(Al u Ga 1-u As} : 0.2 <u
A structure in which a solar cell constituted by <0.4) is laminated on a GaAs substrate.

In addition, the present invention relates to gallium indium arsenide arsenide (Ga _x In _1-x N _y As _1-y : 0 ≦ x ≦ 1, 0.00
A solar cell made of gallium arsenide (GaAs) is stacked on a solar cell made of 1 ≦ y <0.2, and further thereon, gallium indium phosphide (G) is formed.
_{a z In 1-z P:} 0.4 <z <0.6), or gallium aluminum arsenide _{(Al u Ga 1-u As} : 0.2 <u <
A structure in which a solar cell constituted by 0.4) is laminated on a Ge substrate.

Further, the present invention provides a gallium indium arsenide (Ga _x In ₁ -xN _y ) having a band gap in a range of 0.75 eV to 1.0 eV on a GaAs substrate or a Ge substrate. As _1-y : 0 ≦ x ≦
On the solar cell composed of 1,0.001 ≦ y <0.2), by laminating a solar cell composed of gallium arsenide (GaAs), on which further, gallium phosphide, indium (Ga _z an In _1-z P: 0.4 <z <0.6) or gallium aluminum arsenide (Al _u Ga _1-u As:
The present invention relates to a stacked solar cell having a structure in which solar cells constituted by 0.2 <u <0.4) are stacked.

According to the present invention, for example, by forming a solar cell using the present nitrogen mixed crystal compound semiconductor on a GaAs substrate and continuously forming a GaAs solar cell, a wavelength larger than 870 nm, which has been difficult in the prior art, is obtained. Can be photoelectrically converted. Further, according to the present invention,
By using Ge having a lattice constant close to that of GaAs instead of the GaAs substrate, it is possible to use a substrate that is lower in cost than GaAs. The lattice constant of GaAs is 5.6533 °, and the lattice constant of Ge is 5.6.
579 °. The cost of the Ge substrate is GaAs.
It is about 1/4 to 1/6 of the cost of the substrate.

Applied physics: 148 pages, by Masahiko Kondo et al.
As described in Vol. 65, 1996, gallium indium arsenide nitride (Ga _x
In _1-x N _y As _1-y : 0 ≦ x ≦ 1, 0.001 ≦ y <
0.2, hereinafter abbreviated as GaInNAs. ) Is known to be a material capable of adjusting the band gap while controlling the composition of In and N to lattice match with the GaAs substrate as shown in FIG.

According to the present invention, the structure of a solar cell is studied,
By forming GaInNAs under the GaAs solar cell so that the lattice constant becomes close to that of GaAs, GaInNAs is formed.
By adopting a structure in which a GaInNAs solar cell having a small band gap is formed so that the lattice matching with the As solar cell is improved and the conversion efficiency of the stacked solar cell is improved, a solar cell with high conversion efficiency can be manufactured at low cost. Will be obtained.

FIG. 4 shows that the inventors calculated GaAs.
It is the result of having predicted by calculation the relationship between the band gap (Eg) of the GaInNAs solar cell formed under the solar cell and the conversion efficiency. In this calculation, the resistance component, the electrode loss, and the like are not considered, and the ideal conversion efficiency is obtained.

[0024] As shown in FIG.
It has been found that when the Eg of nNAs is in the range of 0.7 eV to 0.8 eV, the conversion efficiency of the GaAs / GaInNAs stacked solar cell can be as high as over 30%.

Further, according to the present invention, a GaAs solar cell is formed on a solar cell formed of, for example, a nitrogen mixed crystal compound semiconductor on a GaAs substrate, and GaAs is further formed thereon.
By forming a GaInP solar cell or an AlGaAs solar cell that is lattice-matched to s and has a larger band gap than GaAs, a stacked solar cell that is more efficient than a GaAs / GaInNAs stacked solar cell can be obtained.

This GaInP / GaAs / GaInNA
s stacked solar cell or AlGaAs / GaAs / G
aInNAs stacked solar cells are also GaAs / G
Like the aInNAs stacked solar cell, it can be formed on a Ge substrate which is lower in cost than a GaAs substrate.

[0027] Ga lattice-matched to GaAs _z In _1-z P
(0.4 <z <0.6), Al _u Ga _1-u As (0.2 <
u <0.4), the band gap is 1.8 to 2.0 eV,
The band gap of GaAs is 1.43 eV.

FIG. 5 shows GaInP / GaAs / GaIn.
GaInP and GaAs
Is a result of calculating and predicting the relationship between the band gap and the conversion efficiency of the GaInNAs solar cell in consideration of the aforementioned band gap. Similar to FIG. 4, this calculation does not consider the resistance component, the electrode loss, and the like, and obtains the ideal conversion efficiency.

As shown in FIG. 5, when the band gap is in the range of 0.75 eV to 1.0 eV,
It has been found that the conversion efficiency of the GaInP / GaAs / GaInNAs stacked solar cell exceeds 35%.

[0030]

(Embodiment 1) FIG. 1 is a sectional view showing Embodiment 1 of a laminated solar cell of the present invention. Referring to FIG. 1, n-type GaInN is formed on an n-type GaAs substrate (1).
Base layer (4) made of As (thickness: 3 μm, n (n-type carrier concentration) = 1 to 2 × 10 ¹⁷ cm ⁻³ ), p-type GaI
Emitter layer (2) made of nNAs (thickness: 0.1 μm)
m, p (p-type carrier concentration) = 2 to 4 × 10 ¹⁸ cm ⁻³ )
To form a GaInNAs solar cell (2), on which a base layer (7) (thickness) of n-type GaAs is formed via a tunnel junction layer (6) of p ^++- type GaAs, n ^++- type GaAs or the like. Length: 3 μm, n = 1 to 2 × 10 ¹⁷ cm ⁻³ ),
Emitter layer (8) made of p-type GaAs (thickness: 0.5
μm, p = 2-4 × 10 ¹⁸ cm ⁻³ ), and p-type GaI
Window layer (9) made of nP (thickness: 0.03 μm, p = 1
× 10 ¹⁸ cm ^-3 ) and a p-type GaAs contact layer (10) (thickness: 0.1 μm, p> 1 × 10 ¹⁹ cm ^-3 ) for lowering the contact resistance with the electrode.
By forming the solar cell (3), a two-junction stacked solar cell was formed.

Here, the composition of GaInNAs is GaAs
The band gap was determined so as to be lattice-matched to s and to be 0.7 eV or more and 0.8 eV or less as shown in FIG. In the first embodiment, Ga _0.7 In _0.3 N
_0.03 As _0.97 . Crystal growth was performed using metal organic chemical vapor deposition (MOCVD).

As raw materials for crystal growth, triethyl gallium (TEG), trimethyl indium (TMI), arsine (AsH ₃ ), and dimethylhydrazine (DMH)
y) was used. The p-type dopant and the n-type dopant are diethyl zinc (DEZ) and silane (SiH ₄ ), respectively.
Was used. The crystal growth pressure is 76 Torr and the growth temperature is 5
The temperature was set at 00 to 600 ° C.

In the first embodiment, the crystal growth is performed by using the organic metal growth method. However, other growth methods such as the molecular beam growth method (CBE) can be sufficiently used. Further, instead of p-type GaInP, p-type Al _u Ga _1-u As is used as the window layer.
(Where u is greater than 0.7 in the case of a window layer).

Using a solar simulator, the characteristics of the obtained laminated solar cell were measured using AM1.5, 1sun (100
(mW / cm ² , no light collection), a conversion efficiency of 31% was obtained. The cost of producing the solar cell in this case was about 900 yen per cm ² .

The solar cell having the structure of the first embodiment
When 87 pieces were made, 30%
The above conversion efficiency was obtained.

Using a n-type Ge substrate instead of the GaAs substrate (1) in Example 1, the characteristics of a stacked solar cell having the same structure were evaluated.
%. In this case, the cost of the Ge substrate is GaAs.
Since the cost is about 1/6 to 1/4 of that of the s substrate, the cost of the solar cell was about 600 yen per 1 cm ² .

When 100 solar cells having this structure were manufactured, conversion efficiency of 27% or more was obtained in 78 of them.

The conversion efficiencies obtained according to the present invention are the same as those of GaI formed on an InP substrate according to the prior art.
nAs solar cell and GaAs formed on GaAs substrate
The conversion efficiency of a stacked solar cell constituted by mechanically stacking a solar cell and a solar cell is 28.8% (H. Mats
ubara et al. : Proc. of PVS
EC-9, p. 533, 1996).

Further, in the case of a solar cell manufactured by the conventional technique, since a more expensive InP substrate is used together with a GaAs substrate, and a stacking process is added, a 1 cm
^The cost was high at about 1480 yen per ² units.

Furthermore, in the prior art, only 54 solar cells out of 100 solar cells could achieve a conversion efficiency of 27% or more due to variations in the stacking process.

Example 2 FIG. 2 is a sectional view showing Example 2 of the laminated solar cell of the present invention. Referring to FIG.
A base layer (4) made of n-type GaInNAs (thickness: 3 μm, n = 1 to 2 × 10 ¹⁷ ) on an n-type GaAs substrate (1)
cm ⁻³ ), an emitter layer (5) made of p-type GaInNAs (thickness: 0.1 μm, p = 2 to 4 × 10 ¹⁸ c)
m- ³ ) to form a GaInNAs solar cell (2),
On top of this, a base layer (7) made of n-type GaAs (thickness: 3 μm, n = 1 to 2 × 10 4) through a tunnel junction layer (6) made of p ⁺⁺ -type GaAs, n ⁺⁺ -type GaAs or the like. ¹⁷ cm
^-3 ), an emitter layer (8) made of p-type GaAs (thickness:
0.5 μm, p = 2 to 4 × 10 ¹⁸ cm ⁻³ ), and a window layer (9) made of p-type GaInP (thickness: 0.03 μm,
GaAs solar cell (3) with p = 1 × 10 ¹⁸ cm ⁻³ )
Is formed, and an n-type GaIn is formed thereon via a tunnel junction layer (6) made of p ^++- type GaAs, n ^++- type GaAs or the like.
P base layer (12) (thickness: 3 μm, n = 1 to 1)
2 × 10 ¹⁷ cm ⁻³ ), p-type GaInP emitter layer (13) (thickness: 0.1 μm, p = 2 to 4 × 10 ¹⁸ c)
m ^-3 ) and p-type Al _v In _1-v P (0.4 <v <0.
Window layer (14) consisting of 6) (thickness: 0.03 μm, p =
1 × 10 ¹⁸ cm ⁻³ ) and a p-type GaAs contact layer (10) (thickness: 0.1 μm, p> 1 × 10 ^19).
cm ^-3 ) to form a GaInP solar cell (11) to form a three-junction stacked solar cell.

Here, the composition of GaInNAs is GaAs
The band gap was determined so as to be lattice-matched to s and to be not less than 0.75 eV and not more than 1.0 eV as shown in FIG. In Example 2, Ga _0.9 In _0.1
N _0.03 As _0.97 . Also in Example 2, the metal organic vapor phase epitaxy was used for crystal growth. In addition, p as a window layer
Instead of p-type GaInP, p-type Al _u Ga _1-u As (where u is greater than 0.7 in the case of a window layer) may be formed.

When the characteristics of the obtained laminated solar cell were evaluated using a solar simulator, the conversion efficiency was 35%.
was gotten. In this case, the cost of producing a solar cell is 1c
m was about 1050 yen per ^2.

The solar cell having the structure in Example 2 was replaced with 100
When 33 pieces were produced, 33%
The above conversion efficiency was obtained.

When the characteristics of a stacked solar cell having a similar structure using an n-type Ge substrate instead of the GaAs substrate (1) were measured under the same conditions as above, a conversion efficiency of 33% was obtained. Was. In this case, the cost of the solar cell is 1c
m was about 740 yen per ^2. In addition, when 100 solar cells having this structure were manufactured, a conversion efficiency of 30% or more was obtained in 72 of them.

These conversion efficiencies are, according to the prior art, I
a GaInAs solar cell formed on an nP substrate;
The conversion efficiency of a laminated solar cell formed by mechanically stacking a GaAs solar cell on an As substrate and further forming a GaInP solar cell on the GaAs solar cell is 33.3% (T. Takamoto eta
l. : Proc. of the 26th IEEE
(PVSC, 1997).

In addition, in the method according to the prior art, a more expensive InP substrate is used in addition to the GaAs substrate, and a stacking process is added.
^The cost was high at about 1550 yen per ² units.

Further, in the conventional process, only 48 solar cells out of 100 solar cells could obtain a conversion efficiency of 30% or more due to variations in the stacking process.

Example 3 Example 3 of the present invention
In place of the GaInP solar cell (11) of Example 2, a base layer made of n-type AlGaAs, p-type AlGa
An AlGaAs solar cell is formed by an emitter layer made of As, a window layer made of p-type AlGaAs (Al composition in the window layer> 0.7), and a p-type contact layer.
An aAs / GaAs / GaInNAs stacked solar cell was constructed.

When the characteristics of this stacked solar cell were evaluated using a solar simulator, it was confirmed that conversion efficiency almost equivalent to that of the stacked solar cell of Example 2 could be obtained.

As described above, according to the present invention, as compared with a conventional solar cell obtained by using two kinds of expensive substrates of InP and GaAs and performing a mechanical lamination process, the present invention has the same or better performance. The solar cell having the characteristics described above can be formed continuously using only a GaAs substrate or only a less expensive Ge substrate, and a significant cost reduction can be achieved.

[0052]

According to the present invention, a stacked solar cell having high conversion efficiency can be easily obtained at low cost.

[Brief description of the drawings]

FIG. 1 is a structural diagram for explaining a first embodiment of the present invention.

FIG. 2 is a structural diagram for explaining a second embodiment of the present invention.

FIG. 3 is a diagram showing a relationship between a lattice constant and a band gap of a GaInNAs nitrogen mixed crystal compound system.

FIG. 4 is a diagram showing the relationship between the band gap of GaInNAs and the conversion efficiency of the stacked solar cell in the GaAs / GaInNAs stacked solar cell.

FIG. 5 is a diagram showing the relationship between the band gap of GaInNAs and the conversion efficiency of the stacked solar cell in the GaInP / GaAs / GaInNAs stacked solar cell.

[Explanation of symbols]

Reference Signs List 1 n-GaAs substrate 2 GaInNAs solar cell 3 GaAs solar cell 4 n-GaInNAs base layer 5 P-GaInNAs emitter layer 6 n ⁺⁺ / p ^++- GaAs tunnel junction layer 7 n-GaAs base layer 8 P-GaAs emitter layer Reference Signs List 9 P-GaInP window layer 10 P-GaAs contact layer 11 GaInP solar cell 12 n-GaInP base layer 13 P-GaInP emitter layer 14 P-AlInP window layer

Claims (10)

[Claims] 1. III-V having a zinc blende type crystal structure
In group III compound semiconductors, nitrogen is 0.1% in group V mixed crystal ratio
It has a structure in which a solar cell composed of a semiconductor material having a band gap larger than that of the nitrogen mixed crystal compound semiconductor material is stacked on a solar cell composed of the nitrogen mixed crystal compound semiconductor material containing above. Stacked solar cell. 2. Gallium indium arsenide nitride (Ga _x I)
n _1−x N _y As _1−y : 0 ≦ x ≦ 1, 0.001 ≦ y <0.
The stacked solar cell according to claim 1, wherein the stacked solar cell has a structure in which a solar cell made of gallium arsenide (GaAs) is stacked on the solar cell made of (2). 3. The stacked solar cell according to claim 2, wherein the stacked solar cell is formed on a gallium arsenide (GaAs) substrate. 4. The stacked solar cell according to claim 2, wherein the stacked solar cell is formed on a germanium (Ge) substrate. 5. The lamination according to claim 1, wherein a band gap of the nitrogen mixed crystal compound semiconductor is in a range from 0.7 eV to 0.8 eV. Type solar cell. 6. Gallium indium arsenide nitride (Ga _x I)
n _1−x N _y As _1−y : 0 ≦ x ≦ 1, 0.001 ≦ y <0.
On the solar cell constituted by two), by laminating a solar cell composed of gallium arsenide (GaAs), further gallium phosphide, indium thereon (Ga _z an In
_The solar cell according to claim 1, wherein the solar cell has a structure in which _1-z P: 0.4 <z <0.6).
3. The stacked solar cell according to item 1. 7. Gallium indium arsenide nitride (Ga _x I)
n _1−x N _y As _1−y : 0 ≦ x ≦ 1, 0.001 ≦ y <0.
A solar cell composed of gallium arsenide (GaAs) is laminated on the solar cell composed of 2), and further a gallium aluminum arsenide (Al _u Ga _1-u A) is further formed thereon.
2. The stacked solar cell according to claim 1, wherein the stacked solar cell has a structure in which solar cells configured by s: 0.2 <u <0.4) are stacked. 8. The stacked solar cell according to claim 6, wherein the stacked solar cell is formed on a gallium arsenide (GaAs) substrate. 9. The stacked solar cell according to claim 6, wherein the stacked solar cell is formed on a germanium (Ge) substrate. 10. The laminated solar cell according to claim 6, wherein a band gap of the nitrogen mixed crystal compound semiconductor is in a range of 0.75 eV or more and 1.0 eV or less. battery.