JPH09199738A - Solar cell - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce a surface re-combination rate of a semiconductor/insulating film interface and to improve conversion efficiency by providing a semiconductor having a work function different from that of a semiconductor layer having a wider forbidden band width than that of an insulating member or that of a semiconductor serving as a solar cell. SOLUTION: A solar cell consists of a semiconductor A (solar battery) 1, an insulating member (or a semiconductor C having a wider forbidden band width than that of the semiconductor A) 2 and a semiconductor B3. It is important to provide the semiconductor B3 having a higher work function than that of the semiconductor A1. It is of course possible to reduce surface re-combination even if the solar cell consists entirely or partly of the semiconductor A (solar battery) 1 and the semiconductor B3 without the insulating member 2 or the semiconductor C. Moreover, by providing the semiconductor B3 having a lower work function than that of the semiconductor A1 serving as a solar cell, the electron density in the vicinity of the surface can be increased.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a solar cell having reduced surface recombination.

[0002]

2. Description of the Related Art Needless to say, reduction of surface recombination is important for improving conversion efficiency and characteristics such as fill factor, open circuit voltage and short circuit current in solar cells. Heretofore, in order to reduce surface recombination, the surface of the semiconductor has been covered with a high quality oxide film. In this case, reducing the surface recombination means reducing the interface state density of the semiconductor / oxide film interface. Today, the interface state density of unit eV and unit power of less than 10 per square cm has been realized by passivation using a thermal oxide film, but it is still necessary to reduce the surface recombination rate in order to improve the conversion efficiency. . However, in order to further reduce the interface state density with a high quality thermal oxide film,
A cleaner atmosphere is required and enormous capital investment is required to achieve this. In contrast to such a method of reducing the interface state density, a method of bending the band near the semiconductor surface and making a difference in electron / hole concentration can reduce the surface recombination. Next, we introduce three typical methods.

One is a floating inversion layer near the base layer which is in contact with the oxide film (a layer having a conductivity type different from that of the base layer, and floating means that this layer is not directly connected to the terminal). There is a method of forming. This typical solar cell is, for example, the 1994 IEEE First World Conference on Photovoltaic Energy Conversion Proceedings (1994 IEEE First World Conference).
on Photovoltaic Energy Conversion Conference
Record) Volume 2, pp. 1278-1282.

The second method is to arrange a gate electrode via an oxide film. Theoretically, if an external voltage is applied to such a MOS structure, an accumulation layer or an inversion layer can be formed freely, and surface recombination can be reduced. This typical solar cell is, for example, Solar Energy Materials and Solar Cells (Solar En
ergy Materials and Solar Cells) Volume 2, 127
It is described on pages 8 to 1282.

The third method is to introduce fixed charges into the oxide film or to dispose the fixed charge layer through the oxide film.

[0006]

In the prior art using the floating inversion layer described above, surface recombination at the Si / SiO ₂ interface on the back surface is reduced, while the lifetime in the floating layer is reduced and the entire solar cell is reduced. Recombination increases. Therefore, there is a drawback that the conversion efficiency of the solar cell is not improved but is decreased. In addition, the conventional technique using the MOS structure has a drawback that a separate power supply is purposely required to apply a voltage to the gate electrode. There is also a problem that the gate electrode to which a voltage is applied from the outside is short-circuited with the base substrate due to a cause such as a pinhole, and not only the characteristics of the solar cell are improved but also no power is generated. The method of introducing fixed charges into the oxide film has the drawbacks that the fixed charges are not stable and that the fixed charge density cannot be made high enough to bend the band edge of the solar cell surface. After all, high-efficiency solar cells have not been achieved by the surface recombination suppressing methods by band bending in the vicinity of the above-mentioned three types of semiconductor surfaces. Still, a method of reducing the interface state density, that is, a method of avoiding contamination of foreign substances as much as possible at the time of manufacturing a solar cell is used, and the capital investment for that is enormous.

In order to solve the above problems, the present invention has an object to improve the conversion efficiency by reducing the surface recombination ratio at the semiconductor / insulating film interface without reducing the bulk recombination. . Another object of the present invention is to realize a solar cell that does not require any other power source, is easy to process, and can stably reduce surface recombination.

[0008]

The above-mentioned object is to provide a solar cell in which at least a part of the semiconductor surface is covered with an insulator or a semiconductor having a forbidden band width larger than that of a semiconductor constituting the solar cell. A semiconductor layer having a larger forbidden band width than that of a semiconductor forming a body or a solar cell, a semiconductor layer is provided on a part or the whole of the semiconductor, and the work function of the semiconductor layer is larger than that of a semiconductor forming an insulator or a solar cell. It is solved by having a work function of a semiconductor in contact with the inside of a semiconductor having a band width and a different value.

In addition, the work function of the semiconductor layer provided on a part or the whole of the semiconductor with the large forbidden band width is larger than that of the semiconductor that constitutes the insulator or the solar cell through the semiconductor having the forbidden band width. It is solved by having a value lower than the work function of the semiconductor in contact, or a semiconductor layer having a work function different from the work function of the semiconductor in contact through the semiconductor having the large forbidden band is arranged on the light-receiving surface. Will be solved.

Further, the above problem is that the semiconductor layer provided on a part or all of the semiconductor having the large forbidden band width is
The problem is solved by being electrically separated from the contact and the electrode wiring portion, or the impurity concentration of the semiconductor layer provided in a part or all of the semiconductor is set to 10% of the impurity concentration of the semiconductor in contact with the inside through the insulator. It is solved by doubling.

[0011]

BEST MODE FOR CARRYING OUT THE INVENTION In order to solve the above-mentioned problems in a solar cell, the structure of a part of the surface of the solar cell according to the present invention is shown in FIG. 1 with a semiconductor A (solar cell) 1 and an insulator (or A semiconductor C) 2 and a semiconductor B3 having a forbidden band width larger than that of the semiconductor A are used. here,
It is important to dispose the semiconductor B3 having a work function different from that of the semiconductor A1. Of course, the surface recombination can be reduced even if the semiconductor A (solar cell) 1 and the semiconductor B3 are partially or entirely formed without completely interposing the insulator 2 or the semiconductor C.

The principle of reducing surface recombination (recombination at the interface between the semiconductor A1 and the insulator or the semiconductor C3) in the case of the above structure is the Shockley-Lead-Hole (SRH) recombination theory. Can be simply explained. First, as shown in FIG. 2, when two types of semiconductor layers having different work functions 6 (which have almost the same meaning as chemical potential at room temperature) are brought close to each other, electrons and holes move so that the chemical potentials become the same level. To do. Specifically, electrons gather on the surface of the semiconductor side having a large chemical potential (holes decrease) based on the vacuum level 8 (holes decrease), and holes gather on the surface of the semiconductor side having a small chemical potential (electrons decrease). As a result, an electric double layer is formed on the semiconductor surface, and the potential barrier 14 is formed. Unlike normal chemical equilibrium, this phenomenon occurs even if semiconductors do not contact each other. Due to such movement of electrons and holes, the electron concentration and hole concentration on the semiconductor end face change. It is considered that the maximum number of pairs that can be recombined when the electron concentration and the hole concentration are equal at the end face of the semiconductor is actually the P-type silicon / oxide film because of the difference in the electron and hole trapping cross sections. At the interface, the recombination number is maximum when the hole concentration is 2-3 orders of magnitude higher than the electron concentration. On the contrary, even if the interface trap density is high, the recombination rate decreases as the ratio of the hole concentration to the electron concentration becomes far from this.

From the above principle, it is possible to dispose a metal having a different work function with respect to the semiconductor forming the solar cell. However, the semiconductor has a larger work function difference with respect to the type than a metal, and the work function can be arbitrarily changed by doping impurities. Further, in general, metal has a high reflectance with respect to visible light and a high absorption coefficient, so that it is impossible to dispose it on the light receiving surface unless it is made very thin. If it ’s a semiconductor,
It is also possible to arrange it on the light receiving surface by selecting a material whose band gap energy is higher than the photon energy of visible light.

Further, in the present invention, in order to solve the above-mentioned problems, not only a semiconductor having a work function different from that of a semiconductor constituting the solar cell is arranged around the solar cell through an insulating film, but Take measures such as. That is, a semiconductor B having a small work function is arranged with respect to the semiconductor A constituting the solar cell, the electrodes and wirings of the semiconductor B and the solar cell are electrically separated, and the semiconductor B is disposed on the light receiving surface side of the solar cell The semiconductor B to be placed in
The impurity concentration of is set to 10 times or more the impurity concentration of the semiconductor A constituting the solar cell.

The semiconductor having a small work function is arranged with respect to the semiconductor constituting the solar cell via an insulator or a semiconductor layer having a wide band gap in order to increase the electron concentration near the surface. Particularly in a silicon solar cell, the electron capture cross-section at the interface with the oxide film is higher than the hole capture cross-section by 2 to 3 orders of magnitude, so surface recombination occurs when the hole concentration is 2 to 3 orders of magnitude higher than the electron concentration. Is the maximum. To use the work function difference and prevent the conditions that maximize surface recombination from being met, place a semiconductor with a large work function to increase the hole concentration near the interface, and a semiconductor with a small work function. It is easier to arrange the to increase the electron concentration near the interface.

Regarding the second point, the following situation can occur. When the semiconductor B arranged via the insulator or the semiconductor layer C having a wide band gap is electrically connected to the electrode or the wiring, the potential of the semiconductor layer B is naturally equal to that of the electrode. As shown in FIG. 3, even if the inversion layer is formed by using the work function difference between the semiconductor A1 and the semiconductor B3, when the solar cell is irradiated with sunlight, the operating voltage 16 is applied to the semiconductor B and the semiconductor B Will change the Fermi level of. Since this change tends to eliminate the inversion layer of the solar cell, it may increase surface recombination. The purpose of electrically separating the semiconductor B arranged via the insulator or the semiconductor layer C from the electrodes and wirings is to leave the band bent at the operating voltage as it is.

The third arrangement of the semiconductor having a work function different from that of the semiconductor constituting the solar cell on the light receiving surface side not only reduces the surface recombination but also reduces the impurity concentration of the surface layer, In addition, it is for use as an antireflection film. In a normal double-sided contact solar cell, the light-receiving surface is covered with a high-concentration layer having a conductivity type opposite to that of the base. In order to improve the efficiency of solar cells, surface recombination of this layer, SR
Not only is H recombination reduced, but it is also necessary to reduce Auger recombination from high concentrations. However, if the impurity concentration in this layer is lowered, the electric resistance becomes high, and as a result, the fill factor becomes low and high conversion efficiency cannot be obtained. On the other hand, by arranging a semiconductor of the same conductivity type on the insulator in the lightly diffused diffusion layer, the above problems can be solved and a higher open circuit voltage, higher fill factor and higher conversion efficiency than before can be obtained. Further, the surface reflectance can be reduced by selecting a semiconductor having a refractive index lower than that of the semiconductor forming the solar cell and optimizing the film thickness.

The fourth semiconductor described above is arranged with a semiconductor concentration of 10 times or more the impurity concentration of the semiconductor constituting the solar cell, as shown in FIG. 4, the potential on the side of the semiconductor 1 constituting the solar cell. This is to raise the barrier. In the case of semiconductors having similar dielectric constants, if the impurity concentration of one semiconductor 3 is set to 10 times the impurity concentration of the other semiconductor 1, the potential barrier in the vicinity of the insulating film of the high-concentration semiconductor is increased. It is 1 when compared to the low concentration
It will be less than 1/0. Here, it goes without saying that the space charge region can be increased when the impurity concentration of the semiconductor forming the solar cell is lower. When the interface state density between the semiconductor and the insulating film is high, Fermi level pinning occurs. This effect can be reduced by increasing the impurity concentration.

[0019]

Embodiments of the present invention will now be described with reference to the drawings.
FIG. 5 is a diagram showing a first embodiment of a solar cell according to the present invention, FIG. 6 is a diagram showing a manufacturing process of the above embodiment, FIG. 7 is a diagram showing separation of a two-layer film and a wiring electrode in the above embodiment, 8 is a diagram showing a second embodiment of the present invention, FIG. 9 is a diagram showing another application of the above-described embodiment, and FIG. 10 is a diagram showing a third embodiment of the present invention.

First Embodiment A first embodiment of the solar cell according to the present invention is shown in FIG. 5, and in FIG. 6 showing the manufacturing process thereof, as shown in FIG. 6 (a), on a P-type crystalline silicon substrate 22. The concavo-convex structure 18 for confining light was formed by anisotropic etching on the light incident side surface of the. Next, the thermal oxide film 27 for the diffusion mask is set to 0.2 μm.
The hole 28 of the high-concentration phosphorus diffusion portion was formed by a known photo-etching method. The N-type high-concentration diffusion layer 20 having a junction depth of 0.5 μm was formed by thermally diffusing POCl ₃ . Next, as shown in FIG. 6 (b), the light incident side surface of the thermal oxide film 27 is removed, and the junction depth is 0.2 μm over the entire light receiving surface.
The n-type low concentration diffusion layer 21 of m was formed by thermal diffusion of POCl ₃ . After that, the thermal oxide film 27 for the diffusion mask is set to 0.2.
μm was formed. Next, as shown in FIG. 6C, a hole is formed in the boron diffusion portion 29 on the back surface by a known photo-etching method. The P-type diffusion layer 23 was formed by thermally diffusing BBr ₃ . The junction depth at this time was 0.4 μm. Next, the thermal oxide film 27 was removed, and as shown in FIG. 6D, a passivation oxide film 24 was formed by thermal oxidation to a thickness of 10 nm. After that, the ZnSe film 25 is made of solid Zn and Se as raw materials (beam pressure is 1.2 × (1
/ 10 ⁷ ) Torr, 1.8 × (1/10 ⁷ ) Torr),
A film thickness of 2.0 μm was deposited at a growth temperature of 250 ° C. At this time, nitrogen gas doping was performed so that the impurity concentration of the above layer was 1.0 × 10 ¹⁸ / cm ³ (N ₂ beam pressure: 1.2 ×
(1/10 ⁷ ) Torr). It was observed by electron microscopy that the film was polycrystalline. While the band gap energy of silicon is about 1.1 eV, this P-type Zn
Since the band gap energy of the Se film is about 2.8 eV, almost all the light that can be absorbed by this layer is absorbed by the silicon layer and does not reach this layer. Therefore, although the reflectance on the back surface is slightly reduced, this layer causes almost no photoelectric conversion loss. After that, a contact hole 30 is formed on the back surface by a photo-etching method, and Al 26 is deposited on the entire back surface as shown in FIG. 6 (e). Also on the light-receiving surface, a contact was opened by a photo-etching method, Ti / Ag17 was vapor-deposited, and the metal vapor-deposited except the electrode portion was removed by lift-off.

After the above example was completed, the electrical characteristics were measured by a solar simulator (AM1.5, 100 mW / cm ² ).
When measured under the following conditions, the conversion efficiency was 23.2%, the fill factor was 0.825, the open circuit voltage was 0.693 V, and the short circuit current density was 40.6 mA / cm ² . This conversion efficiency is 0.9 compared with ordinary solar cells without ZnSe film.
%high.

The work function of the P-type silicon used as the substrate in the present invention is about 4.9 eV, whereas the work function of the P-type doped ZnSe film is about 6.2 eV. It is considered that due to this work function difference, the band of the P-type semiconductor layer (solar cell base layer) near the silicon / oxide film interface was bent to the conduction band side, and the surface electron concentration was extremely reduced. As a result, the recombination on the back surface side was reduced as described above, and the solar cell characteristics were improved. Actually, the lifetime was measured by the photoconductivity decay method without electrodes and the surface recombination rate was estimated. As a result, when the ZnSe film was not attached, the surface recombination rate was about 600 cm / s. On the other hand, it was about 50 cm / s when the ZnSe film was attached.

After passing through the step of forming the passivation oxide film 24 to a thickness of 100 nm in FIG.
Instead of the e film, an intrinsic AlAs film is deposited by MBE method at a substrate temperature of 580 ° C. to a film thickness of 2.0 μm, and then the Al film is continued.
_{A 0.7} Ga _0.3 As film 25 was deposited as a cap layer to a thickness of 1.0 μm. Band gap energy of the AlAs layer and Al _0.7 Ga _0.3 As as described above, since each is about 2.15eV to about 1.95 eV, photoelectric conversion loss hardly occurs even by this layer. After that, electrodes were attached through the same steps as above to complete the present invention.

This electrical characteristic is measured by the solar simulator (A
When measured under M1.5, 100 mW / cm ₂ ),
The results obtained were a conversion efficiency of 23.2%, a fill factor of 0.820, an open circuit voltage of 0.689 V, and a short circuit current density of 41.0 mA / cm ² .

In the present invention, the work function of the intrinsic AlAs film is about 3.6 eV, and it is considered that the band of the base layer near the oxide film is bent toward the valence band due to the work function difference with silicon. As a result, the surface hole concentration was reduced and the recombination on the back surface side was reduced. Also for this, with the electrode not attached, the lifetime was measured by the photoconductivity decay method and the surface recombination rate was estimated. As a result, AlAs / Al _0.7 Ga _0.3 A
The surface recombination rate was about 600 cm / s when the two-layered membrane of s was not attached, whereas AlAs / Al _0.7 Ga
_It was about 10 cm / s when a _0.3 As two-layer film was attached.

Further, as shown in FIG. 7, the AlAs / Al _0.7 Ga _{0.3 As} two-layer film 25 and the wiring electrode 26 of the above-mentioned embodiment are used by using a phosphoric acid and sulfuric acid type etching solution by the photo-etching method.
Separated. AlAs / Al _0.7 Ga _0.3 separated at this time
The metal film 31 on the As2 layer film 25 is not removed because the back surface reflectance is not reduced. When this electrical property was measured under a solar simulator (AM1.5, 100 mW / cm ² ), conversion efficiency was 23.8% and fill factor was 0.1.
821, open circuit voltage 0.707V, short circuit current density 41.0m
The result was A / cm ² . The improvement in open-circuit voltage and conversion efficiency was especially seen compared to the conventional one.By adopting this configuration, the band bent to the valence band side at the operating voltage and the open-end condition returned to the flat state. Probably because it is gone.

The above results are obtained at the interface state density of silicon / oxide film of 1.0 × 10 ¹⁰ / cm ² eV, but the same method is adopted even when the interface state density is increased by one digit. By doing so, a solar cell having almost the same characteristics could be obtained.

Second Embodiment FIG. 8 showing a second embodiment of the present invention was prepared as follows. The N-type diffusion 21 was performed on the P-type crystalline silicon substrate 22 by using the same process as in the first embodiment. At this time, the surface impurity concentration of the N layer other than directly under the finger portion is 1.
The diffusion depth was 0 × 10 ¹⁸ / cm ³ and the diffusion depth was 0.6 μm. After that, P-type diffusion 23 was performed as in the first embodiment, and a passivation oxide film 24 of 20 nm was formed by a thermal oxidation method. Further, after forming both side electrodes 17 and 26 on the front surface and the back surface, as shown in FIG. 8, 210 nm of N-type ZnS polycrystalline film 32 was deposited on the light receiving surface by MBE method at a substrate temperature of 220 ° C. At this time, Cl (ZnCl powder) is used as a dopant.
Was used to dope so that the impurity concentration of this layer would be 1.0 × 10 ¹⁸ / cm ³ . Since the band gap energy of the ZnS film is about 3.8 eV, it is about 330 nm.
It can only absorb light of the following wavelengths. Since the intensity of sunlight with a wavelength of 330 nm or less is almost equal to 0, this layer causes almost no photoelectric conversion loss.

After the completion of this example, the electrical characteristics were measured under a solar simulator (AM 1.5, 100 mW / cm ² ) and the conversion efficiency was 22.9% and the fill factor was 0.9.
825, open circuit voltage 0.689V, short circuit current density 40.3m
The result was A / cm ² . This conversion efficiency is N type ZnS
It is 0.6% higher than a normal solar cell without a membrane.

In this embodiment, the work function of the N layer silicon is about 4.2 eV, whereas the work function of the N-type doped ZnS film is about 4.0 eV. It is considered that this work function difference causes the band of the N-type semiconductor layer (solar cell base layer) near the silicon / oxide film interface to be further bent to the valence band side, increasing the surface electron concentration and decreasing the surface hole concentration. .
This makes it possible to reduce the electric resistance without increasing the impurity concentration of the N layer, and by lowering the impurity concentration of the N layer, in addition to surface recombination, SRH recombination in the N layer and Auger Recombination is reduced and the characteristics of the solar cell are improved. Further, since the ZnS film arranged on the light receiving surface has a refractive index of about 2.4, it serves as an antireflection film for silicon of the substrate. Conventionally, a silicon oxide film is used as a passivation film and at the same time as an antireflection film, and the optimum film thickness is about 110 nm. By disposing the ZnS film on the light receiving surface, the silicon oxide film can be thinned, and the oxidation time and the oxidation temperature can be reduced in the oxidation process. As a result, the interface state density can also be reduced, and the reduction in bulk lifetime can be further prevented.

The same structure is applied to a GaAs solar cell in the embodiment shown in FIG. In this case, the intrinsic AlAs layer 37 corresponds to the silicon oxide film, and Zn
The P-type GaN layer 36 corresponds to the S film. Since the GaN layer also has a bandgap energy of about 3.5 eV, it can absorb only light having a wavelength of about 360 nm or less. Therefore, similar to the above reason, photoelectric conversion loss due to this layer hardly occurs.

In the manufacturing method of this embodiment, first, an N-type GaAs substrate 41 is prepared, etching is performed with a sulfuric acid-based etching solution, and then rinsed with running water for 10 minutes. Then, it is loaded into the MBE device and, using a well-known process, a 0.5 μm thick N-type GaAs is formed.
The buffer layer 40 has a thickness of 3.0 μm and an impurity concentration of 2.0 × 1.
0 ¹⁷ / cm ³ N-type GaAs layer 39, 0.5 μm thick with an impurity concentration of 5.0 × 10 ¹⁸ / cm ³ P-type GaAs layer 38, intrinsic AlG
An aAs (Al: Ga = 0.7: 0.3) layer 37 is deposited in order of 10 nm. The growth temperature at this time is 580 ° C. Here, the device is once taken out of the apparatus, and after mounting the metal mask for electrode formation, the high-concentration P-type GaAs layer 35 is deposited again, and then the front surface electrode 33 and the back surface electrode 42 are vapor-deposited to complete the electrode portion. Then, the substrate temperature is set to 500 by the gas source MBE method.
A P-type GaN layer 36 having a thickness of 0.5 μm is deposited on the light-receiving surface side at 0 ° C., and then an antireflection film 34 is vapor-deposited to obtain the solar cell of this embodiment. In the step of depositing the GaN layer, NH ₃ gas was used as the nitrogen source and Mg was used as the P-type dopant.

By adopting the above structure, P-type Ga
P-type AlGaAs (Al: Ga = 0.7:
The interface state density with the 0.3) layer is changed to the intrinsic AlGaAs ((A
l: Ga = 0.7: 0.3). Further, as a result of disposing a highly-doped P-type GaN layer on the outer side thereof, a work function difference (the work function of the P-type GaAs layer 38 is about 5.4 eV, whereas the P-type GaN layer is
The work function of the aN film was about 6.2 eV), so that the band at the edge of the GaAs layer could be bent to the conduction band side, and surface recombination could be reduced.

After completion of this example, the electrical characteristics were measured under a solar simulator (AM 1.5, 100 mW / cm ² ), conversion efficiency was 23.8%, fill factor was 0.9.
84, open circuit voltage 1.02V, short circuit current density 27.8mA /
The result was cm ² . This conversion efficiency is 1.2% higher on average than that of GaAs solar cells produced by the MBE method.

Third Embodiment A solar cell as a third embodiment of the present invention is shown in FIG.
This example is manufactured as follows. First, the P-type crystalline silicon substrate 22 was thinned with a nitric acid etching solution to a substrate thickness of 100 μm, and then a concavo-convex structure 18 for light confinement was formed on the surface of the substrate 22 on the light incident side by anisotropic etching. Next, a thermal oxide film for a diffusion mask is formed to a thickness of 0.2 μm, and a well-known photo-etching method is used to form a hole in the high-concentration phosphorus diffusion portion.
An N-type high-concentration diffusion layer 23 was formed by thermal diffusion using Cl _{3 so as} to have a junction depth of 0.6 μm and a surface impurity concentration of 1.0 × 10 ²⁰ / cm ³ . Again, a thermal oxide film for a diffusion mask was formed to 0.2 μm, a boron diffusion portion on the back surface was perforated by a well-known photo-etching method, and a P-type diffusion layer 20 was formed by thermal diffusion using BBr ₃ . The junction depth at this time is 0.4 μm, and the surface impurity concentration is 5.0 × 10 ¹⁹ / cm ³ . Next, the passivation oxide films 19 and 24 were formed to a thickness of 10 nm by thermal oxidation. After that, a contact hole is formed on the back surface by photo-etching, and
Then, the emitter (n +) electrode 44 and the collector (p +) electrode 45 were separated by photo-etching. Furthermore, using metal gallium, trimethylaluminum, and NH ₃ gas as raw materials, a substrate temperature of 600 is obtained by a gas source MBE method.
An AlGaN (Al: Ga = 0.4: 0.6) film 43 having a film thickness of 2.0 μm was deposited at a temperature of ° C. AlGaN has poor selectivity, so SiO
Also grows on ₂ . At this time, Si (disilane) was used as a dopant, and the impurity concentration of this layer was 1.0 × 10 ¹⁸ / c.
Doping was performed so as to obtain m ³ . When the film was observed with an electron microscope, it was found to be polycrystalline. Since the AlGaN (Al: Ga = 0.4: 0.6) film also has a bandgap energy of about 4.0 eV, there is almost no photoelectric conversion loss due to this layer.

After completion of the above example, the electrical characteristics were measured under a solar simulator (AM 1.5, 100 mW / cm ² ), conversion efficiency 23.5%, fill factor 0.80, open voltage 0.710 V. , Short circuit current density 41.5
The result was mA / cm ² . This conversion efficiency is AlGaN
(Al: Ga = 0.4: 0.6) 1.2% higher than that of a normal solar cell without a film.

In this embodiment, the work function of the P-type silicon used as the substrate is about 5.0 eV. On the other hand, N
Mold-doped polycrystalline AlGaN (Al: Ga = 0.4: 0.
6) The work function of the film is about 4.0 eV. It is considered that this work function difference causes an inversion layer in the vicinity of the silicon / oxide film interface, and the surface hole concentration is extremely reduced. As a result, the surface recombination of the light receiving surface was reduced as described above, and the characteristics of the solar cell were improved. The lifetime was measured by the photoconductivity decay method without electrodes and the surface recombination rate was estimated. As a result, AlGaN (Al: Ga = 0.4: 0.
6) The surface recombination rate is about 60 when the membrane is not attached.
While it was 0 cm / s, AlGaN (Al: Ga = 0.
It was about 10 cm / s when the membrane was installed.

Further, the impurity concentration of the above embodiment is 1.0.
× 10 ¹⁸ / cm ³ AlGaN (Al: Ga = 0.4: 0.
6) Instead of the film, an AlGaN (Al: Ga = 0.4: 0.6) film having an impurity concentration of 5.0 × 10 ¹⁹ / cm ³ was redeposited. The electrical characteristics of this embodiment are shown in the solar simulator (A
When measured under M1.5, 100 mW / cm ² ),
The results obtained were a conversion efficiency of 23.6%, a fill factor of 0.798, an open circuit voltage of 0.712 V, and a short circuit current density of 41.6 mA / cm ² . This is AlGaN (Al: Ga = 0.4: 0.
6) By increasing the impurity concentration of the film, it is possible to increase the barrier of silicon constituting the solar cell and reduce the influence of Fermi level pinning.

Although crystalline silicon or crystalline GaAs solar cells have been described as examples, the present invention is applicable to polycrystalline or amorphous solar cells of group IV, III-V or II-VI group compounds. Needless to say, is also applicable.

[0040]

As described above, the solar cell according to the present invention is
A solar cell in which at least a part of the semiconductor surface is covered with a semiconductor having a forbidden band width larger than that of an insulator or a semiconductor constituting the solar cell, and a forbidden band larger than that of the semiconductor constituting the insulator or the solar cell. A semiconductor layer is provided on a part or the whole of a semiconductor having a width, and the work function of the semiconductor layer is in contact with the inside through a semiconductor having a forbidden band width larger than that of a semiconductor forming an insulator or a solar cell. By having a work function and a different value, surface recombination can be reduced without deteriorating the bulk lifetime. Also, the above method requires no additional power supply. As a result, it is possible to obtain a high-performance solar cell that does not change over time (deteriorate) without passivation with a high-quality oxide film and has improved conversion efficiency, fill factor, open-circuit voltage, short-circuit current, etc. It will be possible. Further, these effects make it possible to keep the investment for the solar cell manufacturing equipment, which has been conventionally expected, low.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of a solar cell according to the present invention.

FIG. 2 is a band diagram showing the principle of the present invention.

FIG. 3 is a band diagram showing the principle of the present invention.

FIG. 4 is a band diagram showing the principle of the present invention.

FIG. 5 is a diagram showing a first embodiment of a solar cell according to the present invention.

6 (a) to 6 (e) are views showing the manufacturing process of the above-described embodiment.

FIG. 7 is a diagram showing separation of a two-layer film and a wiring electrode in the above example.

FIG. 8 is a diagram showing a second embodiment of the present invention.

FIG. 9 is a diagram showing another application example of the above embodiment.

FIG. 10 is a diagram showing a third embodiment of the present invention.

[Explanation of symbols]

1 ... Semiconductor A (constituting solar cell) 2 ... Semiconductor B (work function is different from semiconductor A) 3 ... Insulator or semiconductor C (forbidden band width is larger than semiconductor A) 4 ... Surface electrode 6 ... Work function 17 ... Ti / Ag front surface electrode 26 ... Al back surface electrode 30 ... contact hole 33 ... Au / Zn front surface electrode 42 ... Au / Ge-Au back surface electrode 44 ... collector back surface electrode 45 ... emitter back surface electrode

Claims (5)

[Claims] 1. A solar cell in which at least a part of the semiconductor surface is covered with a semiconductor having a larger band gap than an insulator or a semiconductor constituting the solar cell, and the semiconductor constituting the insulator or the solar cell. A semiconductor layer is provided on part or all of a semiconductor having a larger forbidden band width, and the work function of the semiconductor layer is an insulator or through a semiconductor having a larger forbidden band width than a semiconductor included in a solar cell. A solar cell having a work function of a semiconductor in contact with the inside and a different value. 2. The solar cell according to claim 1, wherein a work function of the semiconductor provided on a semiconductor having a forbidden band width larger than that of an insulator or a semiconductor constituting the solar cell,
A solar cell having a lower value than the work function of a semiconductor in contact with the inside of an insulator or a semiconductor having a forbidden band width larger than that of a semiconductor constituting the solar cell. 3. The solar cell according to claim 1, which has a work function different from the work function of the semiconductor in contact with the inside through the insulator or the semiconductor having a forbidden band width larger than that of the semiconductor constituting the solar cell. A solar cell in which a semiconductor layer is arranged on the light-receiving surface side. 4. The solar cell according to claim 1, wherein the semiconductor layer provided on a part or the whole of a semiconductor having a forbidden band width larger than that of an insulator or a semiconductor constituting the solar cell is contact and electrode wiring. A solar cell characterized by being electrically separated from the section. 5. The solar cell according to claim 1, wherein the impurity concentration of the semiconductor layer provided on a part or all of a semiconductor having a forbidden band width larger than that of an insulator or a semiconductor constituting the solar cell is set to the above-mentioned. A solar cell, wherein the impurity concentration of a semiconductor in contact with the inside through an insulator is 10 times or more.