JPS58107687A - Semiconductor device - Google Patents

Abstract

PURPOSE:To make it possible to flatten the output and to store electric power, by providing a redox system which is provided with a storage effect only by an area on which light is irradiated so as to form a unitary body with a solar battery. CONSTITUTION:A non-single crystal semiconductor 12 including a PIN junction is provided. A redox 1 is arranged so as to contact with an N or P layer 16 of the semiconductor. An opposing electrode 28 is further provided. Incident light 10 is inputted into the side of a light transmitting substrate 30. This is different from semiconductor electrode reaction using a conventional redox. Electrons or holes generated in said semiconductor 12 are imparted to an interface 9 with the redox 1, and the redox reaction is yielded.

Description

Detailed Description of the Invention The present invention integrates a photoelectric conversion device that generates electrons and holes through light irradiation with a redox (reduction-oxidation reaction solution), thereby flattening output, storing power, Furthermore, it is intended to reduce the use of electricity at night.

Conventional photoelectric conversion devices, particularly solar cells, can generate photovoltaic force by irradiation with light. However, this solar cell produces output only when it is irradiated with sunlight and is proportional to the irradiation intensity, so the output fluctuates greatly depending on whether it is sunny or cloudy.

When installed on the roof of a general household, there are drawbacks such as no photovoltaic power being generated when illuminating at night, which is a major problem for practical use in civilian use.

A method of using a secondary battery to compensate for these problems is known. However, secondary batteries have major drawbacks, such as being expensive and requiring a new installation location.

The present invention compensates for these drawbacks and is characterized by providing a redox system that is integrated with the solar cell and has a power storage effect only in the area that is irradiated with light.

That is, for example, light passing through glass, which is a light-transmitting substrate, is caused to generate a photovoltaic force by a first semiconductor having a PIN junction. The P or N layer provided on one surface is taken out to the outside through ITO, a multilayer film of ITO and tin oxide, or other transparent conductive film provided between the glass and the semiconductor. On the other hand, the N or P layer provided on the back side contacts the redox and provides electrons or holes to the redox. Then, in terms of the charge relationship of this redox, the number of charges is changed as follows: Rn+→R(n+α)+...(1) Qn+→Q(n-β)+...(2) One is to increase the n valence to n+α valence, and the other is to decrease the m valence to m−β valence, thereby storing it as electrical energy. In particular, redox in which R=Q n=m=2 α=β=1 is practically preferable.

For example, Fe (bipyridine) (C104) (simply referred to as FBP), Fe (phenanthroline) (C10), Ru (
bipyridine)■, (C104)■ (simply referred to as RBP)
Ru (phenanthroline) (2) and (C104) (2) are preferred. In particular, RBP has excellent physical properties, but FBP, which is inexpensive and non-toxic, is preferred from a practical standpoint.

Further, as a solvent for this redox, pribilene carbonate (hereinafter simply referred to as PPK) or acetonitrile (CH3-CN) is representative of anhydride, that is, a stable non-toxic solvent with little electrode reaction.

Of course, water, a chromium bath solution, a mixture of the above-mentioned iron bath solution and chromium solution, etc. may be used as the redox in a system using a semiconductor as an electrode using the PIN junction of the present invention.

In particular, when water is used, hydrogen is generated as an example of a semiconductor having an N layer in contact with redox, and oxygen is released on the P layer side which receives holes or emits electrons. Therefore, it was possible to separate and generate oxygen and hydrogen by integrating it with a photoelectric conversion device using a non-single crystal semiconductor having a PIN junction, particularly an amorphous or semi-amorphous semiconductor.

Furthermore, the present invention has a significant difference compared to the conventionally known redox system using a platinum-titanium oxide semiconductor electrode system. That is, since titanium oxide has an energy band width of about 4 eV, only the ultraviolet rays of the irradiated light can effectively create electron-hole pairs and separate them. Therefore, the efficiency is extremely low for continuous light having a wavelength of 500 nm, such as sunlight. Therefore, in the present invention, Eg
(Energy band width) of 1.0 to 2.5 eV In particular, to generate electrons and holes with irradiation light, Eg = 1.5 to 1.
Another feature is that non-single crystal silicon having 8e■ is used. That is, this silicon, especially amorphous or semi-amorphous silicon, has a light absorption coefficient for light, especially sunlight, at a short wavelength of 3500 to 5000 A, which is 10 to 30 times larger than that of single crystal silicon.
The thickness of the I layer (intrinsic or substantially intrinsic conductivity type) in the IN junction may be 0.3 to 1μ, typically 0.5μ;
Furthermore, for the P or N layer, SixC1-x (0≦
x≦1) In particular, the optical energy band width for x = 0.2 to 0.7 is 1.8 to 2.8 eV and the I layer is 1.0 to 1.9 eV.
Typically, it can be set wider than 1.7 to 1.8 eV. Therefore, this P or N layer has a density of 50 to 500
For this reason, light is first irradiated from the side of the transparent substrate, especially the glass substrate, and the semiconductor having the PIN junction provided on the electrode of this substrate has a total thickness of 0.3 to 1 μm. Since it does not need to be thin, redox could be placed in close contact with the window surface of this semiconductor.

FIG. 1 is a vertical sectional view (A) of a redox system in a titanium oxide system showing the conventionally known Honda-Fujishima effect and a corresponding energy band diagram (B).

In the drawing, a semiconductor (5) made of titanium oxide provided on a metal substrate (6), a counter electrode (4), and a redox (1) such as water is filled between the two electrodes. When this area is irradiated with light, especially ultraviolet light (10), it becomes as shown in Figure 1 (B).
As shown in the figure, electrons and holes are generated in titanium oxide (5), and at the redox interface (9), the holes in N-type titanium oxide react with water to generate oxygen.

Furthermore, hydrogen is generated at the interface (11) of the counter electrode (4), making it possible to accomplish a photochemical reaction using a so-called semiconductor electrode.

However, in this conventionally known system, the optical energy band width (referred to as Eg) is 3 to 3.5 e Therefore, only the ultraviolet rays of sunlight can be used.

Therefore, the efficiency with which the incident light energy is effectively used for oxidation/reduction reactions (simply called redox reactions) is 0.1.
% or less, the efficiency was extremely poor and it was impossible to put it into practical use at all.

The present invention eliminates such drawbacks and makes effective use of irradiated light.
The main component is a non-single-crystal semiconductor made of micro-polycrystals with a size of A, and hydrogen or fluorine is added to these non-single-crystal semiconductors to neutralize recombination centers.
It is characterized by using a material to which 0.1 to 30 mol% of a halogen element such as chlorine is added.

Furthermore, the semiconductor in contact with the redox surface is difficult to dissolve.
Another feature is that ixC1-n (0≦x≦1) (hereinafter simply referred to as silicon carbide) is used. In addition, this handmade material allows a PIN junction to be provided, and the P layer and N layer are 1.6 to 3.5 mm thick with a thickness of 50 to 500 A. Wide E of OeV
silicon or germanium or SixGe1- which has an intrinsic or substantially intrinsic conductivity type.
Another feature is that n (0≦x≦1) is used.

Since a semiconductor having such a PIN junction has a non-single crystal structure, its light absorption coefficient is 10 to 30 times larger than that of a single crystal structure, and furthermore, it has a light absorption coefficient of 1.5 to 1, which is optimal for generating electron-hole pairs with irradiation light. Since .8 eV was used, more than 70% of sunlight could be absorbed with a semiconductor thickness of 0.3 to 1 μm. Therefore, an extremely significant feature is that it has become possible to accumulate energy in the redox system using irradiated light with an efficiency of 10% or more.

The present invention will be illustrated below according to the drawings.

Figure 2 shows a non-single crystal semiconductor (
12) and its N or P layer (16), a redox (1) is arranged, and a counter electrode (28) is provided.

In this drawing, unlike the conventional semiconductor electrode reaction using redox, the incident light (10) is performed from the transparent substrate (30) side, and the electrons or holes generated in this semiconductor (12) react with redox (1). at the interface (9), causing a redox reaction.

Particularly in this case, the semiconductor (15) is made of non-single crystal silicon made by plasma CVD as an intrinsic or substantially intrinsic semiconductor, and the P or N layer (14), the N or P layer (
16), carbon or silicon carbide with SixC1-x (0≦x<1) was used to improve the dissolution resistance as a semiconductor surface. In particular, the P or N layer (14) on the incident light side has a thickness of 1.6 to 3e to reduce absorption loss of the irradiated light.
It is extremely important that V is typically 1.8 to 2.3 eV, and a thickness of 50 to 200 A was sufficient. The electrode (20) in this example is made of ITO (a mixture of indium oxide and tin oxide) or has a 100 to 3
A double structure in which tin oxide was formed to a thickness of 00A or a light-transmitting electrode made of tin oxide was used.

The thickness of the I layer is typically 0.4 to 0.3μ.
A semiconductor mainly composed of silicon and having a non-single crystal structure, particularly an amorphous or semi-amorphous structure, having a thickness of 6 μm was used. A semiconductor represented by SixGe1-x (0<x<1) may be used for this purpose.

Furthermore, 50-500A for N or P layer (16)
carbon or silicon carbide having a thickness of . This semiconductor has an Eg of 1.7 to 3. OeV, typically Si
x=0.1 to 0.8 in xC1-n (0≦x<1)
was made possible. The semiconductor having this electrode interface (9) is hardly soluble due to a redox reaction, and if the redox substance is an oxide such as water, it is extremely important that it does not oxidize and become an insulating film. Therefore, the use of silicon carbide or carbon instead of silicon is a major feature in the practical application of the present invention.

In particular, silicon is typically used in general semiconductors, but
Such silicon is not preferred because it forms an oxide, which is an insulator, on the surface.

Further, in FIG. 2 (4), an aqueous solution, typically water, is used as the redox. Furthermore, tin oxide was formed on the surface as a counter electrode (28) to form an electrode. Then, among semiconductors (
16) can be made N-type to serve as a cathode, and a counter electrode can be provided as an anode. With such a structure, both the semiconductor electrode and the counter electrode are stable, hydrogen can be released from (43) on the cathode side, and oxygen can be released from (44) on the anode side (28).

In such a structure, since hydrogen is emitted from the semiconductor side, the semiconductor electrode does not melt, and since the tin oxide on the anode side is an oxide, it does not deteriorate even when oxygen is generated, resulting in extremely high reliability. I was able to have it. In particular, with this structure, the irradiated light does not pass through the aqueous solution, so it is not reflected on the water surface, and it can be decomposed into oxygen and hydrogen using electrical energy generated by a so-called solar cell and stored. Ta. The conversion efficiency is not 1% or less, which is conventionally known, and in particular, to generate oxygen and hydrogen, the load (25) may be set to 0Ω and turned on. In that case, the conversion efficiency is 10 to 18
% as a non-single crystal semiconductor, and the substrate (
30) 10 cm of PI to be provided on the upper transparent electrode
It was possible to manufacture a semiconductor having an N layer for 100 to 300 yen, and its contribution from an industrial perspective was extremely large.

FIG. 3 shows the energy band diagram at FIG. 2 A-A'. The numbers correspond to FIG. 2(A).

FIG. 2(B) shows another embodiment of the present invention.

That is, a transparent electrode (20) is placed on a transparent substrate (30),
Semiconductor (12), P or N layer (14), I layer (15)
, an N or P layer (16), and a counter electrode (28),
The same materials as in FIG. 2(A) were used for them. Redox (1) has a semi-permeable membrane such as an ion conversion membrane (7) in its center.
A redox (2) on the semiconductor electrode side and a redox (3) on the counter electrode side are provided.

If the irradiated light (10) is used as output as it is, a load (24) may be used, and if the redox reaction is to occur and energy is stored in the redox (1), the load (24) may be used.
25) is set to 0Ω, and a current is applied to turn it on. In addition, a load (23) may be used to release the stored energy at night or the like.

In this FIG. 2(B), a non-aqueous solution was used.

In particular, the solutes include Fe (bipyridine), (C104), (FBP)Fe
(phenacerorin or phenanthroline (phena
nthroline))■, (C10■)■Ru (bipyridine)■, (C104)■(RBP)Ru (phenaceroline)■, (C10■)■, and propylene carbonate acetonitrile CH3-CN was used as a solvent.

Especially due to its low cost and non-toxicity, FBP
was used. Although expensive, RBP has higher efficiency. In particular, the solubility of FBP in the solvent propylene carbonate is extremely high, making it possible to store a large amount of energy in a small volume.

By doing so, in the general formula Rn-1→R(n+α)+anode Qn-1→Q(n+β)+cathode, in particular, the reaction where R=Q=FBP n=m=2 α=β=1, That is, the reaction of FBP2+→FBP3+anode and FBP2+→FBP1+cathode could be carried out. In this case, the load (
25) should be turned on close to 0Ω. Moreover, when releasing the accumulated energy, the load (23) could be used to cause the reverse reaction. Moreover, if it is used simply as an output of a photoelectric conversion device, a circuit system of a load (24) may be used. The electrode (15) on the semiconductor was made of tin oxide or the like in a comb shape.

In this way, energy can be stored in a non-aqueous solution, especially on the back side of a photoelectric conversion device installed on a roof, etc., which is convenient for supplying power for nighttime lighting in general households, and further stabilizes the output power. It has the great feature of being able to be supplied with

FIG. 4 shows another embodiment of the present invention in which the redox reaction system using a non-aqueous solution shown in FIG. 2(B) is further developed.

In the drawing, it has a complementary structure in which semiconductor electrodes are used on the cathode side and the anode side.

That is, in one semiconductor electrode (12), the N layer (19
), an I layer (18), and a P layer (17). In this complementary structure, the semi-permeable membrane, especially the ion conversion membrane (7) is used, and in order to store energy by light irradiation, the load (u) should be close to 0Ω and the switch (26) should be turned on. do it. Then, due to the irradiation light (10), the redox (1) causes a reaction similar to that shown in FIG. 2(B) on the cathode side (2) and the anode side (3).

If the stored energy is used, the load (2
3) should be repeated.

An energy band diagram corresponding to the cross section taken along line A-A' in this drawing is shown in FIG. 5 with corresponding numbers. In the drawing, electrodes (20) and (21) in contact with the semiconductor are transparent conductive electrodes, and emission electrodes (28) and (29) are made of, for example, tin oxide. This extraction electrode can be used not only for tin oxide, but also for tungsten oxide, niobium oxide,
Strontium titanate, titanium oxide, iron oxide, iron titanate, and P or N type silicon carbide can be used. This electrode material is (28) or (43) in FIG.
) can be similarly applied. The rest is the same as in the case of FIG. 2(A) or (B).

FIG. 6 shows another embodiment of the invention. In this drawing, similar to FIG. 2(B), semiconductor electrodes are provided with comb-shaped electrodes, and a semi-permeable membrane (7) is provided as a complementary structure. Switch (16) for redox reaction
The photoelectric conversion device generates a load (3
5) and (34), and the release of the stored energy can be utilized using a load (23). Of course, the discharge of stored energy may be performed using the electrodes (46), (47) instead of the electrodes (28), (29).

FIG. 7 shows a further development of the semiconductor device of FIG. 2(B). That is, the semiconductor (12) gives electrons or holes to one of the redoxes in the redox (1) by the irradiation light (10), and the holes or electrons are transferred to the electrode (
13). When this electrode (13) becomes an anode or a cathode, it may be made of P or N type silicon carbide, and is made and used in the same manner as the semiconductor (12) by providing non-single crystal silicon carbide on a metal plate by plasma vapor phase method. I liked that. Thus, the redox reaction could be carried out by bringing the load (22) close to 0Ω. Also, if this irradiation light is sufficient, new redox is generated (41), (42)
), for example, FBP can be supplied, and the reacted redox can be stored in other storage tanks as FBP and FBP from (43) and (44), respectively. By storing such a redox reaction solution in another storage tank, energy could be stored for a long time and on a large scale. Further, the current may be supplied through a load. This technique is shown in FIG. 2(B).Furthermore, in the present invention, a non-single-crystal semiconductor having a PIN structure is used, and redox contacts the N or P layer of this semiconductor to increase the efficiency of the reaction current. This is a major feature. However, even if the entire surface of this semiconductor is coated with titanium oxide, tin oxide, etc., a redox reaction can be caused in the same way. In particular, since tin oxide and titanium oxide are N-type, it is convenient to form them on the surface of the N-type semiconductor of the PIN junction.

[Brief explanation of drawings]

FIG. 1 shows a semiconductor device for redox reaction using a conventional semiconductor device and its energy band diagram. FIGS. 2(A) and 2(B) schematically show the semiconductor device of the present invention. FIG. 3 is an energy band diagram corresponding to FIG. 2(A). FIG. 4 shows another semiconductor device of the present invention. FIG. 5 is an energy band diagram corresponding to FIG. 4. 6 and 7 show other semiconductor devices of the present invention.

Claims (1)

[Claims] 1. A first semiconductor having a PIN junction, a redox layer in contact with a P or N layer constituting the junction, and an N or P layer in contact with the redox and forming a pair with the P or N layer. a second semiconductor having a second semiconductor, a film semiconductor provided within the redox and separating the two semiconductors, and first and second electrodes in contact with each redox separated by the film. semiconductor device. 2. In claim 1, redox is caused by light irradiation as Rn4→R(n+α)+ Qn+→Q(n+β)+ where Q is the molecule n, m is the ion valence α, and β is changed by light irradiation. What is claimed is: 1. A semiconductor device that causes a reaction represented by a valence equation, causes a reverse reaction when light irradiation is stopped or weakened, and extracts electrical output from first and second electrodes. 3. In claim 2, R and Q are Fe.
1. A semiconductor device comprising a propylene carbonate solvent containing a solute in which n and m are divalent and α and β are monovalent (pipyridine). 4. In claim 1, in the semiconductor having a PIN junction, the P layer and the N layer have an optical energy band width of 1.6 to 3.0 eV and are of intrinsic or substantially intrinsic conductivity type. A semiconductor device characterized in that the I layer having a voltage of 1.0 to 1.9 eV is formed of a non-single crystal semiconductor. 5. In claim 4, the I layer is mainly made of silicon or germanium, and the P or N layer is made of Si.
A semiconductor device characterized by having xC1-2 (0<x<1) as a main component.