JPS6041269A - Semiconductor device - Google Patents

Abstract

PURPOSE:To obtain a continuous junction in a nonsingle crystal semiconductor device having a junction by adding carbon, nitrogen or oxygen to at least one of semiconductor layers. CONSTITUTION:A semiconductor device in which nonsingle crystal semiconductor layers made of silicon, germanium or silicon carbide and having 0.2-200atom% of hydrogen or halogen element are laminated, and PIN, PNP, NPN, NIN, or PIP junction is formed, carbon, nitrogen or oxygen is uniformly diffused and added to at least one of the layers. Since such an element is added, the semiconductor material has a wide energy band width, and a junction in which adjacent semiconductors are continued is composed.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is characterized in that - non-single-crystal semiconductors having a conductivity type, that is, amorphous or polycrystalline semiconductors, are stacked in three layers, and at least one of them is a semiconductor material constituting this semiconductor. PIN, PNP, NPN, N
The present invention relates to providing non-monocrystalline semiconductors, such as amorphous or polycrystalline, with IN or PIF junctions.

The present invention provides an amorphous or polycrystalline semiconductor such as silicon, germanium, or silicon carbide, and a non-single-crystalline semiconductor in which carbon, nitrogen, or oxygen is added as an additive sufficiently uniformly dispersed in the semiconductor. At or near this boundary where semiconductors having conductivity types are adjacent to each other, at least one non-single crystal semiconductor has (2) different energy hands from the adjacent semiconductor, and the transition between each energy band is continuous. Concerning what is done.

Furthermore, a PN or PI or NI junction is provided at or near this boundary, resulting in a PIN of the structure of the present invention.
, PNP, NPN, NIN or pH" junction and irradiating this junction with light to generate a photovoltaic force.

Conventionally, when semiconductors having different energy bands are brought into contact with each other, a so-called heterojunction is formed at the boundary. For example, when GaP and GaAs are bonded together, since they are both single crystals, the interface between the two energy gaps or energy gaps (hereinafter referred to as Eg) has a mismatched step shape as seen in Figure 1. A heterojunction has been formed. Due to inconsistency, the other row is Ga
In the junction of Al8 (1) and GaAs (2), a notch (3) in the conduction band (8) and a jump (4) in the valence band (9) occur, and in addition, an interface state (5) occurs at the boundary. ) has occurred.This interface state (Interface 5tate
s (hereinafter referred to as Ns) (3) Therefore, electron or hole carriers meet this Ns at this junction, recombine, and disappear.

As a result, the lifetime of carriers is reduced, and when an attempt is made to use this junction to generate a specific effect, such as photovoltaic force, the photo-excited charge will disappear before generating photovoltaic force. There was a big drawback. Furthermore, when attempting to obtain the characteristics of a semiconductor PN junction diode, the reverse pairing breakdown voltage is weak and the diode becomes a soft diode. Figure 1 (A> is for a 71-P junction, but in Figure 1 (B) which is an N-N junction,
A major drawback was that spikes (6) were generated in addition to Ns (7), which hindered the movement of electrons, which were majority carriers. The present invention prevents the occurrence of such nicks, jumps, and spikes. That is, a major objective is to cause the energy band to change continuously at this junction. Furthermore, the present invention has eliminated or extremely reduced the generation of Ns caused by dangling bonds and crystal defects, which have conventionally existed in heterojunctions due to crystal lattice misalignment at the interface (4). It is characterized by By having such a structure, that is, a continuous junction from an energy band perspective, it has become possible to develop new semiconductor devices that make use of this energy band difference.

The present invention will be explained below based on examples.

The present invention applies to semiconductors having amorphous (pure amorphous or polycrystalline in the order of 5 to 100 short range) or polycrystalline structures such as silicon, germanium, and silicon carbide having one conductivity type. The basis of the present invention is to uniformly disperse and add carbon, oxygen, nitrogen, etc.

Furthermore, uniform dispersion in the present invention refers to a direction in which the quantum theoretical waves of the additives locally interact with each other.

When a film is formed on a composite substrate partially or completely coated with a metal film, etc. on a metal, semiconductor, insulator, or insulator-F such as glass or ceramic,
A material to be a semiconductor, such as silicon (5) element, is formed as a film using silane, dichlorosilane, or other silicide gas. For this reason, a silicide gas such as silane or dichlorosilane and a carrier gas such as hydrogen are placed on the entrance side of a reactor made of heat-resistant glass such as quartz or stainless steel, and impurities that determine conductivity in the semiconductor such as phosphorus, arsenic, and boron. can be introduced using phoshine, arsine, and diborane. In addition, carbide, nitride, and oxide gases such as methane, ammonia, and oxygen can be mixed in. In addition, a vacuum pump was used for evacuation, so that the inside of the reactor could be evacuated to 0.001 torr.

The substrate is held in a susceptor and put into the reactor, the reactor is evacuated to 0.1 = 10 torr, and the substrate is subjected to high frequency heating of 1 to 50 MHz or a combination of it and radiation heating, and then Excited or decomposed a reactive gas. These reactive gases are formed as a film on the substrate. At this time, this film is kept at room temperature -500°C below the temperature of the substrate.
The structure was amorphous at temperatures up to 350° C. and polycrystalline at 350° C. to 900° C.

If the substrate has a single crystal and the film on it is epitaxially grown at 900(6)°C or higher, it will become a single crystal, but it has been experimentally proven that these single crystal semiconductors do not have the structure of the present invention. It was possible. The first feature of the present invention is the use of a non-single crystal coating. When impurities exhibiting N-type conductivity in semiconductors such as phosphorus and arsenic are mixed into this non-single crystal film at a concentration of 1014 to 1022 cm-3 using phosphin (PH3> and arsine (Asllt)), the so-called On the other hand, adding diborane (B4■t) at a similar concentration results in a P-type semiconductor.Furthermore, if these impurities are not added at all, the intrinsic or device bank ground level It has become a so-called essentially intrinsic semiconductor due to the mixing of impurities.This non-single-crystal film contains 0.2 to 200 at. It is added in concentration.

These have the effect of binding to dangling bonds of silicon, suppressing the generation of recombination centers, and electrically neutralizing (inactivating) them. Addition of the hydrogen or halogen element (7) at the same time as the formation of the semiconductor film or after the film formation was an extremely important element for commercializing the present invention industrially. In the present invention, the addition of an impurity for neutralizing the recombination center is achieved by electrically activating a reactive gas and simultaneously activating the added hydrogen or halogen element. Further, in embodiments of the present invention, carbon,
Nitrogen and oxygen were uniformly dispersed and added to the semiconductor. The carbon used was CI, C1■C. Nitrogen is ammonia (NH
), hydrazine (Ni2), and oxygen was HO or OL. These mixtures include N, O, Not
IcH, OR, other alcohols, COLICO, etc. are introduced into a reactor using a carrier gas such as hydrogen, and additives are added such as nitrogen and oxygen or carbon and oxygen.
More than one type may be added. If you try to add oxygen, nitrogen, etc. after forming a single crystal semiconductor film, silicon oxide (Eg = 8eV) or silicon nitride (Eg = 5.5eV)
) and was nothing more than an insulator. However, if these additives are added electrically or using a combination of electricity and heat (8) at the same time as the silicon film is formed, the semiconductor becomes 1. 1eV to 3eV (SiC>, 5.5e
An intermediate value between V (Si1 N4 ) and 8 eV (Sift) could be obtained. The Eg of this film was measured by photoluminescence or optical excitation method.

Since this Eg has a non-single-crystal structure in both semiconductors, there is no specific Ns only at the interface, as is known in single-crystal heterojunctions, and the energy band is the conduction band and the valence band. Both electronic bands were able to form independent step-like continuity or smooth continuity.

The degree of Eg at these different joints is determined by the film formation rate of 0.
The doping rate was adjusted to 1 to 10 μ/min, and the doping amount of the additive was adjusted to 0.
This was achieved by adjusting to N10FF or continuously adjusting the steps. But the important thing is that this is different! ! At or near the boundary of Eg, Ns does not occur due to lattice mismatch, etc. seen in heterojunctions of single crystal semiconductors, although this may also be due to the manufacturing method, and the conduction band and valence band, which are the edges of Eg, do not occur. There were no notches, spikes, etc.

(9) or was found to be substantially non-existent. This is Eg
This is presumed to be due to the addition of hydrogen or halogen element impurities and the fact that they are determined according to the stoichiometric ratio. The above is an example using a low pressure CVO (chemical vapor deposition) method or a glow discharge method.

In the present invention, it is not necessary that one of the two semiconductors having different Eg be a pure semiconductor and the other only a semiconductor to which additives have been added. In both cases, the same type of additive is used in different amounts, for example, one is 1015, . 1
018 C111-3 If the other is added in an amount of 0.01 to 30%, the present invention can be carried out. Furthermore, one of the carbon
It goes without saying that it is sufficient to change the type of additive, for example, from 5 to 10%. As is clear from the above-mentioned theory, implementation method, and results, the present invention is effective against notches and spikes, which are conventionally unexpected elements that occur when materials of different Eg are joined at or near the junction, which is extremely important for the operation of semiconductors. etc. and the interface-specific Ns are eliminated,
This is essentially due to the elimination of (10) factors that are attributable to the joining of materials with different lattice constants. Therefore, an important gist of the present invention is a semiconductor having a non-single crystal structure that eliminates lattice misalignment in a microscopic sense.

By having such a non-single crystal structure and neutralizing the recombination center with hydrogen or halogen, it was possible to complete a semiconductor device having a so-called continuous junction, in which the energy gap is continuously changed according to the stoichiometric ratio. .

Figure 2 shows E in two non-single crystal semiconductors that led to the present invention.
This is an example in which g is changed. In Figure 2 (A), the junction is the boundary, and the non-single crystal semiconductor (11) is N type and W (WID
E) Eg (wide energy gap), the non-single crystal semiconductor (13) is of P type with L (NALI, ow) 17g (narrow energy gap). Figure 2 (B) shows the same type of P conductivity type, in which the non-single crystal semiconductor (11) is W.
-t+g and the non-single crystal semiconductor (14) is L -E
It is g. Furthermore, FIG. 2<c) and (D) are a PP junction and a PN junction, respectively. FIG. 2(E) shows a smoothly continuous NP junction.

FIG. 2(F) shows a stepwise NP junction.

(11) FIG. 3 shows an embodiment of the structure of the present invention in energy bands. That is, this invention applies to PI, NI or PN.
Have two junctions, PNP junction, NPN junction, PIN
It has a conductivity type of a junction, a pH' junction, and an NIN junction, and has a different energy band than at least the adjacent semiconductor among the three semiconductors.

Figure 3 (A) shows W (Eg,) L (Egt) W
(Eg,) is an NPN l-transistor. L's Eg
The P-type base region of , can promote Eg-determined recombination. That is, in the emitter (30〉, base (31), and collector (32), the minority carriers of electrons can easily drift (15) to EgL from [!g, but the reverse diffusion causes holes to This results in a bipolar transistor with excellent frequency characteristics.

Figure 3 (B) is L (Eg+) W (Egz)
, L (Eg7) PNP) transistor. That is,
The impurities are also removed between the source (33) and drain (35) (
A controlled current could be passed through the channel forming region (12) provided on the outside without mixing in the channel (34). FIG. 3(C) shows the NTr' configuration of the L-W-L NIP.

FIG. 3(D) shows a W-V/-L l''IN configuration.

This can be expected to provide high conversion efficiency (15 to 30%) for so-called photocells or solar cells that emit light depending on the W value. The P layer (36) on the incident light side has a wide Eg
Therefore, the optical loss here is small (, one layer of the active layer (
37) allows effective photoelectric conversion. In addition,
These generated carriers are also electrons (40) and holes (40').
) is also a continuous bond, so there is no failure on the bonding surface, and N (3
8), it was possible to drift P (36).

In addition, it had the effect of preventing the backflow of electrons (40 P (36)), resulting in a significant improvement in efficiency.

Figure 3 (E) is W-W-L NPN, (F) is L-W
--W PNP) transistor.

FIG. 3(E) and <F) were effective in improving the ultra-high frequency characteristics as in FIG. 3(A).

Further specific examples are shown below.

(13) Specific example 1 This specific example has the structure shown in FIG. 3(D).

A P-type non-single crystal layer (36) is formed on the electrode on the glass with 5 carbon atoms.
Silane, methane, and diborane were mixed to a concentration of ~30 at % and formed by a glow discharge method. Add 1 to 5 carbons
A ■-type non-single-crystal layer (37) was formed by a glow discharge method using silane and methane so that the silane and methane were added at a concentration of 5%. Next, silane and phosphine were mixed to form an N-type non-single crystal semiconductor layer (38), and two junctions were provided. Finally, an aluminum electrode was provided, and the conversion efficiency of this solar cell was measured with the structure shown in Figure 3 (D), and was found to be 2.07% (open circuit voltage 0.
57V.

Short circuit current 0.95mA, fill factor 0.387. Area 0.
10 cd, series resistance 353Ω, parallel resistance 2089Ω).

Specific Example 2 In this specific example, a metal film electrode is provided on a ceramic having the structure shown in FIG. 3(A), and a non-single crystal semiconductor layer (20
) by a glow discharge method using a mixture of silane, methane, and phosphin (14) and a P-type non-single crystal semiconductor layer (31) with silane and a mixture of silane and diborane (14). The fJi layer was formed by mixing methane and phosphin. This NP
N type Jtij (20) with N l-transistor 2.3
eV, P-type layer (31) is 1.5eν, N-type layer (31)
2. OeV in the energy band lJ. Its frequency characteristics exhibit a high frequency characteristic of 12 MIIz at 1.5V between the emitter and base and 10V between the heath and collector.

This additive may be determined depending on the purpose of the application.

However, these are merely means for further industrially disseminating the present invention.

In the present invention, silicon, germanium, and silicon carbide are used as semiconductor materials. However, other so-called GaA
It goes without saying that a compound semiconductor such as S, GaAIP, or GaP may be used. In addition, the antireflection coating in photosensitive devices such as solar cells has a thickness of λ/4, which depends on n (n is the refractive index of the semiconductor), and is doped with hydrogen or a halogen element, as well as the additive C.
, N or 0 may be sufficiently increased and used as an insulator for lower silicon nitride or lower silicon oxide (SiO or 5iOx) (15).

As is clear from the above description, in the embodiments of the present invention, semiconductors are mainly made of silicon. However, the present invention is not limited to silicon, and can also be applied to germanium, silicon carbide, etc. according to the applied semiconductor device. The objective is to achieve appropriate control of g, and in order to put this into practical use, a non-single crystal semiconductor to which hydrogen to neutralize Ns or a halide such as chlorine is added at a concentration of 0.1 to 200 atomic %. It was used as a basic material in
Additives such as nitrogen and carbon are stoichiometrically 10+5 to 102
For example, 0.1 to 80% carbon, 0.01 to 10% nitrogen, and 101S to 10210% oxygen.
2O was added stepwise or continuously, so even if semiconductors with different Eg are adjacent to each other,
The generation of Ns due to lattice misalignment etc. could be suppressed at the interface. Furthermore, the conductivity types of P-type, N-type, The possible glow (16) lies in the fabrication using discharge or reduced pressure chemical vapor deposition (CVD).As a result, the thickness of one semiconductor can be reduced to 0.0
The concentration of P or N type impurities can be freely controlled within the range of 1μ to lOμ, and the concentration of P or N type impurities can be controlled within the range of 1014 to 1022cm-9. It was found that multilayer junctions can be easily fabricated. In addition, large-scale production can be carried out continuously in the same reactor, which is a major feature that opens the way to completely new industrial fields.

[Brief explanation of the drawing]

FIG. 1 shows an energy band diagram of a conventional heterojunction. 2 and 3 show an embodiment of the invention. (17) 1 palanquin

Claims (1)

[Claims] 1. The first, second and third non-single crystal semiconductors to which hydrogen or halogen elements are added are r'IN and I'NP. At least one of the semiconductors provided to constitute an NPN, NIN or PIF junction is characterized in that an additive capable of changing an energy band is added to another semiconductor material adjacent to the semiconductor. Semiconductor equipment. 2. First, second and third non-single crystal semiconductors doped with 0.2 to 200 atomic % of hydrogen or halogen elements are provided in a stacked manner, and the first, second or third non-single crystal semiconductor The semiconductor is made of silicon, germanium or silicon carbide,
Another first, second or third semiconductor adjacent to the semiconductor has a wide energy band in which the semiconductor material is doped with carbon, nitrogen or oxygen, and the semiconductors adjacent to each other continuously form two junctions. (1) A semiconductor device characterized by being provided with the following. 3. A semiconductor device according to claim 1 or 2, characterized in that the semiconductor has a thickness in the range of 0.01 μm to 10 μm.