JPH0329378A - Manufacture of electronic device using diamond - Google Patents

Abstract

PURPOSE:To enable a sufficient visible light to be emitted by forming a diamond on a substrate which is equipped with an insulating surface and a semiconductor on the diamond selectively, by forming an impurity region by adding the impurities to the semiconductor as a mask, and by forming a pair of electrodes on the impurity region and the semiconductor. CONSTITUTION:A diamond 2 is formed on a substrate 1 which is equipped with an insulating surface, a semiconductor as a buffer layer 3 is formed on the upper side of the diamond 2, the semiconductor is eliminated selectively, and impurities are added to the upper part of the diamond 2 of the eliminated region selectively for forming an impurity region 5. An electrode 9 is formed at one part on the impurity region 5 through an electrode 29 or the buffer layer 3 and voltage is applied between a pair of electrodes 29 and 9 which are provided on the upper side of the substrate 1, thus enabling a visible light to be emitted from a part with the impurity region 5 of the diamond 2 as a center.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a method for manufacturing an electronic device using diamond, particularly a visible light emitting device.

"Prior Art" Regarding light emitting devices, red light emission has already been achieved for more than 10 years by using 1-2 compound semiconductors such as GaAs. However, this light-emitting element emits red light, and it is extremely difficult to emit blue or green light, and it is even impossible to emit continuous visible light such as white light using crystalline materials.

Attempts to make light-emitting devices using diamond have already been demonstrated by the present inventor, for example, in Patent Application No. 146 filed in 1982.
No. 930 (filed on September 17, 1982).

Considering the size of the market for light-emitting devices, diamond has the advantages of being heat resistant and extremely chemically stable, and the raw material is carbon, which is an inexpensive material.
The potential for its industrial mass production is extremely large.

``Conventional disadvantages'' However, it is difficult to make light-emitting devices using diamond stable.
Moreover, no method for producing it with high yield or the structure necessary for it has been shown so far.

A conventional visible light emitting device using diamond has a structure in which one electrode is provided below the substrate and the other electrode is provided above the diamond, allowing current to flow in the vertical direction. However, when diamond has a polycrystalline structure, current flows locally in areas where current flows more easily, such as grain boundaries, and a large amount of heat is generated in the current concentrated area, making it difficult to see We investigated the drawback that it does not emit light. As a result, the following facts were found.

In the vertical flow method, there is too much variation in manufacturing yield. Since the ohmic junction or Schottky junction at the electrode section does not have a sufficiently stable function, a voltage higher than necessary must be applied. Furthermore, the degree of Schottky junction in the voltage varies from element to element, and a high manufacturing yield cannot be expected.

Additionally, diamond is generally easy to make into I-type (intrinsic) and P-type conductivity types, but it is extremely difficult to make diamond with N-type conductivity, and as a result, PIN or PN junctions are constructed using only diamond. It was difficult to do so.

Furthermore, no artificial control method has been proposed for the recombination center that controls the light source.

OBJECT OF THE INVENTION The present invention is designed to obviate such drawbacks. That is, diamond was formed in the form of a thin film on a substrate having an insulating surface, a pair of electrodes were disposed above the diamond, and a current was passed in the lateral direction to further reduce the influence of polycrystalline grain boundaries. Furthermore, a semiconductor of silicon or silicon carbide having an N-type or P-type conductivity is arranged as a buffer layer between the electrode and the diamond having a low-resistance light-emitting region. By embodying this N-type conductivity type, which is difficult to form in diamond, with silicon or silicon carbide, we attempted to achieve current injection by constructing a PN or PIN junction since the luminescent center is in diamond. Further, in the present invention, an impurity region that emits light using this semiconductor is intentionally provided using a self-line process.

One of the technical ideas of the present invention is that if impurities are added externally to a region that should emit light to form a controlled formation, the electrical resistance of this region will be one order of magnitude higher than that of a region to which no other impurities are intentionally added. They discovered the physical property that they are smaller and allow current to flow more easily in a concentrated manner, and tried to proactively apply this to the construction of electronic devices. Then, carriers (electrons or holes) are efficiently injected into the luminescent region of the diamond by applying a voltage between a pair of electrodes, and recombination occurs between the luminescent centers, between bands (between valence bands and between single valence bands), or between the emission centers. The luminescence center is located between one band (conduction band or valence band).

``Structure of the Invention'' The present invention relates to a method of manufacturing an electronic device for emitting visible light by forming diamond on a substrate having an insulating surface and passing current laterally thereon. The present invention
A single or multilayer layer (hereinafter also referred to as a buffer layer) of silicon carbide (SixC.-.0<X<1) or a semiconductor such as silicon having a P or N type on the upper surface of the diamond, and on this buffer layer. Conductive electrodes are provided in a strip-shaped, comb-shaped, donut-shaped, etc. pattern. Impurities are implanted into the diamond in the region where there is no buffer layer by ion implantation or the like using this electrode as a mask and controlling the accelerating voltage in a self-aligned manner.

Ion implantation is the preferred method for creating luminescent centers with a uniform concentration because it can be implanted without causing any difference in the concentration of added impurities at grain boundaries or in the bulk, regardless of the shape or morphology of the diamond. .

The region to which this impurity is added, ie, the impurity region, becomes a light emitting region. Furthermore, another electrode is provided on the upper surface of this impurity region. When an electrode is provided here, an electrode may be provided on the buffer layer at the same time. By applying a pulse, direct current, or alternating current between the pair of electrodes formed on the upper side, visible light is generated, particularly in the impurity region. This impurity region, that is, the light emitting region, does not exist under the buffer layer, and in the present invention, impurity ions are implanted in a self-aligned manner into the region where the buffer layer does not exist, using the buffer layer as a mask. This is used as an impurity region. This impurity region is annealed as necessary to form another electrode in a part of the impurity region. Then, only two types of photomasks are required to manufacture the electronic device of the present invention, and an extremely high manufacturing yield can be expected.

The present invention uses a P- or N-type semiconductor as the buffer layer, and in particular, forms a semiconductor of silicon, silicon carbide, or a multilayer thereof, and as a result, the semiconductor layer is interposed on the diamond, and heat treatment is performed at 500°C or higher. By providing a pair of electrodes that can allow current to flow in the lateral direction without applying after forming the pair of electrodes, reliability under long-term actual use conditions has been improved. That is, the structure is as follows: upper electrode = P-type or N-type semiconductor (for example, silicon or silicon carbide); diamond without an impurity region; diamond with an impurity region that becomes a light emitting region; upper electrode provided on the impurity region; or An upper electrode was formed via another buffer layer. The structure is such that the upper electrode and the diamond to which no impurities have been added are not in direct contact with each other, and the other electrodes are in close contact with the diamond to which impurities have been added at a high concentration, creating a stable bond between the diamond and the semiconductor. It is.

Furthermore, in order to more effectively generate blue light emission, the present invention adds Zn (zinc), an element of Group IIb of the Periodic Table of Elements, as an impurity added to the diamond. Cd (cadium), as well as O (oxygen), which is a group VIb element, and S
An element selected from (sulfur), Se (selenium), and Te (tellurium) was added by ion implantation. Further, a compound of carbon and OH such as methanol (CIIsOH) was used for the diamond coalescence.

Semiconductors contain B (boron), 81 (anoleminium), Ga (gallium), which are elements in the MB group of the periodic table of elements.
In (indium) or N (nitrogen), which is an element of the VB group. P (phosphorus), As (arsenic), and Sb (antimony) were added to make it P or N type. This may be added to diamonds, but the color changes from blue to green.

When ion implantation is used, it is possible to create damage in the diamond and simultaneously implant impurities into all parts at a uniform concentration, making it possible to create more recombination centers or emission centers.

If impurities were added only by a diffusion method, the impurities would tend to concentrate at grain boundaries, and some of the added impurities would be polarized at the grain boundaries, resulting in a decrease in activity, which was undesirable.

The region to which impurities are added by this implantation has electrical conductivity greater than 1 order of magnitude compared to the region to which no impurities are added. Therefore, when a voltage is applied between the pair of electrodes, injected carriers intentionally flow concentrated in this impurity region, and electrons and holes tend to recombine with each other via recombination centers. This recombination process allows light to be emitted.

When this ion implantation method is used, even if thermal annealing is subsequently performed at 200 to 1000°C in an oxygen-containing atmosphere, such as oxygen, NOx, or air, the damage remains and the strain energy in an atomic sense is reduced. Since it is only relaxed, the luminous efficiency can be increased by adding oxygen, which is an element of group VIb of the periodic table of elements, in addition to the impurities already implanted.

As a result, local concentration of current is prevented by flowing the current in the lateral direction, and the current flows uniformly in the impurity region added by ion implantation into the diamond, resulting in interband transition, band-to-band recombination centers, or luminescence centers. Recombination of carriers occurs due to transitions between recombination centers or between emission centers. The idea is to emit visible light according to the energy band interval (gap) of the recombination. In particular, the visible light can emit blue and green colors according to the energy band width of this transition. It is also possible to create continuous light such as white light by creating an energy level at the recombination center between the two.

The structure of the types and conductivity types of impurities that more actively emit blue light is shown.

On a substrate having an insulating surface, impurities of group 1b, such as (CHff)2Zn, are added to diamond together with ClhOI+ and hydrogen, and a film is formed by plasma vapor deposition.

A semiconductor serving as a buffer layer above the diamond is formed as an N-type semiconductor. The semiconductor is selectively removed, and on top of the diamond in the removed area, the periodic table of elements VI
Impurities of group b or group IIb, especially group VIb, such as S,
Se was selectively added to form an impurity region. An excellent method is to provide another electrode or another electrode via a buffer layer on a part of this impurity region, and to apply a voltage between a pair of electrodes provided on the upper side of the substrate.

As a structure of opposite conductivity type and a type of impurity, 0, S. A diamond doped with Se and Te is formed. And those diamonds are II 2S I II zSe + II 2
Te+ (cll z) 2 Sl (cll i)
2se + (Cllz)zTe is formed by adding CI+3011 and hydrogen into a diamond film using an ite-flame method. Further, the semiconductor serving as the upper buffer layer is of P type, and the impurity region is doped with group IIb or Vtb impurities, particularly group IIb impurities such as Zn,
Cd is added by ion implantation to form an impurity region.

An electrode or another electrode via a buffer layer is formed on this impurity region, and an electrode is also formed on the semiconductor serving as the buffer layer. It may be of a so-called reverse conductivity type.

The present invention will be described below according to examples.

"Example 1" In the present invention, FIG. 1 shows the manufacturing process thereof. As shown in FIG. 1(A), a diamond (2) was formed on a substrate (1) having an insulating surface on which a silicon nitride film was formed, using a magnetic field microwave CVD apparatus shown in FIG. A method for forming a diamond film using a magnetic field microwave CVD apparatus is disclosed in Japanese Patent Application No. 61-292859 (Thin Film Formation Method (H) filed on December 8, 1988, filed by the present inventor. The outline is shown below.

The silicon nitride film (1-2) is heated to zero by a known plasma vapor phase method.
．． A silicon semiconductor (1-1) substrate formed to a thickness of 1 to 0.5 μm was immersed in a mixed solution of alcohol mixed with diamond particles, and ultrasonic waves were applied for 1 minute to 1 hour. Then, many minute damages can be formed on the substrate (1) having this insulating surface. This damage can be the source of the subsequent diamond-shaped core. This substrate (1) was placed in a magnetic field microwave plasma CVD apparatus (hereinafter also simply referred to as a plasma CVD apparatus). This plasma CvD device uses a microwave oscillator (18) to generate microwave energy of a frequency of 2.45 GIlz up to 10 KW. Attenuator (16L quartz window (
45) into the reaction chamber (19). In addition, using Helmholtz coils (17) and (17'), the magnetic field is adjusted to a maximum of 2 to create a resonance surface of 875 Gauss.
．． Added up to 2KG. this? Board inside the file (1)
was placed on the holder (13) by holding the substrate (14).

In addition, the position within the reactor was adjusted using the substrate position moving mechanism (42), and the vacuum was drawn to 10-'' to 10-6 torr.After this, methyl alcohol (C I
I30H) or ethyl alcohol (Cz1l
A gas having a C-011 bond such as s01I), for example alcohol (22), is mixed with hydrogen (21) to a volume of 40 to 200 χ
(When it is 100% by volume, Ctl3011:II■-1:
1) and introduced.

If necessary, dimethylzinc (Zn (Cllx) z) was added to Zn(C!h)z/CH 3011 = 0. 5-3
The system (23) was uniformly added as χ (volume χ) into the capsular membrane. If you want to make this diamond P-type, add trimethylboron (B(CH3)l) as a P-type impurity to the system (
23) from B (CIl:l) :l/CI13011
= 0.5 to 3x was introduced to make the diamond P-type.

Furthermore, as a dopant, S, which is a Vtb group element,
When Se and Te are added, from the system (24), for example, (+1■S is (CL) zS)/C! hOtl
=0. 1 to 3"X may be added. For diamond growth, the pressure in the reaction chamber (19) is evacuated from the exhaust system (25) to 0.0L to 3 torr, for example 0.
．． It was set to 26 torr. A magnetic field of 2.2KG (kilogauss) is applied from (17) and (17') to the substrate (1).
) or its vicinity is set to 875 Gauss. I added 4KW to the microwave. In addition to this microwave energy, auxiliary thermal energy is added from the holder (13) to raise the temperature of the substrate to 200-1000'C, for example 80'C.
The temperature was set to 0°C.

Then, the carbon in the alcohol, which was decomposed and turned into plasma by this microwave energy, was elongated on the substrate, making it possible to elongate many single-crystal diamonds into columnar shapes. At the same time, in addition to this diamond, graphite fraction is also likely to be formed, which reacts with oxygen and hydrogen and revaporizes as carbon dioxide or methane gas. As a result, crystallized carbon or diamond (2) was deposited on a substrate (2 hours) with a thickness of 0.5 to 3 μm, e.g., an average thickness of 1.3 μm (filming time 2 hours), as shown in FIG. 1(A). 1) I was able to raise it.

That is, in FIG. 1(A), a substrate (
1) Diamond with Zn or B added on top (2)
or undoped (state where no impurities are intentionally added)
A diamond (2) was formed.

P-type conductivity type silicon or silicon carbide (S
ixC. 0 < At the same time, silicon carbide (SixC1
-, Q<Xd) to a thickness of 300 to 0.3 μm. This formation is performed using plasma CVD similar to diamond.
Create using equipment.

These films are doped with different impurities such as P-type and N-type, so a multi-chamber system is used to create a reaction chamber for diamond film and a reaction chamber for N-type semiconductor i1fc film, and connect them to each other for mass production. That is valid.

In FIG. 1(B), the buffer layer (3) was selectively removed using the first photomask to obtain FIG. 1(B). As shown in FIG. 1 (Il), this photoresist (4)
, (4''), using the buffer layers (3) and (3') as masks and using an accelerating voltage of 50 to 200 KeV, S or Se was implanted by ion implantation to I x 10''' to 5 x
The impurity region (5) was formed by doping to a concentration of 10"cm-', for example 6XIO19cm-'.
As shown in Figure (C), the ends of the buffer layers (3), (3') and the end (20) of the impurity region (5) can be made to coincide or approximately coincide with each other. Therefore, when a current is passed through the buffer layer to the impurity region, it is possible to prevent variations in the applied voltage due to variations in alignment accuracy from product to product. After that, the photoresist on the buffer layer (3) was removed. The whole was heat-treated in oxygen or air as necessary to remove lattice strain in the impurity region, and then oxygen was added thereto. The whole was immersed in dilute hydrofluoric acid to remove the silicon oxide components on the buffer N(3), (3'').

In the next manufacturing process shown in FIG. 1(D), 0.05 to 0.5 of molybdenum and tungsten (29-1)
A buffer layer was formed to a thickness of μm. At this time, (9-1) may be formed using the same material in the same process. Furthermore, on top of this, aluminum ξ (29-2), (9-2
) as an electrode member for wire bonding from 0.5 to
It may be formed to have a thickness of 2 μm.

Thereafter, this electrode member was selectively removed by photoetching using a second photomask (2), and the electrode (9
-1). (9-2) i.e. (9) and (29-1)
, (29-2), that is, (29) was formed. That is, a photoresist was selectively formed and removed by a known dry etching method using plasma.

Next, this electrode (9-2). Wire bonding (8) on (29-2). (2B) was applied. Furthermore, the entire structure was coated with a silicon nitride film (6) as an antireflection film.

This was carried out after providing a light emitting element on a lead frame and wire bonding. FIG. 1(D) shows this structure.

In addition, the whole thing is molded with translucent plastic,
It is effective to improve moisture resistance and mechanical resistance.

In the structure shown in FIG. 1(D), the voltage between the pair of electrodes (9) and (29) is 10 to 200 V (DC to 10 V).
0H2 duty ratio 1) For example, a voltage of 50V was applied. Then, a current (11) could be passed through the electrode (9), the buffer N (3), the diamond without an impurity region (2), the diamond with an impurity region (5), and the electrode (29). The resistance of the impurity region (5) is more than an order of magnitude lower than that of other diamonds to which no impurities have been added, so the current flows through this impurity rather than through the diamond which is also below this impurity region. A semiconductor (3
) and a region (impurity region) (5
) could emit light to the outside (upward).

That is, it became possible to emit visible light, particularly blue light of 475 nm±5nn+, from a portion of the diamond centered around the impurity region (5). The strength was 17 candelas/II12.

"Example 2" In this example, the perfect pattern is shown in FIG. 2(A). The manufacturing process is roughly the same as that of Example 1 shown in FIG.

That is, 0.5 to 3 μm on the substrate (1) having an insulating surface.
For example, undoped diamond was formed with an average thickness of 1.2 μm. Thereafter, a P-type silicon or silicon carbide semiconductor (3) (
SixC. -x O<X<1) was formed as Huffer N(3). After this, the semiconductor is selectively removed using a photo-etching method (of the first mask), and the buffer layer (3
) were left selectively.

Next, Zn, which is an element in Group B of the Periodic Table of the Elements, was deposited on top of the diamond (2) at 9.5
An impurity region (5) was created by implanting ions to a concentration of .

In this example, as the impurity for the luminescent center, an element of Group IIb, rather than Group VIb of the Periodic Table of Elements, was used as the main component.

Furthermore, without providing another huffer layer in this impurity region, aluminum was directly used as an electrode (29) with a thickness of 1.5 μm.
The thickness was set at . At the same time, the electrode (9) was also provided using a second photomask (2) and heat treated in the atmosphere at 400-500'C.

Aluminum is directly brought into close contact with the impurity region as one electrode.

Other steps were the same as in Example 1.

In this embodiment as well, a protective antireflection film (6) is formed on the impurity region (5).

A voltage of 40 V was applied between the pair of electrodes (29) and (9).

Then, blue light emission with a wavelength of 480 nn+ could be observed from this point. The intensity was 14 candela/mt, which was darker than Example 1. However, it was possible to put it into practical use.

"Example 3" A completed vertical cross-sectional view of this example is shown in FIG. 2(B). The manufacturing process is roughly the same as in Example 1. A large number of electrodes were provided in a comb shape to create a large-area light emitting element.

In Example 1, diamond (2) was formed on a substrate (1) having an insulating surface while adding oxygen.

On top of these are N-type silicon carbide semiconductors (3), (3”),
(3゛) was formed.

After this, the diamond (2) was added Se (
Selenium) was ion-implanted using an accelerating voltage of 50 to 200 KeV at I
'' to form an impurity region (5). Then, the end of this semiconductor and the end of the impurity region (20) are
I was able to get one loss or almost a match. Furthermore, P-type semiconductors (29-1), (29
-1') was selectively formed as another buffer layer.

This is annealed at 750 to 900''C in the atmosphere, and oxygen is added at a higher concentration to the impurity region (5), and oxygen, selenium, and two VIb group elements are added to eliminate the lattice strain. Ta.

Semiconductor (3), (3'), (3'' that forms the buffer layer)
) (29-1) and (29-1') were eluted with dilute hydrofluoric acid. Next, aluminum was applied to the electrodes (9), (9') . to a thickness of 2 μm on these semiconductors. (9
”)+(29-2).(29-2゛).

After scribing and breaking this electronic device and placing it closely on the lead frame or stem, wire bond (8)
．． (28) was formed.

Finally, the same silicon nitride film as in Example 1 was formed as an antireflection film (6). Because the light emitting area is large and because a buffer layer is interposed between both electrodes and the diamond,
In addition to having long-term stability, the wavelength is 490±10 nm.
It was possible to produce greenish-blue luminescence of 29 candela/mt.

"Effect" In the conventional light-emitting device using diamond that allows current to flow in the vertical direction, a voltage of 40V is applied between the electrode and the substrate.
However, in the present invention, the diamond is placed on a substrate with an insulating surface, and the diamond reaches a temperature of nearly 60"C, and because the upper electrode and the diamond are in close contact with each other, the diamond reacts and deteriorates. A pair of electrodes was placed on top of the diamond, and carriers were injected into the impurity region in the lateral direction of the diamond.The structure was a self-line structure in which the buffer layer and the impurity region coincided or roughly coincided, and a light shielding effect was achieved. The structure is such that an impurity region for emitting light is formed in close contact with a certain semiconductor layer.As a result, even when a pulse voltage of 40 to 100 V is applied, in addition to emitting visible light, High brightness can be achieved because the emitted light can pass through the anti-reflection film and be emitted to the outside without any obstacles.Furthermore, the electrode material does not diffuse into the impurity region that is the light emitting part. Even when the light was applied continuously for about one month, no decrease in the luminance was experimentally observed.

The present invention mainly shows the case where one light emitting element is manufactured.

However, it is possible to create a light-emitting device using multiple diamonds on the same substrate, form electrodes, and then scribe and break them to an appropriate size to make each diamond a single unit.
Alternatively, it is effective to use a light emitting device in which a large number of light emitting sources are integrated on the same substrate, for example, a light emitting device in a matrix array.

Furthermore, the method of the present invention uses only two types of photomasks, and can be expected to have an extremely high yield. For example, when manufacturing a 0.8 mm x 0.8 11111+ LHI on a 4-inch wafer, 10' LEDs could be manufactured on one side from the same wafer.

In the present invention, diamonds are mainly shown in the form of polycrystalline thin films. However, it goes without saying that if this diamond is a single crystal diamond, better physical properties such as higher brightness and luminous efficiency can be expected.

However, it has the disadvantage of being more expensive.

In the present invention, the substrate having an insulating surface is not limited to a substrate in which a silicon nitride film is formed on silicon, but may also be a substrate in which other insulators or silicon carbide are formed. Further, any material that does not allow current to flow downward, such as crystalline diamond, which has sufficient insulation properties can be used as a substrate for carrying out the present invention.

Including such light-emitting devices, diodes, transistors, resistors, and capacitors are integrally made using the same diamond and silicon semiconductors above or below the diamond, and are combined to form an integrated electronic device. is valid.

[Brief explanation of drawings]

FIG. 1 shows a manufacturing process and a longitudinal sectional view of the diamond electronic device of the present invention. FIG. 2 shows a longitudinal sectional view of another electronic device according to the invention. FIG. 3 shows an example of a magnetic field microwave apparatus for forming diamond on a substrate for use in the present invention. 1 ・ ・ ・ ・ ・ ・ 2 ・ ・ ・ ・ ・ ・ 3.3'. 3” .29-1,29 4 ・ ・ ・ ・ ・ ・ 5 ・ ・ ・ ・ ・ ・ 6 ・ ・ ・ ・ ・ ・ 7 ・ ・ ・ ・ ・ ・ 8, 28 ・ ・ ・ ・ ・ l1・・ ・ ・ ・ ・ ・ 9.29-2.29-2' ・ 13・ ・Substrate・Diamond l゛・・・Buffer layer・Photoresist・Impurity region・Antireflection film・Electrode・Bonded wire・Injected current path・Upper electrode・Holder・Attenuator・Magnet・Microwave oscillator 19・・ ・ ・ ・ 20・ ・ ・ ・ ・ 21, 22, 23.24 25・ ・ ・ ・ ・ 42・ ・ ・ ・ ・ ■,■・ ・ ・ ・Reaction chamber, end of impurity region, doping system, exhaust system・Movement mechanism ・Photo-etching process

Claims (1)

[Claims] 1. A step of selectively forming diamond on a substrate having an insulating surface and a semiconductor on the diamond, and adding an impurity to the diamond in the region from which the semiconductor has been removed using the semiconductor as a mask. 1. A method for manufacturing an electronic device using diamond, comprising the steps of: forming an impurity region using a diamond; and forming a pair of electrodes on the impurity region and on the semiconductor. 2. A method for manufacturing an electronic device using diamond according to claim 1, wherein the impurity is added by an ion implantation method of an element from group IIb or group VIb of the periodic table of elements. 3. A step of selectively forming diamond on a substrate having an insulating surface and a buffer layer on the diamond, and adding an impurity to the diamond in the region where the buffer layer has been removed using the buffer layer as a mask to form an impurity region. and forming an electrode on the buffer layer, and forming another electrode on the impurity region or another buffer layer on the impurity layer, and forming an electrode on the electrode. A method for manufacturing an electronic device using diamond, characterized by: