JPH059948B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a semiconductor device formed by combining a single crystal semiconductor and a non-single crystal semiconductor, and particularly to a photoelectric conversion device having a phototransistor function.

``Prior Art'' Conventionally known photoelectric conversion devices having a phototransistor function have been made only of single crystal semiconductors. A typical vertical cross section is shown in Figure 1.

As is clear from the figure, a base 2 and an emitter 3 are formed by diffusion on a single crystal semiconductor substrate 1 by planar technology, and an opening is formed in a silicon oxide insulating film 4, and an electrode lead 5 of the emitter and an electrode of the base are formed. A lead 6 is provided, and most of the substrate is used as a collector 7, and an electrode 8 is provided thereon.

In such a configuration, the light 10 is transmitted to the emitter 3
The light is guided through the upper silicon oxide insulating film 4 to the depletion layer region between the base and emitter of the single crystal semiconductor, and photocarriers are generated in this portion.

``Problems to be Solved by the Invention'' In such a structure, the emitter 3 has a high concentration of N ⁺ conductivity type with respect to the incident light 10, and
Eg (optical energy band width is hereinafter referred to as Eg)
is, for example, 1.1 eV when silicon is used as the semiconductor, so there is a large amount of light absorption loss in this emitter region.

Furthermore, when a silicon semiconductor is used, the width in the thickness direction of the active region (depletion layer region between the emitter and base) due to the light irradiation 10 becomes an N ⁺ type emitter with a high concentration (impurity concentration 10 ^{19 to} 10 ²¹ cm ^{- 3} ), - Due to the P-type base (impurity concentration 10 ¹⁸ cm ^-3 ), it is formed only in an extremely thin thickness.

However, since the conventional photoelectric conversion device having a phototransistor function uses a single crystal semiconductor, it is an indirect transition type, and the absorption coefficient for visible light is also very small. That is, although the depletion layer region needs to be thick in order to generate a large number of photo carriers, it is only formed as thin as ~0.1 μm for the reasons mentioned above. For this reason, there was a problem that the probability of occurrence of photocarriers was extremely small. Furthermore, when a silicon semiconductor is used, its Eg is 1.1 eV, which deviates from the human visual sensitivity characteristic, resulting in high photosensitivity to long wavelength light.

Because of these drawbacks, their use in mechatronic optical sensors and the like has not been satisfactory, and there has been a demand for photoelectric conversion devices with phototransistor functions with even better photosensitivity as light-receiving elements for optical communications.

"Means to solve problems" The present invention compensates for these drawbacks.

In the present invention, a single crystal semiconductor, such as a silicon semiconductor, has an Eg of 1.1 eV, and although it is possible to detect infrared light, the visibility of visible light is not sufficient. Furthermore, even in the case of infrared light, photocarriers are not sufficiently generated because the light absorption coefficient of a single crystal semiconductor is small.

For this reason, we have further improved the conventionally known phototransistor using a single-crystal semiconductor, and created a depletion layer region between the emitter of the photosensitive region or the emitter base (hereinafter, these are collectively referred to as emitter regions). ), it is characterized by the use of a non-single crystal semiconductor with a large light absorption coefficient.

That is, a collector region of one conductivity type provided with a single crystal semiconductor, a base region of a single crystal semiconductor of a conductivity type opposite to the collector region on this collector region,
This photoelectric conversion device is provided with an emitter region made of a non-single crystal semiconductor that generates photovoltaic force upon irradiation with light doped with hydrogen or a halogen element on the base region.

As a non-single crystal semiconductor in such an emitter region,
Semiconductors for generating photocarriers (hereinafter simply referred to as I-type semiconductors) with a large light absorption coefficient that are intrinsic or substantially intrinsic (P- or N-type impurities are added at a low concentration of 10 ¹⁵ to ^{10 17} cm ^-3) . layer or I layer)
and a P- or N-type semiconductor layer having an emitter electrode function is provided on this semiconductor layer.

Furthermore, this P or N type semiconductor layer is made of a semiconductor having a wider Eg than the I type, such as Si _X C _1-X
(0<x<1), or use a micro-polycrystalline semiconductor with a crystal grain size of 50 to 200Å with low light absorption loss or a polycrystalline semiconductor with a fiber structure having uniaxial single crystallinity. I can,
Therefore, the structure allows light incident from the emitter side to effectively reach the I layer and generate a large number of photo carriers.

Further, a plurality of base regions made of a single crystal semiconductor are provided in the collector region, and a plurality of emitter regions made of a non-single crystal semiconductor are provided on the base region, and each emitter region has a specific wavelength range on the light irradiation surface side. By providing a filter that transmits this light, it is possible to create a sensor that is sensitive to color. Furthermore, by providing this filter to correspond to light elements such as red, green, and blue,
It is possible to create a sensor with the same level of color sensitivity as humans.

In particular, since an amorphous or semi-amorphous semiconductor is used in the I layer as the photoactive semiconductor layer,
Its Eg is 1.6 to 1.8 eV for silicon,
Its light absorption coefficient is also approximately 10 times larger than that of single crystal silicon. Therefore, it has extremely high conversion efficiency for visible light, and its photosensitivity characteristics are the same as human visibility, so it can act as a ``human eye with a light amplification function.''

Further, a semiconductor having an amorphous structure or a semi-amorphous structure can be used for the I layer as a photoactive semiconductor layer, and an arbitrary wavelength band can be selected by changing the material used. For example, in the case of germanium, its Eg is in the range of 1 to 1.2 eV, and
For _{Si x Ge 1 -x} (0 < wavelength range can be selected. Therefore, it is extremely convenient as a light receiving sensor for optical communications using optical fibers.

Further, in the phototransistor structure, a forward bias is applied between the emitter and the base.
Therefore, in the case of a non-single crystal semiconductor such as an amorphous semiconductor, the reverse breakdown voltage is about 5 V, or only 0.5 to 2 V is applied in the forward direction, so there is no problem in terms of reliability. On the other hand, ~200V, typically 10-20V, is applied between the base and collector. However, this provides a PN junction or a PIN junction,
In addition, by removing lattice defects in single crystal semiconductors, it is possible to withstand voltages of 20 to 200 V in practical use without any problems.

For this reason, a non-single crystal semiconductor doped with hydrogen or a halogen element is used for the depletion layer region to which forward bias is applied, and a non-single crystal semiconductor doped with hydrogen or halogen is used for the region to which reverse bias is applied. The leakage current can be measured to be reduced to 10 ^-9 to 10 ^-11 A (when 10 V is applied).

This means that if a non-single-crystal semiconductor is used in the depletion layer region between the base and collector, a current of 10 ^-4 to 10 ^-6 A (when 5V is applied) can be obtained. By combining a single-crystal semiconductor with a non-single-crystal semiconductor as in the invention, the advantages of each semiconductor can be brought out and the shortcomings of each can be compensated for, and the structural effect is remarkable.

Examples are shown below according to the drawings.

Example 1 FIG. 2 is a vertical sectional view showing the manufacturing process of the photoelectric conversion device of the present invention.

In FIG. 2A, a silicon semiconductor having a (100) plane was used as a single crystal semiconductor substrate. This board 1
is made up of an N ⁺ type semiconductor layer 7 and an N type (including I type or N ⁻ type) semiconductor layer 17 . Further, this surface was thermally oxidized in oxygen at 1100°C to form a silicon oxide film 4 with a thickness of 0.1 to 0.5 μm, and using this as a mask, P-type impurities were selectively doped to a thickness of 200 Å to 1 μm. .

This dope may be formed, for example, by a thermal diffusion method using BSG (boron glass) or by an ion implantation method.

Thus, the single crystal semiconductor 1 shown in FIG. 2A comprises an N ^- N substrate having a collector function, and a base 2 made of a single crystal semiconductor on top of the N - N substrate. Further, on the base 2 of such a single crystal semiconductor, an intrinsic or
A substantially intrinsic non-single crystal semiconductor layer doped with a valent impurity such as boron at a concentration of 10 ¹⁵ to 10 ¹⁷ cm ^-3 was formed to a thickness of 500 to 5000 Å by plasma CVD.

When producing a semiconductor whose main component is amorphous or semi-amorphous silicon to which hydrogen or a halogen element such as fluorine or chlorine is added as this non-single crystal semiconductor, silane, dichlorosilane or silicon fluoride gas is used at 13.56 By applying high frequency power with a frequency of MHz and using the glow discharge method,
Formed at a temperature of ~350°C.

This semi-amorphous semiconductor is disclosed in a patent application filed by the present inventor (Semi-Amorphous Semiconductor Patent Application 1988-26388).

Furthermore, an N-type non-single crystal semiconductor layer 12 having a fiber structure was formed on the upper surface to a thickness of 100 to 300 Å.

This semiconductor with a fiber structure has single crystallinity in the uniaxial direction, and even at low temperatures of 200 to 250 degrees Celsius,
Can be made with low high frequency power of ~10W. This semiconductor having a fiber structure is disclosed in a patent application filed by the present inventor (Semiconductor having a fiber structure and its manufacturing method Patent Application No. 57-87801 S57.5.24). Furthermore, the inventors of the present invention
We are planning to make a presentation on semiconductors with a fiber structure at the Japan Society for Applied Physics conference, so please refer to the presentation and the proceedings.

A semiconductor with this fiber structure has an Eg of 1.5
~1.8 eV, its optical absorption coefficient is as small as that of a single crystal semiconductor, and its electrical conductivity is 10
Since it has an extremely large value of ~300 (Ωcm) ^-1 ,
This is convenient for constructing the emitter of the phototransistor of the present invention.

Furthermore, if the N layer 12 is formed of an amorphous silicon semiconductor, its absorption coefficient is large, so light does not reach the depletion layer region where photocarriers occur sufficiently, which is not preferable. For this reason, when using silicon alone as a semiconductor, micropolycrystals (crystal grain sizes of 100 to 300
Å) It is effective to use a semiconductor having a fiber structure.

In addition, when an amorphous semiconductor is used as the N layer 12, a W (N type semiconductor layer) with a larger energy band width than that of the I layer 11 is produced by using Si _x C _1-x (0<x<1) or the like. -N (I-type semiconductor layer)
With the (WIDE Eg-NARROW Eg) structure, it was possible to reduce the decrease in transmitted light due to light absorption in this N-type semiconductor layer 12 and increase the amount of light reaching the depletion layer region.

In FIG. 2B, a transparent conductive film 1 is further formed on this upper surface.
3 For example, ITO (indium oxide to which 0 to 10% tin oxide is added) is formed to a thickness of 600 to 750 Å by vapor deposition using an electron beam. After this, using a photomask 14, ITO 13, N layer 12, I layer 1
1 is selectively etched to form an emitter region.

Furthermore, as shown in FIG. 2C, a photoresist is coated on this semiconductor, holes are selectively made, aluminum is vacuum-deposited in the holes, and aluminum is then lifted off to form a transparent conductive film electrode that forms an emitter. An emitter lead 5 was made from No. 13, and a base lead 6 was made from base 2 at the same time.

In this example, the aluminum electrode was fabricated by the lift-off method in order to prevent aluminum from associating and impregnating ITO during vacuum evaporation, thereby preventing the light transmittance of the transparent conductive film from decreasing.

Finally, as shown in FIG. 2C, the silicon oxide film 1
5 was overcoated to a thickness of about 0.5 μm. This silicon oxide film was also formed by a plasma vapor phase method or a vacuum evaporation method. The non-single crystal semiconductor layer of the substrate was formed at a temperature of 300° C. or lower to prevent a decrease in reliability due to release of recombination center neutralizing elements such as hydrogen in the semiconductor to the outside.

After this, the extraction electrode 8 for the collector region was formed of aluminum on the back surface of the semiconductor substrate 1, and then the external extraction electrode 16 was bonded to form a photoelectric conversion device having a phototransistor function.

FIG. 3 shows an energy band diagram corresponding to the vertical section taken along line A-A' in FIG. 2C.

In this figure, a base region 2 and collector regions 17 and 7 are made of a single crystal semiconductor 1, and an N-type semiconductor 12 and an I-type active layer 11 in an emitter region are both made of a non-single crystal semiconductor 3. Light 10 is irradiated from the transparent conductive film 13 side, and electrons 29 and holes 31 are generated.

At this time, a forward bias is applied between the base region 2 and the emitter region 12, and a reverse bias of 5 to 50 V is applied between the base region 2 and the collector regions 17 and 7. By doing so, h _fe =
We were able to create a photoelectric conversion device with over 5000 phototransistor functions.

Of course, when not irradiated with light, it can act as a simple transistor, and even in such a case, the emitter region 1
2,11 (Eg is 1.7 to 2eV) has a large Eg, so holes flowing in the opposite direction from the base region to the emitter region are blocked, making it much more effective than a planar transistor made only of single crystal semiconductors. We were able to achieve an excellent amplification effect.

In particular, in this example, since amorphous or semi-amorphous silicon was used for the active layer, the visibility was the same as that of the human eye, making it possible to create a so-called unprecedented phototransistor for visible light.

Example 2 In this example, non-single crystal germanium was used to form the active semiconductor layer 11 in the manufacturing process shown in FIG. As a result, its energy band width is 1 eV, and since it is a non-single crystal semiconductor, its light absorption coefficient is also large. For this reason, it was an excellent sensor for infrared light.

That is, after forming the base region 2 in the semiconductor substrate as shown in FIG. 2A, the base region 2 is formed by plasma vapor phase method.
Germanium gas is introduced at a temperature of 200 to 300°C, and an active germanium layer 11 is laminated on the base region 2 to a thickness of 500 to 5000 Å. Further, as the N-type semiconductor 12 in the emitter region, a silicon non-single crystal semiconductor having an N ⁺ fiber structure was used. By doing this, it has a W-N structure of N-type semiconductor layer (1.7 eV) - active layer (1.0 eV) and the light absorption coefficient of the N ⁺ layer is small, so much infrared light is transmitted to the active germanium layer 1.
We were able to supply 1.

The fabrication process other than this step was based on Example 1.

Thus, leakage current in the reverse direction is reduced for infrared detection.
A phototransistor with h _fe = ~10 ⁴ of 10 ^-9 A or less can be made, and in the case of this example, especially NPN
(including the NIPIN type), the carrier is electronic, and its frequency response speed is fast, making it possible to use it as a light receiving sensor for optical communications.

Embodiment 3 FIG. 4A shows a vertical sectional view of another photoelectric conversion device having a phototransistor function according to the present invention.

In the figure, since the light receiving part has a large area (1 cm ² or more), auxiliary electrodes 18 are placed on the transparent conductive film 13 on the N-type semiconductor layer 12 in the emitter region with a width of 10 to 30 μm and an interval of 200 to 1000 μm. They are set up. Further, a bias voltage is applied between the emitter region 3 and the collector region 7 from the electrodes 16 and 8, and as a result, the N-type semiconductor layer 12 in the emitter region and the active layer 1
1 - base region 2 in the forward direction, and P type base region 2 - I or N ^- type semiconductor layer 17 - N ⁺ type collector region 7 is biased in the reverse direction.
No base electrode is provided. Even with this, a sufficient phototransistor effect could be confirmed.

In this embodiment, the active layer 11 is formed over the entire semiconductor layer. Also, this active layer 11
visible light as non-single-crystal silicon, and non-single-crystal germanium, Si _x Sn _1-x (0<x<1), Si _x
It is preferable to use Pb _1-x (0<x<1) for detection in the infrared or visible-infrared region.

Example 4 FIG. 4B shows another photoelectric conversion device of the present invention.

In this drawing, an N-type collector region 7,
The N ⁻ layer 17 and the P type base region 2 are made of a single crystal silicon semiconductor 1, and the emitter region is provided with an active layer 11 made of a non-single crystal semiconductor 3. In this embodiment, unlike the previous embodiments, the manufacture of the N layer above this is omitted. In the drawing, an aluminum comb-shaped electrode 18 is manufactured by a lift-off method, and an anti-reflection film is formed on its upper surface by depositing SiO or TiO ₂ by an electron beam evaporation method.

Furthermore, in addition to this configuration, a distance of 2 to 2 Å can be used to flow a tunnel current between the active layer 11 and the electrode 18.
It is also possible to fabricate a silicon nitride or silicon carbide film with a thickness of .

The rest is the same as in Examples 1 and 2.

Example 5 FIG. 4C shows another embodiment of the invention.

This embodiment is constructed by configuring a plurality of base regions within the collector region.

In the drawing, three P-type base regions 2, 2', 2'' of a single crystal semiconductor are collector regions 7, 17.
is placed above. An active layer 11, an emitter region 12, and a transparent conductive film 13 are provided as a non-single crystal semiconductor 3 on a single crystal semiconductor substrate 1, and an external lead electrode 5 is made of aluminum. The three photoelectric conversion device parts 25, 26, 27 share a collector. On the incident light 10 side, red,
Three types of color filters 20, 21, and 22, such as blue, yellow, or red, green, and yellow, are provided, and the device has an integrated structure in which the sensitivity of each color can be selected and detected.

A photoelectric conversion device with such a structure has made it possible to distinguish colors in the same way as the human eye, and it has become possible to amplify and detect the signals.

Of course, it goes without saying that a so-called integrated structure in which the collectors are not shared and each is independent may also be used.

Furthermore, by using a semiconductor with a wider Eg than the base and collector on the emitter side,
Reduction of 1/f noise as a transistor, and
It was possible to simultaneously increase h and _fe .

In the examples of the present invention, NPN, NIPN or NIPIN type was used. This is because the carrier mobility of holes in non-single crystal semiconductors is only a few tenths of that of electrons. However, it goes without saying that a PNP type (including PINIP type) using this hole may also be produced.

``Effects of the Invention'' As is clear from the above explanation, the present invention combines the characteristics of single crystal semiconductors and the characteristics of non-single crystal semiconductors, and has unprecedented high breakdown voltage and micro leakage characteristics. We were able to create a photoelectric conversion device with high light sensitivity that is visible to the human eye.

Furthermore, by using a semiconductor with a wider Eg than the base and collector on the emitter side,
Reduction of 1/f noise as a transistor, and
It was possible to simultaneously increase h and _fe .

[Brief explanation of the drawing]

FIG. 1 shows a vertical sectional view of a conventional phototransistor. FIG. 2 shows the manufacturing process of the photoelectric conversion device of the present invention. FIG. 3 is an energy band diagram corresponding to FIG. 2C. FIG. 4 shows a vertical sectional view of another photoelectric conversion device of the present invention.

Claims (1)

[Scope of Claims] 1. A single crystal semiconductor substrate having a semiconductor region of one conductivity type serving as a collector region, a base region of a single crystal semiconductor of a conductivity type opposite to that of the collector region provided on the substrate, and a base region of a single crystal semiconductor of a conductivity type opposite to that of the collector region. 1. A photoelectric conversion device comprising: an emitter region made of a non-single crystal semiconductor doped with hydrogen or a halogen element on at least a portion of the region and generating photovoltaic force upon irradiation with light. 2. In claim 1, the emitter region is provided by laminating an intrinsic or substantially intrinsic semiconductor layer and a semiconductor layer of a conductivity type opposite to that of the base region on the semiconductor layer. A photoelectric conversion device characterized by: 3. A single crystal semiconductor substrate having a semiconductor region of one conductivity type serving as a collector region, a base region of a single crystal semiconductor of a conductivity type opposite to the plurality of collector regions provided on the substrate, and at least one semiconductor region on the base region. A plurality of emitter regions made of a non-single crystal semiconductor doped with hydrogen or a halogen element and which generates photovoltaic force upon irradiation with light, and light in a specific wavelength range on the light irradiation surface side of each of the emitter regions. A photoelectric conversion device characterized by being provided with a filter that transmits light. 4. In claim 3, the emitter region is provided by laminating an intrinsic or substantially intrinsic semiconductor layer and a semiconductor layer of a conductivity type opposite to that of the base region on the semiconductor layer. A photoelectric conversion device characterized by: