JPH0230591B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a semiconductor device using a static induction thyristor that can be turned on and off completely by light.

Specifically, conventional thyristors with a pnpn four-layer structure have the disadvantage that it is difficult to switch off using the gate electrode, and even if the gate can be used to turn off the switch, the speed is extremely slow. On the other hand, an electrostatic induction thyristor (hereinafter referred to as an SI thyristor), which has a diode structure with a gate, has the advantage that it is extremely easy to shut off using the gate, and that the shutoff time is quick. Figure 1 shows a cross-sectional view of a typical structural example of an SI thyristor.

FIGS. 1a to 1d are cross-sectional views of structural examples of SI thyristors in the circuit configuration of the present invention, which will be described later.

FIGS. 1a and 1b are cross-sectional views of typical examples of surface gate structures of SI thyristors. FIG. 1c shows an example of a buried gate structure, and FIG. 1d shows an example of a cross-sectional structure of an insulated gate type SI thyristor.

In FIGS. 1a, b, and c, p ⁺ regions 11 and 14 are anode regions and gate regions, n ⁺ region 13 is a cathode region, and n ⁻ region or i region 12 constitutes a channel. Typically the semiconductor material is silicon. 11', 13', 14' are Al, Mo, W,
The anode electrode, cathode electrode, and gate electrode are made of Au or other metals, low-resistance polysilicon, or a multilayer structure thereof.
Reference numeral 15 denotes an insulating layer of SiO ₂ , Si ₃ N ₄ , Al ₂ O ₃ , AlN, etc., or another insulating layer, or a composite insulating layer or multilayer insulating layer of these. The n region 16 has a relatively high impurity density, is formed in a thin layer, and is a layer for suppressing hole injection from the anode.

In FIG. 1d, p ⁺ region 21 is an anode region, i
Area 22 is an area that constitutes a channel,
The n ⁺ region 23 is a cathode region, and the n region 27 is a region for suppressing hole injection from the anode. The p region 28 has a structure in which it reaches the surface at a predetermined point in the vertical direction in the figure, and is usually directly connected to the cathode region by an electrode. 25 is
This is the insulating layer mentioned above. 21', 23', and 24 are the aforementioned anode electrode, cathode electrode, and insulated gate electrode. Regarding the operation of such SI thyristors, the dimensions of each region, and the impurity density, please refer to the patent publication No. 62-21275 "Electrostatic Induction Thyristor" written by the present inventors.
detailed in the specification.

SI thyristors control conduction and cutoff by controlling the potential distribution near the cathode with gate voltage, so they have the advantage of being able to easily cut off direct current and at high speed. Figure 1a
With the structure shown in FIG.
Compared to the structure shown in Figure 1a, the structure shown in FIG. However, it has the disadvantage that the reverse breakdown voltage is low. Therefore, in order to use the SI thyristors shown in Fig. 1b, c, and d where reverse breakdown voltage is required, a Schottky diode or the like must be connected in series.

Figures 2a and 2b show the symbol marks for junction type and insulated gate type SI thyristors. A diode is formed on the drain side of a static induction transistor (hereinafter referred to as SIT).

FIG. 3 shows a typical structural example of a conventionally well-known photosensitive semiconductor element. Figure 3 a to d
1 shows an example of the structure of a photosensitive element integrated or connected to the gate of an SI thyristor in the circuit configuration of the present invention, which will be described later. a is a photoconductive element. An i region 32 that can almost be regarded as an insulator is provided on the n ⁺ region 31 . 31', 3
2' is an ohmic electrode, and in this example, 3' is an ohmic electrode.
2' is a transparent electrode made of In ₂ O ₃ , SnO ₂ or the like.
When the electrode 32' is difficult to become an ohmic electrode, the transparent electrode 32' may be provided after a very thin n ⁺ region is provided on the surface of the i region 32. The light is irradiated,
When electron-hole pairs are generated in the i-region 32, current begins to flow. In this figure a, the electrode 32'
When 31' is used as a shot electrode, if a voltage is applied to the shot electrode side so that the potential is lower than that of 31', it also operates as a shot photo diode that is sensitive to light. L represents incident light. Figure b shows a phototransistor.
The n ⁺ region 44, the p region 43, the n ^- region 42, and the n ⁺ region 41 are an emitter region, a base region, a high resistance layer, and a collector region, respectively. 41' is a collector electrode, and 44' is an emitter transparent electrode. The n ⁺ region 44 and the p region 43 are made thin like a normal bipolar transistor. Most of the light is absorbed in the n ^- region 42. When a positive voltage is applied to 41', photoexcited electrons flow into the n ⁺ collector region and are absorbed, and holes flow into the p base region, which is a floating region, and are accumulated. When holes are accumulated in excess, the potential barrier to electrons in the base region decreases, allowing electrons to flow from the emitter region to the base region and into the collector region. Figure c shows a thyristor excited by light. n ⁺ region 55, p ⁺ region 51 is a cathode region,
This is the anode area. 55' is a transparent electrode, and 51' is an anode electrode. When light is irradiated with a positive voltage applied to 51', electrons and holes that are photoexcited in the i region flow into the n region 52 and the p region 54, respectively. The barrier potential decreases, and electrons are injected from the cathode region and holes are injected from the anode region, resulting in conduction. The optical thyristor in figure c has carrier injection amplification mechanisms on both sides. However, it is highly sensitive to light. Figure d is an example of a p ⁺ in ⁺ photodiode. 63' is a transparent electrode, and 61' is an electrode. 6
1' is applied to the positive voltage. In any case of a, b, c, or d, most of the irradiated light is absorbed in the i region or the n ^- region, generating electron and hole pairs. Therefore, if the electric field strength in this region is set to a level slightly lower than the avalanche electric field, a large amount of carrier will be generated by the avalanche multiplication mechanism, and the sensitivity will be further improved. Since FIG. 3 shows a very typical and very simple example, it goes without saying that there are many other variations. Of course, a photosensitive semiconductor element exhibiting a photovoltaic effect may be used.

So far, we have explained SI thyristors and photosensitive semiconductor devices. SI thyristors are extremely unique in that they have a large operating voltage and current, and can perform high-speed switching.
However, when considering high power applications such as DC power transmission, it is almost difficult for a single SI thyristor to handle all of this high power. Therefore,
For example, multiple SI thyristors with a forward blocking voltage of 5,000 volts or 10,000 volts and a current of 1,000 A when conducting are connected in series to increase the withstand voltage, and by further connecting them in parallel, the current can be increased. In order to obtain , a series-parallel connection of SI thyristors is required. When a large number of thyristors are connected in series and parallel, it is important that all thyristors turn on and off in synchronization. If even one thyristor is out of synchronization, the load will concentrate on that thyristor, causing it to burn out, and eventually causing damage to other thyristors. This causes the thyristor to burn out, leading to a major accident. In such cases, it is easier to synchronize the SI thyristor's conduction/cutoff by controlling it with light rather than with electrical signals. In addition, in conventional high-power SI thyristors, large currents also flow through the gate circuits, so the gate circuits at the actual implementation level are extremely large and complex, and the number of safety protection circuits has also increased, resulting in a large number of thyristors being connected in series and parallel. Connection becomes difficult. On the other hand, an SI thyristor controlled by light has already been described by the inventors of the present application in the specification of Japanese Patent Publication No. 1908/1988 entitled "Electrostatic Induction Optical Thyristor". However, in the invention described in Japanese Patent Publication No. 61-1908, the optical control method is related to turn-on operation, that is, optical trigger operation, and the turn-off operation is based on electrical turn-off using a gate circuit. That is, conduction is performed by light, but electrical gate turn-off is performed for interruption.

An object of the present invention is to provide an electrostatic induction thyristor whose conduction and interruption can be controlled by light and a device including the electrostatic induction thyristor.

Another object of the present invention is to provide a device including an optically controlled electrostatic induction thyristor capable of high-speed switching without excessive injection of fractional carriers from the gate at turn-off, and sufficient extraction of fractional carriers at turn-off. The goal is to provide the following. Yet another object of the present invention is to provide a device including electrostatic induction thyristors in which all the thyristors can be easily synchronized when the thyristors are connected in series and parallel for high power applications.

Still another object of the present invention is to provide a device including a static induction thyristor that simplifies the gate drive circuit, facilitates the separation of the high-power hot line and the gate drive signal, and allows multiple series-parallel connections of the thyristors. That's true.

The present invention will be described in detail below with reference to the drawings.

FIG. 4 is a circuit example of an SI thyristor whose operation is controlled by light according to the present invention, in which the aforementioned photosensitive semiconductor element is connected to the gate of the SI thyristor. In Figure 4, Q is an SI thyristor, D ₁ , D ₂
is a photosensitive semiconductor element. Vg is the bias power supply. There are various types of SI thyristors, so the value of Vg should be selected according to them. For example, if a forward voltage of 5000 volts can be blocked by applying a reverse gate bias of -30 volts to the gate, then Vg should be selected to a value of about -30 volts.

FIG. 4a shows an example in which a photosensitive element is connected in series and in parallel to the gate of an SI thyristor. Light irradiation has a complementary relationship. In other words, when D ₁ is irradiated with light, D ₂ is not irradiated with light,
When D ₂ is irradiated with light, D ₁ is not irradiated with light. When D ₁ is conducting,
The relationship is such that D ₂ is in a blocking state.
When D ₁ is conducting, the SI thyristor is cut off. When D ₂ is conducting, the SI thyristor is conducting.

When the reverse breakdown voltage of the SI thyristor is not very high, as in the examples shown in Figure 1 b, c, and d, and when reverse breakdown voltage is required during operation, the SI thyristor is Just connect a Tsutoki diode or the like in series with the SI thyristor.

For example, SI with a maximum forward blocking voltage of 5000 volts
Thyristors are used to convert 1 million volts of DC power into alternating current and
When configuring a DC converter, at least
200 SI thyristors must be connected in series. In such a case, as shown in FIG. 5, the required number of SI thyristors may be connected in series. Since the gates are controlled by light, it is easy to operate all SI thyristors simultaneously and synchronously, even if they are connected in series. Furthermore, since there is no large current wiring or resistance that constitutes the gate control circuit, the gate control circuit becomes simple and lightweight, making it easy to connect a large number of devices in series. For light irradiation, the light may be guided using an optical fiber or the like so that a predetermined location on the photosensitive element is uniformly irradiated. In any case, since the operation is not very fast, a clad fiber is sufficient. Further, in the case of the clad type, the light intensity within the cross section of the fiber is more likely to be uniform. Of course, depending on the application,
It is also effective to use converging fibers. FIG. 5b shows an example in which a resistor _Ro is connected in parallel to SI thyristors connected in series. This is because resistors of equal value are connected in parallel so that the voltage applied to each SI thyristor is uniform when the SI thyristor is in the cut-off state. The higher the resistance value, the better, as long as it can be considered to be smaller than the resistance value of the SI thyristor in the cut-off state. For example, a value such as 1MΩ is chosen. Of course, depending on the application, it may be larger or smaller than this.

Although the example of FIG. 5 shows an example in which the devices are connected in series in the example of FIG.

Furthermore, when handling a large current, the required number of devices configured as shown in FIG. 5 may be connected in parallel. Also in this case, since the gate drive circuit is simplified and the hot line and drive signal are separated, it is easy to connect a large number of thyristors in series and parallel. Although Figures 4 and 5 are all written using junction type SI thyristors, MOSSI thyristors may also be used.

Until now, it has been controlled by light by connecting a photosensitive semiconductor element to the gate of the SI thyristor.
An example of configuring an SI thyristor circuit is described. It goes without saying that the SI thyristor itself can be made into an element controlled by light.

In that case, in the operations shown in Figures 4 and 5,
Naturally, when the thyristor is brought into a conductive state, a means for introducing light into the thyristor is provided and an operation of direct light irradiation is included.

If the structure shown in Figure 1 a, b, c, and d is configured with transparent electrodes on either side so that light reaches the high resistance region, it will immediately become a light-controlled SI thyristor. . The side to which light is irradiated may be the cathode side or the anode side. It is sufficient to use a transparent electrode on either side. If the semiconductor material is Si, the incident depth of a given light is 20 to 30 μm
That's about it. Therefore, in the structure shown in FIG. 1, when light is irradiated from the anode side, the p ⁺ region 11
or the thickness of p ⁺ region 11 and n region 16 (p ⁺
It is desirable that the sum of the thicknesses of the region 21 and the n-region 27) be as thin as possible, and must be at least sufficiently thinner than the depth of incidence of light. For example, the p-region thickness should be about 1 μm or less, the impurity density should be about 1×10 ¹⁹ cm ^-3 or more, the n-region 16 thickness should be about 1 μm or less, and the impurity density should be about 1×10 ¹⁶ cm ^-3 or more. . The depth of the gate region is usually about several μm, although it depends on the gate spacing and the impurity density in that region. Therefore, if the electrodes are made transparent,
It can be controlled with light. If only transparent electrodes such as In ₂ O ₃ or SnO ₂ result in high resistance, metal electrodes such as Al may be placed at strategic points. This will be explained using the example shown in FIG. 1b. I area 12 by light irradiation
electron-hole pairs are generated. If a positive voltage is applied to the anode, the generated electrons flow into the n-region 16 and charge the n-region 16 negatively. When the amount of charge increases to a certain extent, hole injection begins to occur from the anode region. Photoexcited holes and holes injected from the anode flow toward the cathode. Some flows into the p ⁺ gate region and some flows into the cathode region. At this time, the barrier to electrons that had been formed on the entire surface of the cathode is lowered. Electron injection begins to occur from the cathode and it becomes conductive. In order to further reduce the forward voltage drop during conduction, a resistor of a predetermined size may be inserted between the gate and source. The magnitude Rgi of the insertion resistor may be set such that RgiIg is approximately equal to the reverse bias applied to the gate, where the current Ig flowing through the gate is taken as RgiIg. This gate-source resistance can be easily made of polysilicon. Figures a, c, and d may be similarly designed. In FIG. 1d, a resistor is inserted between p region 28 and cathode region 23. In FIG.

FIGS. 6 to 8 are similar to FIGS. 1 a to d,
1 shows an example of the structure of an SI thyristor according to the present invention.

Compared to the structure shown in FIG. 1, FIG. 6 shows a structure in which the current flowing through the gate is reduced and the main current is increased. The structure is such that an insulating layer 17 made of SiO ₂ or the like is provided on the lower surface of the p ⁺ gate region 14 . Gate·
The structure is such that a polysilicon resistor is inserted between the cathodes. Polysilicon resistors are provided at predetermined locations in the vertical direction in the figure. The equivalent circuit is as shown in FIG. 6b. Of course, the gate-cathode resistance Rgi can also be connected externally.

FIG. 7 shows an example in which, instead of connecting a resistor between the gate and the cathode, the potential barrier on the cathode side is also eliminated by accumulation of carriers. For example, a negative voltage is applied to the gate electrode 14, but this voltage is completely masked by the holes flowing into the vicinity of the gate, the potential barrier across the entire surface of the cathode disappears, and electron injection from the cathode occurs violently. . The mechanism of hole injection from the anode is the same as in the previous examples. 15 is an insulating area.

In FIGS. 6 and 7, the electrodes 13' and 14' are made of transparent electrodes such as low-resistance polysilicon, In ₂ O ₃ , and SnO ₂ so that light can enter from the cathode side. This is done to allow heat sinking from the anode side. If you don't need to worry too much about the heat sink, use a transparent anode electrode.
Light may be irradiated from the anode side. Of course, it goes without saying that light may be irradiated from both sides.

Although FIGS. 6 and 7 show a structure in which a gate is provided only on the cathode side, a gate structure may also be provided on the cathode side. If the electrons accumulated in the n-region 16 remain as they are, they will have to go through annihilation due to recombination or the like. This problem can be eliminated by introducing a gate structure on the anode side as well. An example is shown in FIG. Also on the anode side, p ⁺ anode area 11
An n + gate region 18 is provided adjacent to the n ⁺ gate region 18 . An insulating region 19 is provided. If a reverse bias is applied to the anode region 11 and n ⁺ region 18 on the anode side as well as the cathode side, region 1
2 may be an i region with very high resistance instead of n ^- . In the structure of Figure 8, p ⁺ region 11 and n ⁺ region 1
It is sufficient to build a polysilicon resistor between 8 and 8. The impurity density and thickness of regions 12 and 16 are determined according to the application specification of JP-A No. 55-99774 ``Electrostatic Induction Thyristor'' and JP-A No. 55-108768 ``Electrostatic Induction Thyristor''.
The selection may be made as described in the application specification. The gist of the present invention is to integrate or connect a photosensitive element to the gate of an SI thyristor so that it can be quenched (turned off by light) with light, and to transfer carriers accumulated at the gate of the thyristor when the thyristor is on to the thyristor. The thyristor is turned off by irradiating light onto the photosensitive element connected to the gate of the thyristor, making it conductive and discharging it. That is,
Embodiment The circuit configuration shown in FIGS. 4a and 4b and FIGS. 5a and 5b is the main part of the present invention. In the end, the important part of the present invention is that it is turned off by light, and moreover, the gate is turned off by light, that is, it is completely controlled by light. The fact that the hot line can be completely controlled by light is extremely important in terms of safety measures and equipment configuration.

It goes without saying that the configuration of the light-controlled SI thyristor of the present invention is not limited to the example described here. Of course, a light-controlled SI thyristor may be used as the light-sensitive semiconductor device shown in FIGS. 4 and 5. Moreover, it goes without saying that the conductivity type may be reversed. All you have to do is reverse the polarity of the voltage. In the case of a junction type, it is preferable to sandwich an insulating layer such as SiO ₂ between the cathode and gate because this reduces capacitance and increases breakdown voltage at the same time. It is also effective to provide a gate in the cut region even in the case of a junction gate type. Even in a structure in which the gate is provided on the anode side, it is often effective to sandwich an insulating layer. In the case of a junction gate, a short gate may be used instead of a pn junction.

The light source for controlling the SI thyristor may be any light source that can generate electron-hole pairs in the semiconductor and can switch at high speed. It may be a semiconductor laser or a light emitting diode. For example, a GaAlAs-based heterojunction laser or a light emitting diode may be used. Of course, a semiconductor laser or a light emitting diode made of other materials may also be used. Of course, other lasers may also be used. In short, any light source that can provide the necessary intensity and that can be modulated at high speed is sufficient. Depending on the type of laser, a solid or gas laser may be used. These lights may be directly irradiated, or they may be guided to a predetermined location using a fiber.

The optically controlled SI thyristor of the present invention can be manufactured using conventionally known lithography technology, diffusion/ion implantation technology, crystal technology, epitaxial growth technology, etching technology, oxidation technology,
It can be easily manufactured using CVD technology, wiring technology, etc.

According to the present invention, the SI thyristors capable of high-speed switching of large voltages and large currents can be controlled by light, so when multiple SI thyristors are connected in series or parallel, or in series and parallel, all SI thyristors can be controlled by light. It is an extremely effective technology that can easily synchronize the signals, and its industrial value is high in constructing high-power AC/DC converters such as DC power transmission.

According to the present invention, since the gate is not forward biased during turn-off, excessive minority carriers are not injected, and at turn-off, minority carriers are efficiently sucked out from the gate, making extremely high-speed operation possible. Become.

According to the present invention, extremely high-speed switching is possible because no extra resistance is connected to the gate drive circuit.

According to the present invention, the gate drive circuit is simple and lightweight, and since the hot line and the drive signal line are separated by light, it is easy to connect a large number of thyristors in series and parallel.

[Brief explanation of drawings]

Figures 1a to d are examples of cross-sectional structures of electrostatic induction thyristors, Figures 2a and b are symbol marks of electrostatic induction thyristors, and Figure 3 is a diagram of various optically controlled semiconductor elements.
4 is a photoconductive element, b is a phototransistor, c is a photothyristor, d is a photodiode, FIG. Examples of series connection of SI thyristor circuits including photosensitive semiconductor elements, Figures 6a and 7
Figure 8 is a diagram showing an example of the structure of the electrostatic induction thyristor part in the circuit configuration which is the main part of the present invention.
FIG. b is an explanatory diagram of a circuit that lowers the forward voltage drop.

Claims (1)

[Claims] 1. A first gate circuit consisting only of a first photosensitive semiconductor element _D1 and a reverse gate bias power supply connected in series with the first photosensitive semiconductor element D1 is connected to the gate of the electrostatic induction thyristor.
In a configuration in which a second gate circuit consisting of only a second photosensitive semiconductor element _D2 is connected in parallel between the cathodes and connected to the gate and cathode of the electrostatic induction thyristor, the first and second gate circuits are connected in parallel between the cathodes. the first photosensitive semiconductor element, comprising first and second means for respectively irradiating light to the photosensitive semiconductor element;
When D ₁ is irradiated with a light pulse L ₁ and D ₁ is in a conductive state, the second photosensitive semiconductor element D ₂ is not irradiated with light and D ₂ is in a blocked state, and no light is applied to D ₂ . When pulse L ₂ is irradiated and D ₂ is in a conductive state, no light is irradiated to D ₁ and D ₁ is in a blocked state,
The relationship between the irradiation of L ₁ and the irradiation of L ₂ and the
The conduction state of D ₁ and the conduction state of D ₂ are complementary to each other, and when D ₁ is in a conduction state, the electrostatic induction thyristor is cut off, and when D ₂ is in a conduction state, the electrostatic induction thyristor is cut off. 1. A semiconductor device including an electrostatic induction thyristor, wherein the electrostatic induction thyristor is in a conductive state when the electrostatic induction thyristor is in a conductive state. 2. A semiconductor device including a static induction thyristor according to claim 1, characterized in that a Schottky diode or a pin diode having a predetermined reverse breakdown voltage is connected in series with the static induction thyristor. 3. A semiconductor device including a capacitive induction thyristor, characterized in that a plurality of semiconductor devices including a capacitive induction thyristor according to any one of the above items 1 to 2 are connected in series. 4. In order to equalize the voltage applied to each electrostatic induction thyristor when all of the plurality of electrostatic induction thyristors are in a cut-off state, resistors R _o of equal value are connected in parallel between the anode and cathode of each electrostatic induction thyristor. A semiconductor device including the electrostatic induction thyristor according to claim 3, which is connected to the electrostatic induction thyristor. 5. A semiconductor device including an electrostatic induction thyristor, characterized in that a semiconductor device including the electrostatic induction thyristor according to claim 4 is further connected in parallel. 6. The electrostatic induction thyristor according to any one of claims 1 to 5, wherein light irradiation to the photosensitive element is guided by an optical fiber. Semiconductor devices including. 7. A semiconductor device including an electrostatic induction thyristor according to any one of claims 1 to 6, characterized in that a photo-controlled electrostatic induction thyristor is used as the photosensitive semiconductor element. 8. A semiconductor device including a static induction thyristor according to any one of claims 1 to 7, wherein the static induction thyristor has a surface junction gate structure. 9. A semiconductor device including a static induction thyristor according to any one of claims 1 to 7, wherein the static induction thyristor has a buried gate structure. 10. A semiconductor device including the electrostatic induction thyristor according to any one of claims 1 to 7, wherein the electrostatic induction thyristor has an insulated gate structure. 11. A semiconductor device including the electrostatic induction thyristor according to any one of claims 1 to 7, wherein the electrostatic induction thyristor is a photo-controlled electrostatic induction thyristor. 12. The above-mentioned claim characterized in that the electrostatic induction thyristor is an optically controlled electrostatic induction thyristor, and includes means for irradiating light, and includes an operation of transitioning to a conductive state by directly introducing a light pulse. A semiconductor device comprising the electrostatic induction thyristor according to any one of items 1 to 7.