JPS6226194B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical thyristor having a turn-on configuration upon receiving light. This type of conventional optical thyristor includes one shown in FIG. 1, for example. FIG. 1 is a plan perspective view of a pellet 100 of a conventional planar optical thyristor, and FIG. 2 is a perspective view of the pellet 100 in plane A of FIG.
00 is a cross-sectional view. Such pellets selectively diffuse p-type impurities into an n-type silicon semiconductor substrate _nB to form p-type semiconductor layers _pB and _pE , and further selectively diffuse n-type impurities into the p-type semiconductor layer _pB . An n-type semiconductor layer n _E is formed, and the n-type semiconductor layer n _E
This can be obtained by providing a cathode electrode 1 on the surface, a gate electrode 2 on the surface of the p-type semiconductor layer _pB , and an anode electrode 3 on the p-type semiconductor layer _pE .
The semiconductor layer p _E operates as an emitter of the pnp transistor part of the thyristor, and n _B is
It acts as the base of the pnp transistor part and the collector of the npn transistor part, and p _B is npn
acts as the base of the transistor part and the collector of the pnp transistor part, _nE acts as the emitter of the npn transistor part, and thus pn junctions J ₁ , J ₂ and J ₃ respectively act as the emitter junction of the npn transistor part, npn, It operates as the collector junction of both PNP transistors and the emitter junction of the PNP transistor part. Also, 4 in the same figure is
This is an insulating film such as a silicon oxide film to protect the exposed surfaces of pn junctions J ₁ , J ₂ , and J ₃ . In this example, all of the pn junctions J ₁ , J ₂ , and J ₃ are exposed on one flat main surface of the semiconductor substrate, and the exposed surfaces are protected on this surface. The thyristor is called a planar optical thyristor. The thickness of each layer n _B , p _E , p _B , n _E is 50 to 70 μm, 30 to 50 μm, 20 to 25 μm, and 10 to 25 μm, respectively.
The optical thyristor is designed to perform a switching operation on the same principle as a general thyristor which is controlled by an electric signal (gate trigger) when a voltage is applied between the anode electrode 3 and the cathode electrode 1 of the optical thyristor. However, in the optical thyristor, electrons and holes are generated by the energy of the light that hits the pn junction J ₁ exposed on the surface and its vicinity, and the generated electrons and holes are connected between the gate electrode 2 and the cathode electrode 1. It turns on the gap between the anode and cathode by using a minority carrier that is injected into the body by electrical means, and the amount of electrons and holes generated is small and generally Like a thyristor, a gate current of any required magnitude is passed between the gate electrode 2 and the cathode electrode 1 to generate a large amount of minority carriers in the semiconductor layers p _B and n _E and turn them on. Therefore, the structure must be such that it can be easily turned on even by minute minority carriers caused by light energy. Therefore, in the _optical thyristor, the _p
Compared to the type impurity concentration, the n-type impurity concentration on the semiconductor layer _nE side near the p-n junction plane _J1 is made extremely large.
Injection efficiency γ of minority carriers at p-n junction surface J ₁
is increased, and the minority carriers injected from the semiconductor layer n _E to the semiconductor layer p _B easily reach the p-n junction surface.
Efforts have been made to make the width t of the semiconductor layer p _B as thin as possible so as to reach _J2 , that is, to increase the transport efficiency β of minority carriers. On the other hand, since the current amplification rate α of the transistor is approximately equal to the product of γ and β, that is, (α≒γ・β), the semiconductor layer n in the optical thyristor can be
α of the npn transistor portion formed by _E , _pB , and _nB becomes large, so that the optical thyristor is easier to turn on than a general thyristor. For an explanation of why the thyristor turns on more easily when the current amplification factor α is increased by increasing the injection efficiency γ and the transport efficiency β, see, for example, “Semiconductor
Controlled Rectifiers”Prentice―Hall Inc.1964
Since it is well known and has been published in 2015 and in many other documents, we will omit its explanation here. Now, in optical thyristors, as mentioned above, the PNPN four-layer structure has been devised to make it easier to turn on. The disadvantage is that the amount of light is poor. This drawback will be explained using a cross-sectional view of a conventional optical thyristor shown in FIG. An external voltage is applied between the anode electrode 3 and the cathode electrode 1 of the pellet 100 of the photothyristor so that the anode electrode 3 has a positive voltage and the cathode electrode 1 has a negative voltage, and the light 5 is connected to the p-n junction J ₁ It irradiates the exposed pellet surface, that is, the pellet surface where the cathode electrode 1 is present. Then, as the light passes through the very thin oxide film 4, electrons and holes are generated by the energy of the light at the pn junction _J1 just below the oxide film 4. Since the generated electrons have a negative charge, they enter the semiconductor layer _pB , are attracted by the externally applied voltage, pass through the semiconductor layer _pB , and pass through the pn junction _J2 , where most of the externally applied voltage is applied. Head to On the other hand, since the holes have a positive charge, they are attracted by the externally applied voltage and pass through the semiconductor layer n _E layer toward the cathode electrode 1 . That is, electrons and holes generated in the p-n junction _J1 by light cause a gate current to flow between the gate electrode 2 and the cathode electrode 1, and electrically conduct the semiconductor layer _pB.
The electrons injected into the semiconductor layer _n and the holes injected into the semiconductor layer act in the same way, turning on the thyristor. However, the problem here is that in conventional optical thyristors, the pn junction _J1 is connected to the pellet. This means that the amount of electrons and holes generated in response to light is small because there is little exposed under the oxide film 4 on the surface. That is, a small amount of electrons and holes generated is equivalent to a small gate current of the optical thyristor, which means that the thyristor is difficult to turn on. Therefore, in conventional optical thyristors, pn
Even if the injection efficiency γ and the transport efficiency β of electrons and holes of junction J ₁ are designed to be large, when receiving light energy,
Because the exposed part of the p-n junction _J1 , which effectively generates electrons and holes, on the pellet surface is too small,
The drawback was that it was difficult to turn on. In other words, conventional optical thyristors do not turn on unless they receive a large amount of light, resulting in a problem of poor light sensitivity. The present invention aims to solve the above-mentioned conventional problems and provide an optical thyristor with good photosensitivity by devising the structure of the pn junction that generates electrons and holes using light energy. It is something to do. The present invention will be described below by taking a planar optical thyristor as an example thereof. Figure 4 shows a planar optical thyristor pellet 20.
FIG. 5 is a perspective plan view of FIG. Such an optical thyristor of the present invention is
In the configuration of the semiconductor layer _nE acting as the emitter of the npn transistor part, the pn junction _J1 acting as the emitter junction of the npn transistor part, and the cathode electrode 1a, in the conventional optical thyristor shown in FIGS. Although they are different, that is, the semiconductor layer n _E , the pn junction J ₁ and the cathode electrode 1, there is no essential difference in the other configurations. That is, the pn junction _J1 , which operates as the emitter junction of the npn transistor part, has a deep part _J1D and a shallow part _J1S , and accordingly, the semiconductor layer _nE , which operates as the emitter junction of the npn transistor part, is different from the conventional one. It is formed from a portion n _ED that is approximately as thick as that of the optical thyristor and a portion n _ES that is much thinner than that. Furthermore, in the present invention, the thin semiconductor layer n
There is no cathode electrode 1 on _{the ES} , and a thick semiconductor layer n
The cathode electrode 1a is configured to exist only on the surface of _{the ED} . And also the width C of this cathode electrode 1a
is preferably made slightly smaller than the width D of the semiconductor layer n _ED and formed elongated together with the thick semiconductor layer n _ED , with the thin semiconductor layer n _ES disposed around its outer periphery. In addition, the cathode electrode 1a is formed in one or several rows together with the thick semiconductor layer _nED , and a bonding area 1b with a relatively wide area is provided so that it is easy to connect the cathode electrode 1a and bond an external conductor wire thereto. It is preferable to form the cathode electrode 1 as a whole. Furthermore, it is preferable that no oxide film or the like be present on the surface portion of the very thin semiconductor layer _nES and the portion exposed to the pellet surface of the pn junction _J1 . Also, when light hits the semiconductor surface, the light
It reaches up to about 5000 Å and is difficult to reach inside than that, so the thickness of the thin semiconductor layer _nES , that is, the depth from the semiconductor light-receiving surface to the pn junction _J1S , is 500 to 500 Å.
A suitable thickness is 5000 Å, and the suitable thickness of the n _ED is about 10 to 25 μm as in the conventional case. Furthermore, the ratio of the width E of the thin semiconductor layer n _ES to the width D of the thick semiconductor layer n _ED , that is, (E/D) is
It is best to decide within a range that satisfies 1/3 to 1. This is because (E/D>1), the cathode area becomes smaller and there is a risk that it will not be able to handle a large current.
Moreover, when (E/D<1/3), the thin semiconductor layer is small and the effect over the conventional one is reduced. In the following, a case will be described in which a power supply voltage V is applied through the load resistor R _L so that the anode electrode 3 of the semiconductor device obtained in this way becomes a positive voltage and the cathode electrode 1 becomes a negative voltage, and the semiconductor device is operated as a photothyristor. explain. First, a very thin semiconductor layer n _ES top surface (light-receiving surface)
When exposed to light, the light 5 hits the exposed surface of the semiconductor layer _nES and is attenuated into the thyristor pellet 200.
At this time, if the thickness of the very thin layer n _ES of the semiconductor layer n _E is set to about 500 to 5000 Å, the light 5 will pass through the thin layer n _ES and sufficiently enter the p-n junction J _1S. This is where many electron and hole pairs are generated. The electrons generated in this way are transferred to the semiconductor layer p.
Enter _B , pass through it and head to p-n junction J ₂ . On the other hand, since a negative voltage is applied to the cathode electrode 1a, the holes move laterally through the semiconductor layer _nES and reach the cathode electrode 1a through a part of the thick semiconductor layer _nED . In addition, since the light 5 passes through the thin semiconductor layer _nES and hits the pn junction _J1S surface almost perpendicularly, the area of the pn junction that receives the light 5 is Since _the plane of J ₁ is parallel to the light 5 , the p-n junction J The number of electron-hole pairs generated in the _1S section is significantly larger than in the conventional case, and the thyristor is therefore easier to turn on. In this way, the turn-on current based on the electrons and holes generated in the p-n junction _J1S flows in the direction of arrow 6 in _FIG .
Since the semiconductor layer is thin (approximately 5000 Å), many electrons and holes are generated, and based on this, a large turn-on _current tends to flow.
The fact that it is small and thin means that the resistance in the lateral direction, that is, the direction in which the current flows, is large, and as a result, a large potential difference is created by this resistance, and this potential difference causes the electrons and holes at the pn junction _J1D to Allow each injection to occur. Therefore, the arrow 6
After being turned on so that the current flows in the direction of arrow 9, it is subsequently turned on so that the current flows in the direction of arrow 9. Moreover, such a shift in the turn-on position becomes more pronounced depending on the structural and manufacturing characteristics of the optical thyristor. That is, according to the standard manufacturing method of optical thyristors, the semiconductor layer p _B is obtained by diffusing p-type impurities into the n-type substrate n _B , and each n-type semiconductor layer n _ED , n _ES is also diffused. formed by. The injection efficiency γ of the p-n junction obtained by such a diffusion technique is that of the thick n-type diffusion part n _ED γ _1D is that of the thin n-type diffusion part n
The shape diffusion part n _ES is larger than that of γ _1S , and the transport efficiency β between the pn junctions J _1D and J ₂ is larger than that of β _1D as described above.
It is larger than β _1S between p-n junction J _1S and J ₂ ,
Therefore, the current amplification factor (αnpn _D ) of the npn transistor portion formed along the current path 9 becomes larger than the current amplification factor (αnpn _S ) of the npn transistor portion formed along the current path 6, This is because it is easy to turn on. As is clear from the above explanation, according to the optical thyristor according to the present invention, turn-on occurs first along current path 6 shown in FIG. 6, and then turn-on occurs along current path 9. The on-current that mainly flows steadily is in the area indicated by the diagonal line 10 in the same figure, and as in a normal thyristor, it flows almost linearly from the anode electrode 3 to the cathode electrode 1, causing a particularly large voltage drop in the thyristor. Never. Therefore, the optical thyristor according to the present invention can flow a sufficiently large on-current, operate with a small amount of light, and can directly control switching of a large current. In order to further improve the photosensitivity, the area of the pn junction _J1S that receives the light 5 should be made as large as possible, and in order to flow as much on-current as possible after turning on, as shown in Figs. 4 and 5. In this case, the cathode electrode 1 is formed into an elongated shape, and a sufficient shallow semiconductor layer n _ES is provided between the elongated electrodes. As can be seen from the above explanation, it is also possible to make the width C of the cathode electrode 1a smaller than the width D of the deep part _nED of the semiconductor layer _nE , and to arrange it so that it exists within the range of the width D. This is important in improving sensitivity. This is because the cathode electrode 1a is a shallow layer n _ES
If it protrudes above the semiconductor layer n _ES , the area of the pn junction J _1S at the semiconductor layer n ES that receives light becomes smaller, and the potential difference between points 7 and 8 becomes smaller, and the turn of the thyristor at the pn junction J _1D becomes smaller. - This is because there is a drawback that it becomes difficult to turn on. In addition, in optical thyristors, there are almost no usages in which a gate current is passed from the gate electrode except in special cases, so the gate electrode 2 as shown in Figures 1 to 6 is used.
is not necessarily necessary. Furthermore, in the above embodiment, a planar type optical thyristor has been described, but as shown in FIG. 7, the present invention can also be applied to a mesa type optical thyristor (in the case of FIG. 7, there is no gate electrode). Furthermore, in the above embodiment, regarding the pnpn four-layer structure, especially the emitter of the npn transistor portion
Although the improvement of the semiconductor layer that operates as a base has been described, it goes without saying that the present invention can also be applied to a thyristor having a four-layer structure in which n-type and p-type are replaced.

[Brief explanation of the drawing]

FIG. 1 is a plan perspective view of a pellet of a conventional optical thyristor, FIG. 2 is a sectional view of a pellet of a conventional optical thyristor, FIG. 3 is a sectional view of a pellet for explaining the operation of a conventional optical thyristor, and FIG. Figure 4 is a perspective plan view of an optical thyristor that is an embodiment of the present invention;
5 is a sectional view of an optical thyristor which is an embodiment of the present invention, FIG. 6 is a sectional view of essential parts for explaining the operation of the optical thyristor which is an embodiment of the present invention, and FIG.
The figure is a sectional view of a mesa-shaped optical thyristor which is another embodiment of the present invention. 1, 1a, 1b... cathode electrode, 2... gate electrode,
3... Anode electrode, 4... Oxide film, 5... Light, 6... Direction of current flow, 7... One point on pn junction J _1S , 8... One point on cathode electrode 1a, 9... Direction of current flow, 1
0...Direction of current flow, 100 ...Pellet of conventional photothyristor, 200 ...Pellet of photothyristor of the present invention, _J1 , _J1D , _J1S ...pn junction, _J2 ...pn junction, _J3 ...pn junction, _nB , _nE , _nED , _nES ...n-type semiconductor layer, _pB , _pE ...p-type semiconductor layer.

Claims (1)

[Claims] 1 When irradiated with light, electrons and holes are generated and 2
In an optical thyristor having a pn junction that turns on between two metal main electrodes, a part of the pn junction surface is approximately parallel to the light-receiving surface of the pellet and is shallower than the surface by 5000 Å or less, and the pn The other part of the bond is substantially parallel to the light-receiving surface of the pellet, and is sufficiently deep by 5000 Å or more from the surface, and is deeper than the surface.
Without the presence of the metal main electrode on the surface directly above the p-n junction surface existing at a shallow depth of 5000 Å or less, on the surface directly above the p-n junction surface existing at a sufficiently deep location than the surface. An optical thyristor characterized in that the metal main electrode is formed in an elongated manner, and the shallow p-n junction surface is present below the light-receiving surface of the pellet around the outer periphery of the metal main electrode.