JPS595669A - Semiconductor device - Google Patents

Abstract

PURPOSE:To provide a semiconductor device which has excellent mass productivity and a minute electrode structure and wherein metallic electrode foils for a cathode and a gate are not contacted in assembling, by providing two relay metallic electrodes at a part of the main surface of an insulating material, and connecting each relay metallic electrode and a specified part of a silicon element, which is formed into a minute pattern with two comb-shaped electrode metal foils, by using a solder. CONSTITUTION:A relay metallic electrode 31c for a gate and a relay metallic electrode 31e for a cathode are bonded to an annular ceramic insulating material 31a by silver solders 31b and 31d. A composition of metal and an insulating material 31 is formed in this way. A silicon element 1 is contacted with facing protruded parts 31f and 31g of the inner wall of the opening of the composition of the metal and insulating material 31 and the position of the element 1 is fixed. Assembling is performed as follows: the tooth parts of electrode metallic foils 3 and 4 are temporarily soldered to the specified parts of the silicon element 1; a solder 7, the composition of the metal and insulating material 31, the silicon element 1 and the electrode metallic foils 34 which are soldered together, and solders 32 and 33 are set on an anode electrde plate 6; a weight is mounted on the device; and the device is made to pass in a furnace.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a semiconductor device, and particularly to a polarization structure suitable for a semiconductor element having a fine pattern structure.

A gate turn-off transistor (hereinafter referred to as GTO) or a transistor has a fine pattern on the cando side or emitter side of the semiconductor element.

Taking GTO as an example, a lead electrode made of metal foil is attached to the electrode film on the finely patterned n emitter layer and p space layer, and the lead electrode is connected to an external circuit. The reason for adopting a so-called microelectrode structure is that because one large main circuit current is controlled on and off by a control current, the potential drop in the electrode film on the surface of the semiconductor element in the main circuit and control circuit affects the switching characteristics of the element. This is to avoid any influence.

A conventional example of a fine electrode structure is shown in FIG.

In the figure, insulating materials 11a, 12as relay metal electrodes 11
C, 12C% and silver solder 11b, 12 to bond these
A metal/insulating material composite 11.12 consisting of b and 11.12 is a relay body for connecting the comb-shaped electrode metal W33, 4 and an external solder electrode (not shown), and is for candos and gates, respectively. It is bonded together with the element 1 onto the anode electrode plate 6 with solder 7.

Further, the cathode electrode metal foil 3 and the gate electrode metal foil 4 are bonded to the @ and @ films 2 on the silicon element 1 with solder 5, and are bonded to the metal/insulating material composite 11.12 with solder 13.14, respectively.
It is glued with.

Assembly of this fine electrode structure is performed, for example, as follows.

First, 11: The metal foils 3 and 4 are temporarily soldered to the electrode film 2 on the silicon element 1 with the solder 5, and each component and the adhesive solder are set in a fixing jig. The fixing jig is designed so that each component is placed in a predetermined position. Next, the set items are placed on a belt conveyor and passed through a heating furnace. When the fixing jig passes through a high temperature area, the solder melts and the parts are bonded together.

When normal adhesion is achieved, the electrode metal foils 3 and 4 are bonded to the electrode film 2 on the silicon element 1 by the solder 5, as shown in FIG. 2(a).

Is it high in the conventional structure? Figure 2 (1) after passing through M section
), many defects occurred due to contact between the gate metal foil 4 and the cathode 1 (1) and the metal foil 3.

This contact short-circuits the p base layer 11) and the n emitter 1a, thereby causing the silicon element to lose its fl+IJ capability.

As a result of investigating the cause of this defect, the following was discovered.

The gold-plated/insulating material composite 11.12 and the silicon element 1 are fixed by a fixing jig, but due to the dimensional tolerance of the jig, the images were taken as each part passed through the heating furnace. Shifts in various directions due to For example, if the gold plating insulating material composite 11.12 is shifted in opposite directions in the Y-Y direction in FIG.
are partially placed on the metal/insulating phase composites 11 and 12-H, respectively, and since an appropriate weight is placed on the electrode metal foil 3.4 during assembly, the metal/insulating material composite 11. .12, the electrode metal beak foils 3 and 4 shift in opposite directions, resulting in poor contact on the silicon element 1 as shown in FIG. 2 (bJ).

In GTO, the n emitter layer 1a is finely divided in order to control its main current, and the spacing between the electrode films 2 is usually 0.1 to 0.2.
It is about trm. Therefore, metal-insulating composites 11.1
In a structure in which the electrode metal foils 2 and 2 are independently shifted during assembly, there is a high possibility that the gate electrode metal foil 4 and the cathode electrode metal foil 3 will come into contact due to normal jig tolerances.

Furthermore, even if the metal/insulating material composites 11 and 12 are displaced in the same direction along Y-Y in FIG. 1, they will be displaced relative to the silicon element 1, and the cathode will be damaged as shown in FIG. 2(C).

There was also a contact failure between the metal foils 3 and 4 for both gates.

In this case, the silicon element 1 is a metal/insulating material composite 11°12
This is because they shift independently within the tolerance of the fixing jig during assembly.

As described above, in the conventional example, short-circuit defects between the gate and the cathode are likely to occur during assembly, which has been a major problem in mass production of fine electrode structures.

The purpose of the present invention is to assemble the cathode as described above.

It is an object of the present invention to provide a semiconductor device having a shape-only structure that is easy to mass-produce and is free from contact between gate electrode metal foils.

The present invention is characterized in that a silicon element is housed in a hole of an annular insulating material so that its periphery is circumscribed, and two relay metal strips and four electrodes are provided on a part of the main surface of the insulating material. The main feature is that each relay metal electrode and a predetermined portion of the silicon element having a fine pattern are connected by solder using two comb-shaped metal foils.

The present invention will be explained below with reference to an embodiment shown in the drawings.

(Fig. 3 shows an embodiment of the present invention, and the proboscis is the same as that in Fig. 1. The proboscis is given the same reference numeral.

In FIG. 3, a gate relay metal electrode 31c and a cathode relay metal core 4M 31e are bonded to a ceramic monkey-shaped insulation shrine 31a with silver studs 31b and 31d.
An insulating material composite 31 is formed. The silicon element 1 is formed by opposing protrusions 31 on the inner wall of the opening of the metal/insulating material composite 31.
f, 31gK are in contact and positioned. As with the conventional example, the teeth of the pHM metal foils 3 and 4 are temporarily attached to the silicon element 10 at predetermined locations with solder. material composite 31
, Silicon element engineering and electrode metal foil 3. '4 was temporarily attached, solder 32 and 33 were set, a weight was placed on it, and it was passed through a heating furnace. At this time, the solder 5, 7° 32.33 melts, but the silicon element 1, the relay metal' polarization 31 c
, 31 e does not change, so even if a slight vibration is applied, the relative positions of the silicon element 1 and the electrode metal foils 3 and 4 will hardly change due to the weight, so each part will be soldered in a predetermined position. , Figure 2 (b), (C)
No defects such as those shown in will occur.

In this embodiment, as shown in FIG.
L+1.

The distance between the walls is D = 15', and the dimension of the outer edge of the silicon element 1 is d = 15-0.03"1. In this case, the maximum gap is 0.08 mm, and The distance between the gate and the cathode is P = 0.2 rrm, and even if the silicon baby 1 is shifted to the maximum, the distance between the gate and the cathode will not be reached.Generally, the distance between the gate and the cathode is not reached.
D d)yAx 2 ' is preferable. This is because, in the case of GTO, a short circuit between the gate and the cathode is a fatal failure, so when considering the possibility of contamination by foreign matter in the subsequent process, it is necessary to provide twice as much margin.

As described above, according to the iC of the present invention, it is possible to obtain a semiconductor device having a fine electrode structure that can be easily mass-produced without causing contact between the gate and canard metal foils during assembly.

[Brief explanation of drawings]

Figure 1 shows a conventional GTO, where (a) is a plan view seen from the canard side, (b) is a sectional view taken along the line XX in (a), and Figure 2 is the same as in Figure 1 ( Cross-sectional views taken along the Y-Y cutting line of (a), (a) shows the correct adhesion situation, (b)
), (C) are diagrams showing poor adhesion, Figure 3 shows a GTO that is an embodiment of the present invention, (R) is a plan view seen from the cand side, and (b) is 4 is a sectional view taken along the Y-Y cutting line in FIG. 3(a). 1... Silicon element, 2... Electrode film, 3.4 ...
Electrode metal foil, 5.7, 32.33...Solder, 6...
Anode, JJ electrode plate, 31... metal/insulating material composite,
31a...Insulating material, 31b, 31d-...Silver solder, 31
C, 31e"'Figure 1 ((1) <b) Figure 2 (Z) (b) C□) Figure 6 (a (b)

Claims (1)

[Scope of Claims] 1. A ring-shaped insulating material and a semiconductor element housed in its opening so that its outer edge is in contact with each other are bonded onto one electrode plate, and the side of the semiconductor element that is not bonded to the electrode plate; The main surface of the insulating material has a fine pattern, and two relay metal electrodes are bonded to the main surface of the insulating material on the side that is not bonded to the electrode plate. A semiconductor device characterized in that the electrode is bonded to two comb-shaped electrode metal foils by solder. 2. The semiconductor device according to item 1 above, wherein the fine pattern of the semiconductor element has emitter layers arranged in parallel, and the toothed portions of the electrode metal foil are bonded. 3. The semiconductor device according to item 1 above, wherein the semiconductor element is in contact with both opposing inner walls of the insulating material.