JPH08316256A - Pressure contact-type semiconductor device - Google Patents

Abstract

PURPOSE: To provide a pressure contact-type semiconductor device whose life is long and whose reliability is high. CONSTITUTION: While aluminum layers 17a which form a control electrode part and aluminum layers 17b which form a main electrode part are being insulated by an insulating film 19, respective pieces of the aluminum layers 17b are coupled electrically to an aluminum film 41, and a semiconductor element which has constituted a cathode electrode part is pressurized and brought into pressure contact by a pressure contact plate 23.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor device, and more particularly to a pressure contact semiconductor device effective as a high power semiconductor device.

[0002]

2. Description of the Related Art FIG. 3 is an exploded view showing a schematic structure of a pressure contact type semiconductor device. In FIG. 3, 10 is a semiconductor element, which is a gate turn-off thyristor (hereinafter abbreviated as GTO), and 21 is a copper anode. A cap, 22 is a silver (Ag) foil, 23 is a disc-shaped molybdenum plate, 24 is rubber as a cushion member, 25 is a PPS ring, 26 is a molybdenum ring plate, 27 is a ring-shaped silver foil, 27
Is a cylinder made of ceramic, 28 is a copper cathode block, 29 is a weld ring made of Cu-Ni-Co,
Reference numeral 30 is a gate block arranged in the copper cathode block 28.

As shown in FIG. 4, the gate block 30 is composed of a PPS cup 31, a copper gate block 32, a gate lead wire 33, a stainless copper plate 34 and a stainless plate spring 35.

Generally, a molybdenum plate that serves as an external extraction electrode and a high thermal conductivity / stress relaxation material is sandwiched between both surfaces of a Si wafer on which impurity diffusion and electrode patterns are formed. Furthermore, it is pinched from both sides with copper blocks and pressure is applied to adhere them. At this time, a silver foil is interposed between the copper block and the molybdenum plate in order to mitigate adverse effects (such as generation of stress) due to the difference in thermal expansion between the copper material and the molybdenum plate. This copper foil is very soft in terms of material and has an effect of filling a space portion caused by initial strain and surface roughness of the copper block and molybdenum plate. Further, since it has high thermal conductivity, there is no deterioration in thermal conductivity caused by the interposition.

FIG. 5 shows the element structure of the gate turn-off thyristor 10, in which 11 is an N base layer, 12 is a P emitter layer, 13 is an anode / short layer made of an N ⁺ layer, and 14 is a metal layer. Therefore, the P emitter layer 12, the anode / short layer 13, and the metal layer 14 form the anode electrode portion A. Reference numeral 15 is a P base layer, 16 is an N emitter layer, 17a and 17b are aluminum layers, 18 is an insulating layer made of SiO ₂ , and 19 is a polyimide layer as an insulating protective film.

In the semiconductor device of FIG. 5, N-type crystalline Si
A P base region is formed thereon by diffusion or the like. After that, a diffusion window is provided, an N emitter layer is formed, and an oxide film (SiO ₂ ) is formed so that the gate part and the cathode part are not short-circuited.
After forming the electrode, an electrode is formed with aluminum to lead to the outside. Finally, an insulating protective film (currently a polyimide film) is applied to the cathode / gate fine pattern portion to prevent long-term deterioration of the withstand voltage between G and K.

[0007]

FIGS. 6 and 7 show the state of the upper part of the wafer when it is actually pressure-contacted.
0b shows a cooling fin.

When the aluminum electrode portion is pressed against the molybdenum plate, stress concentration portions 40 are generated at the corner portions of the electrode. Further, as the Si wafer is heated and cooled, the stress applied to this portion further increases. The yield stress of the molybdenum plate itself (stress that causes plastic deformation and does not return to its original shape) is extremely low, but it is unlikely to deform, but the yield stress of the cathode aluminum electrode is low, and plastic deformation easily occurs due to pressure and heat generation. Cause

The aluminum electrode that has been plastically deformed by heat generation has an angular shape as a result, and a gap is formed at this portion during cooling. When this heat generation / cooling is repeated, this gap gradually expands, and eventually the aluminum electrode pattern is not contacted at all even under pressure. In this state, the contact area becomes smaller than that in the initial state, which causes deterioration of electrical characteristics, and the heat generated from the Si wafer mainly passes through the aluminum electrode and the molybdenum plate / Cu.
Although it conducts to the block, the thermal conductivity deteriorates as the contact area decreases, and the Si temperature rises, causing thermal runaway (thermal destruction).

Further, since a minute gap is formed between the aluminum electrode and the molybdenum plate, a spark (spark) -like electric current is apt to occur at this portion. When this phenomenon occurs, the temperature instantaneously rises in this gap, which causes thermal destruction and melting of the aluminum electrode.

The present invention has been made in view of the above problems, and an object thereof is to provide a pressure contact type semiconductor device having a long life and high reliability.

[0012]

In order to achieve the above object, a pressure contact type semiconductor device of the present invention has a first main electrode portion on one surface of a semiconductor wafer and a second main electrode portion on the other surface. First metal pressure contact plates are arranged on both sides of a semiconductor element having a main electrode part and a control electrode part, and metal pressure contact blocks are arranged on these first pressure contact plates via foil members. In the semiconductor device in which the semiconductor element is pressed from both sides thereof, the second main electrode portion of the insulating layer is maintained while being electrically insulated from the control electrode portion through the insulating layer. It is characterized in that it is configured by being combined with a metal layer provided on the upper part.

[0013]

The structure has a structure in which molybdenum plates (tungsten plates are also acceptable) are sandwiched between both sides of a patterned Si wafer, and both sides are pressed with Cu blocks via silver foil (aluminum is acceptable). The cathode electrode is bonded to the gate electrode while being electrically insulated from the gate electrode to prevent uneven current distribution due to the electrode not being pressed. By flattening the cathode electrode in a single layer, it is possible to improve the adhesion to the molybdenum plate and to eliminate the stress concentration portion that is applied to the corner portion of the cathode electrode.

[0014]

Embodiments of the present invention will be described below with reference to FIGS.

1 and 2 each show a main part of a pressure contact type semiconductor device according to an embodiment of the present invention. In these figures, the same or corresponding parts in FIGS. 3 to 7 are designated by the same reference numerals. There is.

In the same manner, the anode electrode portion forms the first main electrode portion, the aluminum electrode 17a forms the gate electrode portion which is the control electrode portion, and the aluminum electrode 17b.
Forms a cathode electrode portion which is a second main electrode portion. In this embodiment, as shown in FIG. 1, an aluminum film 41 is formed on the insulating film 19 made of polyimide or Si nitride film.
And each aluminum electrode 17b is electrically joined to the aluminum film 41. After that, the surface of the aluminum film 41 is flattened and pressed against the molybdenum plate 22.

When the insulating layer is formed, the height of the surface of the insulating film and the height of the cathode electrode are usually not aligned. If the aluminum film is laminated as it is by a sputtering apparatus, a vapor deposition apparatus or the like, the surface of the aluminum film is as shown in FIG. A step portion 42 is generated at the position. Therefore, the surface of the aluminum film or the aluminum electrode / insulating film before aluminum film formation is planarized by chemical / mechanical polishing / etching.

By forming an aluminum electrode entirely on the insulating layer, the stressed portion is made uniform over a wide range, and the stress value is also reduced. On the contrary, since a force is applied on the insulating film, the insulating film to be used must be hard. If it is a soft film, this insulating portion will be recessed during pressure contact, and a crack will occur in the aluminum film.

However, the polyimide resin currently used is soft in material and may not have resistance to application of high pressure. Therefore, the material of this insulating layer is, for example, S
It is necessary to change the iN nitride film, the oxide film, the composite layer with another insulating film, and the like.

[0020]

The present invention is as described above. For example, by combining the cathode electrode patterns on the Si wafer into one and flattening it, the contact area when the molybdenum plate and the aluminum electrode are pressure-welded is significantly increased. Increase to. Further, it is possible to improve the imbalance of the current density due to the non-uniformity of the contact that occurs during the pressure welding. Further, by flattening, it is possible to prevent local concentration of stress generated at the corners of the aluminum electrode portion and prevent crushing of the aluminum electrode. Therefore, deterioration of the element due to long-term operation is improved.

[Brief description of drawings]

FIG. 1 is a pattern diagram of a main part of a pressure contact type semiconductor device according to an embodiment of the present invention.

FIG. 2 is a pattern diagram of a main part of a pressure contact type semiconductor device according to an embodiment of the present invention.

FIG. 3 is an exploded view of a semiconductor device having a flat pressure contact structure.

4 is an enlarged cross-sectional view of a part of the semiconductor device of FIG.

FIG. 5 is a pattern diagram of a gate turn-off thyristor showing an example of a device structure.

FIG. 6 is a pattern diagram of a conventional pressure contact type semiconductor device.

7 is a partially enlarged view of the pressure contact type semiconductor device of FIG.

[Explanation of symbols]

10 ... gate turn-off thyristors 15 ... P base layer 16 ... N emitter layer 17a, 17b ... aluminum layer 18: insulating layer (SiO ₂₎ 19 ... insulating plate (polyamide) 23 ... press-in plate (molybdenum plate) 41 ... aluminum film

Claims (1)

[Claims] 1. A semiconductor element having a first main electrode section on one surface of a semiconductor wafer and a second main electrode section and a control electrode section on the other surface of the semiconductor wafer. In the semiconductor device in which the pressure contact plates are disposed, the metal pressure contact blocks are disposed on the first pressure contact plates via the foil members, and the semiconductor element is pressed from both sides thereof, A pressure contact type semiconductor device, characterized in that the main electrode portion is connected to a metal layer provided on the insulating layer while maintaining a state of being electrically insulated from the control electrode portion via an insulating layer. .