JP2714115B2 - Power semiconductor switch device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention relates to a power semiconductor switch device.

(Prior Art) Recently, semiconductor switch devices have been widely used as switches in power circuits. Of these applications, several hundred amps
The power semiconductor switch device used in the above power circuit is usually configured as shown in FIG. That is, a semiconductor chip 3 constituting a semiconductor switching element represented by a thyristor or GTO is interposed between a copper anode electrode 1 also serving as a heat sink and a copper cathode electrode 2 also serving as a heat sink. A dissimilar metal layer 4 is interposed between the electrodes 1 and 2, and a gate electrode line 5 is extracted from between the anode electrode 1 and the cathode electrode 2. In addition,
The space in which the semiconductor chip 3 is accommodated is actually covered with an envelope (not shown). When mounting, a large pressing force is applied between the anode electrode 1 and the cathode electrode 2 as shown by a thick arrow in the figure, thereby causing a thermal resistance between the semiconductor chip 3 and the anode electrode 1 and the cathode electrode 2. Is reduced to a sufficiently small value. The dissimilar metal layer 4 has a role of alleviating generation of a large thermal stress in the semiconductor chip 3 due to a difference in thermal expansion coefficient between the anode electrode 1 and the cathode electrode 2 and the semiconductor chip 3.

However, the conventional power semiconductor switch device configured as described above has the following problems. That is, when the current capacity is to be increased, it is necessary to improve the heat radiation characteristics from the semiconductor chip 3. In the conventional switch device, since the electrode surface of the semiconductor chip 3 is used as a heat radiation path, it is necessary to apply a large pressing force to promote heat radiation from the semiconductor chip 3. However, since a comparative relationship is not always established between the pressure contact force and the heat radiation characteristics, it is difficult to reduce the thermal stress generated in the semiconductor chip 3 by applying a large pressure contact force, which is called cooling. From the aspect, it was difficult to increase the capacity. In order to increase the current capacity, the diameter of the semiconductor chip 3 may be increased. In this case, a large semiconductor wafer is required. However, in order to improve the switching characteristics, it is generally necessary to perform fine processing on a semiconductor wafer, and it is extremely difficult to perform fine processing with uniform accuracy on the entire large semiconductor wafer as described above. For this reason, in the conventional power semiconductor switch device, it is difficult to form a switch having a large capacity and good switching characteristics due to structural restrictions.

(Problems to be Solved by the Invention) As described above, in the conventional power semiconductor switch device, it is structurally difficult to increase the capacity and the performance.

Accordingly, it is an object of the present invention to provide a power semiconductor switch device having a structure capable of realizing a large capacity and high performance and which is easy to mount.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above object, in one example of a power semiconductor switch device according to the present invention, an anode electrode also serving as a heat sink and a cathode serving also as a heat sink are provided. In a structure in which a chip constituting a semiconductor switching element is sandwiched between electrodes and a gate electrode line is taken out from between the two electrodes, a plurality of independent semiconductor switching element chips are arranged in parallel between the two electrodes by the number that satisfies the rated current condition. The semiconductor switching element is constituted by a semiconductor switching element chip group arranged in the above manner. In this example, around each semiconductor switching element chip, a heat transfer means having a function of relaxing thermal stress generated in the chip and a function of transmitting heat to the two electrodes from the entire outer surface of the chip. It is more effective if provided.

(Operation) In one example described above, a switch device having a desired current capacity can be obtained only by selecting the number of semiconductor switching element chips arranged in parallel between the anode electrode and the cathode electrode. . In general, it is much easier to manufacture a small semiconductor switching element chip than to manufacture a large semiconductor switching element chip, and it is possible to increase the precision of fine processing. That is, a small-sized semiconductor switching element chip having uniform characteristics and high performance can be manufactured. Therefore, with the above-described configuration, a system in which a required number of small and high-performance semiconductor switching element chips are arranged in parallel can be adopted, so that a large capacity and high performance can be achieved. In addition, when the chips are divided and the total cross-sectional area is the same, the area of the side surface of the chip can be greatly increased as compared with the case where a single chip is provided. Therefore, heat can be radiated by effectively using the side surface of the chip. For this reason, the problem of heat dissipation which becomes a problem when trying to increase the capacity can be solved. Also,
When this structure is combined with the above-described heat transfer means, the heat radiation characteristics can be further improved while each chip is protected. Therefore, it is possible to solve the problem of heat radiation without employing the pressure welding method.

(Example) Hereinafter, an example is described, referring to drawings. FIG. 1 is a plan view of a power semiconductor switch device according to one embodiment. As shown in FIG. 2, this switch device is roughly divided into, for example, an anode electrode 11 and a cathode electrode 12 which are formed of a copper plate having a thickness of 10 mm and a diameter of 90 mm, and are arranged opposite to each other. Aluminum nitride plate 13 (hereinafter abbreviated as AlN plate) and this Al
Anode electrode 11 and cathode electrode surrounded by N plate 13
As shown in FIG. 1, ten small-sized semiconductor switching element chips 14 (hereinafter simply referred to as chips) are interposed between an anode electrode 11 and a cathode electrode 12. And an insulating envelope 15 that seals the portion.

The AlN plate 13 includes a first AlN plate located on the anode electrode 11 side.
The plate 16 and a second AlN plate 17 located on the side of the cathode electrode 12 are stacked. As shown in FIG. 3 and FIG. 4, when the plates are overlapped with each other, a portion located between the center and the outer edge of the first Al N plate 16 and the second Al N plate 17 has an anode. A hole 1 forming a cubic space 18 having a side of 10 mm between the electrode 11 and the cathode electrode 12, for example.
9 and 20 are provided at equal intervals along a circle drawn with the center as the center point. As shown in FIG. 4, a molybdenum plate 21 is interposed between the anode electrode 11 and the cathode electrode 12 in the ten cubic spaces 18 formed by the holes 19 and 20, respectively. The above-mentioned chip 14 is mounted in a state where it is in a closed state. The total thickness of the two molybdenum plates 21 and the chips 14 is set to a value slightly larger than the thickness of the AlN plate 13.

Each chip 14 has an MO formed using a silicon wafer.
It is composed of insulated gate transistors such as SFET and IGBT. As shown in FIG. 1, the gate electrode lines 22 of these chips 14 are formed along the grooves 23 formed in the mating surfaces of the AlN plates 16 and 17 as shown in FIGS. , Which are led to the center of the device and connected in common. From this connection point, as shown in FIG. 2, a lead wire 24 also extends in the outer peripheral direction along the inside of the groove 23, and this lead wire 24 is guided to the outside through the wall of the envelope 15. .

The power semiconductor switch device thus configured is
As shown by a thick arrow in FIG. 4, the mounting is performed in a state where a pressing force is applied. The molybdenum plate 21 has a coefficient of thermal expansion close to that of the silicon constituting the chip 14. Therefore, the molybdenum plate 21 has a chip shape due to a difference in thermal expansion coefficient between the anode electrode 11 and the cathode electrode 12 and the chip 14.
It contributes to alleviating the thermal stress in the thickness direction generated at 14. Further, the AlN plate 13 has a thermal expansion coefficient close to the thermal expansion coefficient of silicon, is excellent in insulating properties, and excels in thermal conductivity. For this reason, the AlN plate 13 relieves the thermal stress in the radial direction generated in each chip 14 and maintains the insulation between the anode electrode 11 and the cathode electrode 12 in each chip 14.
It contributes to reducing the thermal resistance between the outer peripheral surface of 14 and the anode electrode 11 and the cathode electrode 12.

As described above, in this example, ten independent small chips 14 are arranged in parallel between the anode electrode 11 and the cathode electrode 12 to constitute a switch device. Assuming that the rated current capacity per one chip 14 is 50 A, a switch device having a current capacity of 500 A can be obtained. Of course, chips
By selecting the fourteen numbers, a switch device having a desired current capacity can be obtained. In general, when trying to obtain a current capacity of 500 A with one semiconductor chip and to obtain good switching characteristics, it is difficult to make the accuracy of each part uniform due to manufacturing restrictions, and a high-performance chip is manufactured. It is extremely difficult to do this. However, if the required current capacity is obtained by arranging a plurality of chips 14 in parallel as in this embodiment, it is possible to adopt a method of arranging a required number of small and high-performance chips in parallel. And high performance can be achieved. In addition, when the chips are subdivided and the total cross-sectional area is the same, the area of the side surface of the chip can be greatly increased as compared with the case where a single chip is provided. Therefore, it is possible to effectively radiate heat by effectively using the side surfaces. Therefore, heat can be satisfactorily dissipated without requiring a large pressing force, and the heat stress generated in each chip can be easily alleviated. Stress problems can also be solved. As in this embodiment, when the chips 14 are arranged along the same circle and the gate electrode lines 22 are connected in common at the center of the circle, the switching characteristics of the chips 14 are made uniform. can do.

FIGS. 5 and 6 show only a main part of a power semiconductor switch device according to another embodiment, that is, only a portion corresponding to FIGS. 3 and 4. Therefore, the description of the overlapping part will be omitted.

The power semiconductor switch device according to this embodiment is different from the above-described embodiment in that the shape of the holes 19a and 20a provided in the first and second Al N16 and
And the material of a metal material interposed between the cathode electrode 12 and the metal material. That is, in the holes 19a and 20a, a large-diameter portion 31 is formed in the center as shown in FIG. 6 in a cubic space 18a formed when the first and second AlNs 16 and 17 are overlapped. Is formed in the step hole. The large-diameter portion 31 sandwiches the peripheral portion of the chip 14 when the first and second AlN plates 16 and 17 are overlapped, and has a size such that a slight annular gap is formed on the outer periphery of the chip 14. Is formed. Meanwhile, the chip 14, the anode electrode 11, and the cathode electrode
12 and between the AlN plate 13 and the anode electrode 11 and the cathode electrode 12 are filled with a low melting point metal such as a lead-tin alloy having high conductivity and heat conductivity. The low melting point metal layer 32 is melted and the chip 14 and the anode electrode 11 and the cathode electrode 12 are soldered.

With such a configuration, when the chip 14 generates heat during energization, the low melting point metal layer 32 is melted, thereby
Thermal resistance between the anode electrode 11 and the cathode electrode 12 is greatly reduced. Also, the AlN plate 13 dissipates heat from the outer peripheral surface of the chip 14 to the anode electrode 11 and the cathode electrode 12.
Tell promptly. Therefore, the heat radiation characteristics are improved as compared with the embodiment. In addition, the low melting point metal layer 32 and Al N
The plate 13 relieves the thermal stress generated in the chip 14. Therefore, in this embodiment, it is possible to eliminate the need for a pressure welding jig while achieving a large capacity and high performance.

7 and 8 show only a main part of an electric semiconductor switch device according to still another embodiment, that is, only a portion corresponding to FIGS. 2 and 4. FIG. Therefore,
The description of the overlapping part is omitted.

This embodiment differs from the above embodiments in the configuration of the anode electrode 11a and the cathode electrode 12a and the insulating material filled between the electrodes. That is, on the surface of the anode electrode 11a and the cathode electrode 12a located on the chip 14 side,
As shown in FIG. 8, a concave portion 41 having a diameter larger than the diameter of the tip 14 is provided.
Are provided in each of these concave portions 41.
An AlN plate 42 large enough to completely fill 41 is mounted.
Then, a thin copper plate 43 is welded to the inside thereof so as to cover the concave portion 41. Both ends of the chip 14 are anode electrodes
The copper plate 43 located on the 11a side and the copper plate 43 located on the cathode electrode 12a side are welded. On the other hand, the anode electrode 11
The space existing between the copper plate 43 located on the a side and the copper plate 43 located on the cathode electrode 12a side is filled with AlN powder.

With such a configuration, the chip 14 is welded to the thin copper plate 43, and furthermore, this chip
It is adjacent to the AlN plate 42 having a thermal expansion coefficient close to the thermal expansion coefficient of the silicon constituting 14. Also,
The outer peripheral surface of the chip 14 is in contact with the AlN powder 44, which easily absorbs force and has excellent thermal conductivity. Therefore, also in this embodiment, it is possible to eliminate the need for a press-fitting jig while achieving a large capacity and high performance.

FIG. 9 shows only a main part of a power semiconductor switch device according to still another embodiment, that is, only a portion corresponding to FIG. 8 in the above embodiment. Therefore, the description of the overlapping part will be omitted.

This embodiment differs from the previous embodiment in that the anode electrode
11b and the configuration of the cathode electrode 12b and an insulating material filled between the electrodes. In other words, the portions of the anode electrode 11b and the cathode electrode 12b that face the chip 14 extend over a range larger than the diameter of the chip 14.
And a silicon plate 51 having the same thickness as that of the material constituting the silicon substrate 51 and having a predetermined thickness. A chip 14 is provided in a space formed between the anode electrode 11b and the cathode electrode 12b.
Are filled with silicon plates 53 and 54 whose surfaces are covered with a silicon oxide layer 52.

With such a configuration, the chip 14 has a relatively thin copper layer.
Through 55, it is adjacent to the same silicon plate 52 as the silicon constituting the chip 14. The outer peripheral surface of the chip 14 is also in contact with the silicon plate 53 having the same thermal expansion coefficient via the silicon oxide layer 52. Therefore,
Due to the presence of the silicon plates 51 and 53, the chip 14
Is reduced. Further, the heat generated in the chip 14 can be directly transmitted to the anode electrode 11b and the cathode electrode 12b on the one hand, and can be transmitted to the anode electrode 11b and the cathode electrode 12b via the silicon plate 53 on the other hand. Therefore, also in this embodiment, it is possible to eliminate the need for a press-fitting jig while achieving a large capacity and high performance.

FIG. 10 is a longitudinal sectional view of a power semiconductor switch device according to still another embodiment of the present invention. In this figure, the same parts as those in FIGS. 2 and 4 are indicated by the same reference numerals. Therefore, a detailed description of the overlapping part will be omitted.

In each of the above-described embodiments, a plurality of chips 14 are interposed between the anode electrode 11 and the cathode electrode 12. In this embodiment, only one chip 14 is interposed. Even in such a configuration, the heat generated by the chip 14 is transmitted through the entire outer surface of the chip 14 while the thermal stress of the chip 14 is reduced by the presence of the molybdenum plate 21 and the AlN plate 13.
11 and the cathode electrode 12 can be transmitted well.
Therefore, the heat radiation characteristics can be improved as compared with the conventional device, and a large capacity can be realized using the same chip.

[Effects of the Invention] As described above, according to the present invention, the problem on heat radiation,
Problems of thermal stress and problems of manufacturing can be easily solved, and large capacity and high performance can be realized. Also, since the capacity can be easily increased, it is possible to omit the pressure welding jig.
The implementation can be simplified.

[Brief description of the drawings]

FIG. 1 is a plan view of a power semiconductor switch device according to one embodiment, FIG. 2 is a sectional view taken along the line AA in FIG. 1, and FIG. FIG. 4 is an exploded perspective view of the power semiconductor switch device, taken along line BB in FIG. 1 and viewed in the direction of the arrow, and FIG. 5 is a partial view of a power semiconductor switch device according to another embodiment. FIG. 6 is an exploded perspective view schematically showing the device, FIG. 6 is a sectional view showing only a main part of the device, FIG. 7 is a longitudinal sectional view of a power semiconductor switch device according to still another embodiment, and FIG. Is a cross-sectional view showing only the main part of the same device, FIG. 9 is a cross-sectional view showing only the main part of a power semiconductor switch device according to still another embodiment, and FIG. FIG. 11 is a longitudinal sectional view of a power semiconductor switch device, and FIG. 11 is a conventional power semiconductor switch device. It is a schematic view. 11,11a, 11b ... Anode electrode, 12,12a, 12b ... Cathode electrode, 13,16,17,42 ... Al N plate, 14 ... Semiconductor switching element chip, 15 ... Envelope, 21 ... ... Molybdenum plate,
22 ... Gate electrode wire, 24 ... Lead wire, 32 ... Low melting point metal layer, 43 ... Copper plate, 44 ... Al N powder, 51 ... Silicon plate, 53,54 ... Surface covered with silicon oxide Broken silicon plate.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Yoshio Kamei, Inventor 1 Kosuka Toshiba-cho, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture (72) Inventor Asako Matsuura 1 Kosuka-Toshiba-cho, Sachi-ku, Kawasaki-shi, Kanagawa Address Toshiba Research Institute, Inc. (56) References

Claims (5)

(57) [Claims] 1. A power semiconductor switch device comprising a chip constituting a semiconductor switching element sandwiched between an anode electrode also serving as a heat sink and a cathode electrode also serving as a heat sink, and a gate electrode line being taken out between the two electrodes. The semiconductor switching element is constituted by a semiconductor switching element chip group in which independent semiconductor switching element chips are arranged in parallel between the two electrodes by a number that satisfies a rated current condition. Semiconductor switch device. 2. The semiconductor switching element chip according to claim 1,
The power semiconductor switch device according to claim 1, wherein the power semiconductor switch devices are arranged along the same circle. 3. A heat transfer means is provided around each of said semiconductor switching element chips to relieve thermal stress generated in each of said chips and to transmit heat from the surface of each of said chips to said two electrodes. The power semiconductor switch device according to claim 1. 4. The power supply according to claim 3, wherein said semiconductor switching element chip is made of silicon, and said heat transfer means is made up of an aluminum nitride layer or a silicon layer embedded in said electrodes. Semiconductor switch device. 5. The semiconductor switching element chip is made of silicon, the heat transfer means is an aluminum nitride layer,
4. The power semiconductor switch device according to claim 3, wherein the power semiconductor switch device is configured to include one selected from a powdery aluminum nitride layer and a silicon layer whose surface is covered with a silicon oxide layer.