Semiconductor power module

Abstract

To provide a semiconductor power module in which an electric field at corners of electrodes and around the corners is reduced while size reduction and voltage increase are achieved.SOLUTION: A semiconductor power module includes a plurality of electrodes which each have different potentials and which are laid over each other with an interval therebetween. As seen in a direction in which the plurality of electrodes are laid over each other, each of the plurality of electrodes has three or more corners; at least a part of the corner of one of the adjacent two electrodes is provided on an outer side relative to the other of the adjacent two electrodes; and at least a part of the corner of the other electrode is provided on an outer side relative to the one electrode.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] This application relates to a semiconductor power module.

Background Art

[0002] As semiconductor power modules equipped with semiconductor elements, there are transfer mold type semiconductor power modules in which the semiconductor elements are encapsulated with thermosetting resins such as epoxy resins, and gel encapsulation type semiconductor power modules in which the semiconductor elements are encapsulated with gel-like resins. In particular, transfer mold type semiconductor power modules are widely used for power control and the like because they are small and have excellent reliability.

[0003] Inside such semiconductor power modules, an insulator is provided between electrodes to ensure insulation between electrode pairs so that dielectric breakdown does not occur in electrode pairs with a potential difference. Locations where insulation is ensured are, for example, between the heat spreader and the cooler, and between overlapping lead frames. If miniaturization and high voltage operation of the semiconductor power module are simply advanced in such locations where such insulation is required, higher electric fields will be generated at these locations. In particular, the corners of the electrodes have high electric fields, leading to a decrease in insulation performance. Therefore, there has been a problem that miniaturization and high voltage operation are difficult at the corners of the electrodes.

[0004] Also, a technique has been disclosed in which lead frames through which current flows in opposite directions are arranged in close proximity so as to overlap vertically, canceling out the inductances with each other in this region, reducing the internal inductance of the semiconductor power module, and reducing the impedance loss associated with high-frequency operation of the operating voltage of the semiconductor power module (for example, Patent Document 1). In this technique, it is necessary to bring the electrodes close to each other until the effect of inductance reduction is obtained. If the electrodes are brought too close to each other, the electric field between the electrodes will exceed the breakdown electric field strength of the insulator, so there is a possibility of dielectric breakdown occurring between the electrodes. Therefore, it is necessary to ensure a certain distance or more between the electrodes, and there has been a problem that the effect of inductance reduction cannot be sufficiently obtained.

[0005] To address these challenges, a semiconductor power module has been disclosed in which, in an electrode pair with a potential difference, one electrode is made to protrude further than the other electrode, thereby mitigating the electric field around the corner of one electrode (for example, Patent Document 2). Furthermore, a semiconductor power module has been disclosed in which one electrode is made to protrude further than the other electrode, and the protruding portion is filled with an insulator having a dielectric constant lower than that of the sealing resin, thereby mitigating the electric field (for example, Patent Document 3). [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2018-137283 [Patent Document 2] Japanese Patent Publication No. 2019-197816 [Patent Document 3] Japanese Patent Publication No. 2012-9815 [Overview of the project] [Problems that the invention aims to solve]

[0007] In the semiconductor power module configuration described in Patent Document 2, it is possible to mitigate the electric field around the corner of one of the protruding electrodes. However, the electric field around the corner of the other electrode that overlaps with the protruding electrode is higher than when the corners of the two electrodes overlap in a plan view. Because the electric field around the corner of the other electrode is higher, the insulation performance around that corner deteriorates, which presents a challenge in miniaturizing and increasing the voltage of the semiconductor power module.

[0008] In the semiconductor power module configuration described in Patent Document 3, the corners of one electrode and the other electrode overlap in a plan view. However, because an insulator with a dielectric constant lower than that of the sealing resin is filled in, the electric field around the corners of the electrodes can be suppressed. However, this requires filling with an insulator with a dielectric constant lower than that of the sealing resin, which increases the manufacturing process and thus the cost. Furthermore, because an insulator must be filled in addition to the sealing resin, the probability of insulation defects such as voids increases, thus posing a challenge in that the risk of a decrease in the insulation performance of the semiconductor power module increases.

[0009] Therefore, the present invention aims to provide a semiconductor power module that is miniaturized and can handle high voltage while suppressing the electric field at the corners of the electrodes and the areas surrounding them. [Means for solving the problem]

[0010] The semiconductor power module disclosed herein comprises a plurality of electrodes, each having a different potential and overlapping with a gap between them, and each of the plurality of electrodes having three or more corners when viewed in the direction in which the plurality of electrodes overlap, and in two adjacent electrodes, at least a portion of the corner of one electrode is located outside the other electrode, and at least a portion of the corner of the other electrode is located outside the first electrode. Each of the multiple electrodes is formed in a plate shape, and when viewed in the direction in which the multiple electrodes overlap, the edges on both sides of the corners of each of two adjacent electrodes overlap, and in each of two adjacent electrodes, the first part of one corner is further out than the first part of the other corner, and the second part of the other corner is further out than the second part of the first corner, so that the shape of one corner and the shape of the other corner are different. It is something that exists. [Effects of the Invention]

[0011] The semiconductor power module disclosed in this application comprises a plurality of electrodes, each having a different potential and overlapping with a gap in between. When viewed in the direction in which the plurality of electrodes overlap, each of the plurality of electrodes has three or more corners. In two adjacent electrodes, at least a portion of the corner of one electrode is located outside the other electrode, and at least a portion of the corner of the other electrode is located outside the first electrode. Therefore, when viewed in the direction in which the two adjacent electrodes overlap, the corner of one electrode does not overlap with the other electrode, and when viewed in the direction in which the two adjacent electrodes overlap, the corner of the other electrode does not overlap with the first electrode. As a result, the electric field at the corners and around the corners of both electrodes can be suppressed. Because the electric field at the corners and around the corners of the electrodes is suppressed, the semiconductor power module can be made to operate at a higher voltage. Furthermore, since no additional insulator is required, as in Patent Document 3, the manufacturing process does not increase and the probability of insulation defects such as voids does not increase, thus the semiconductor power module can be made smaller and to operate at a higher voltage. [Brief explanation of the drawing]

[0012] [Figure 1] This is a schematic plan view of a semiconductor power module according to Embodiment 1. [Figure 2] This is another plan view showing a schematic of the semiconductor power module according to Embodiment 1. [Figure 3] This is a schematic cross-sectional view of a semiconductor power module cut at the AA cross-sectional position in Figure 1. [Figure 4] This is a plan view showing the electrodes of a comparative example semiconductor power module. [Figure 5] This figure shows the normalized electric field value at the corner of the electrode in the comparative example. [Figure 6] This is an equipotential diagram around the electrodes of a comparative semiconductor power module. [Figure 7] This is a plan view showing the corners of two adjacent electrodes in a semiconductor power module according to Embodiment 1. [Figure 8]It is a diagram showing the normalized electric field value at the corner of the electrode of the semiconductor power module according to Embodiment 1. [Figure 9] It is a plan view showing two adjacent electrodes of the semiconductor power module according to Embodiment 2. [Figure 10] It is a plan view showing the corners of two adjacent electrodes of the semiconductor power module according to Embodiment 2. [Figure 11] It is a diagram for explaining the movement of the corners of two adjacent electrodes of the semiconductor power module according to Embodiment 2. [Figure 12] It is a diagram showing the normalized electric field value at the corner of the electrode of the semiconductor power module according to Embodiment 2. [Figure 13] It is an equipotential line diagram around the electrode at the cut position along the C-C cross-section of FIG. 11. [Figure 14] It is a diagram showing an approximate curve of the normalized electric field value at the corner of the electrode of the semiconductor power module according to Embodiment 2. [Figure 15] It is a plan view showing two adjacent electrodes of another semiconductor power module according to Embodiment 2.

Embodiments for Carrying Out the Invention

[0013] Hereinafter, the semiconductor power module according to the embodiment of the present application will be described based on the drawings. In each figure, the same or corresponding members and parts will be denoted by the same reference numerals and described.

[0014] Embodiment 1. Figure 1 is a schematic plan view of the semiconductor power module 101 according to Embodiment 1, with the sealing resin 12 removed. Figure 2 is another schematic plan view of the semiconductor power module 101, with the third lead frame 7 removed from Figure 1. Figure 3 is a schematic cross-sectional view of the semiconductor power module 101 cut at the AA cross-sectional position in Figure 1. The semiconductor power module 101 for power conversion is mainly mounted in a power conversion device that converts desired power into DC or AC voltage. The semiconductor power module 101 has power semiconductor elements inside. The semiconductor power module 101 converts power through the switching operation of the power semiconductor elements.

[0015] <Semiconductor Power Module 101> As shown in Figure 3, the semiconductor power module 101 comprises a heat spreader, power semiconductor elements, a conductive layer, a lead frame, and a sealing resin 12. In this embodiment, the outer shape of the semiconductor power module 101 is formed in a rectangular parallelepiped shape by the sealing resin 12. The heat spreader consists of a first heat spreader 2 and a second heat spreader 5, which are formed in a plate shape and arranged side by side. The power semiconductor elements consist of a first power semiconductor element 3, which is formed in a plate shape and bonded to one side of the first heat spreader 2, and a second power semiconductor element 6, which is bonded to one side of the second heat spreader 5. The conductive layer consists of a first conductive layer 9, which is formed in a plate shape and bonded to the other side of the first heat spreader 2 via a first insulating layer 8, which is an insulating layer, and a second conductive layer 11, which is bonded to the other side of the second heat spreader 5 via a second insulating layer 10, which is an insulating layer.

[0016] The lead frame is a wiring component. The lead frame consists of a first lead frame 1, whose other side is joined to one side of the first heat spreader 2; a second lead frame 4, which extends in the direction in which the heat spreaders are arranged, with one end of its other side joined to one side of the first power semiconductor element 3 and the other end of its other side joined to one side of the second heat spreader 5; and a third lead frame 7, which extends in the direction in which the heat spreaders are arranged, with one end of its other side joined to one side of the second power semiconductor element 6, and at least a portion of its other side overlapping one side of the second lead frame 4 with a gap between them.

[0017] In the following explanation, the upper surface of the semiconductor power module 101 shown in Figure 3 will be referred to as one surface 101a of the semiconductor power module 101, and the lower surface of the semiconductor power module 101 will be referred to as the other surface 101b of the semiconductor power module 101. However, one surface 101a is not limited to the surface that is positioned on top when the semiconductor power module 101 is in use, and the other surface 101b is not limited to the surface that is positioned on bottom when the semiconductor power module 101 is in use. Figures 1 and 2 are plan views of the semiconductor power module 101 shown in Figure 3, viewed from the side of one surface 101a, and show the internal configuration when the sealing resin 12 is removed. The dashed lines in Figures 1 to 3 show the outer shape of the sealing resin 12.

[0018] Details of each component constituting the semiconductor power module 101 will now be described. The first heat spreader 2 and the second heat spreader 5 are flat plate-shaped components that form the base of the entire semiconductor power module 101. In this embodiment, as shown in Figure 1, the shapes of the first heat spreader 2 and the second heat spreader 5 are rectangular when viewed perpendicular to one surface 101a. The shapes of the first heat spreader 2 and the second heat spreader 5 are not limited to rectangles; for example, they may be polygons other than rectangles. In this embodiment, since the shapes of the first heat spreader 2 and the second heat spreader 5 are rectangular, each of the first heat spreader 2 and the second heat spreader 5 has four corners when viewed perpendicular to one surface 101a. The lower left corner of the first heat spreader 2 is referred to as corner 2c.

[0019] As shown in Figure 3, when the direction perpendicular to the first surface 2a, which is one face of the first heat spreader 2, is defined as the height direction (hereinafter referred to as the Z direction), the dimensions of the first heat spreader 2 and the second heat spreader 5 in the Z direction are equal. The first surface 2a of the first heat spreader 2 and the first surface 5a, which is one face of the second heat spreader 5, are in the same plane, and the second surface 2b, which is the other face of the first heat spreader 2, and the second surface 5b, which is the other face of the second heat spreader 5, are in the same plane. The heights of the first heat spreader 2 and the second heat spreader 5 are not limited to these, and the heights of the first heat spreader 2 and the second heat spreader 5 may be different.

[0020] The material used for the heat spreader is a metal with excellent electrical and mechanical properties. Suitable materials for the heat spreader include, for example, an alloy containing one or more of the following metals: aluminum (Al), copper (Cu), silver (Ag), nickel (Ni), and gold (Au), or a composite material (Al-SiC) containing silicon carbide and aluminum. However, the materials for the first heat spreader 2 and the second heat spreader 5 are not limited to these.

[0021] The first power semiconductor element 3 is bonded to the first surface 2a of the first heat spreader 2, and the second power semiconductor element 6 is bonded to the first surface 5a of the second heat spreader 5. Power semiconductor elements include, for example, IGBTs (Insulated Gate Bipolar Transistors), FWDs (Free Wheel Diodes), and MOSFETs (Metal Oxide Semiconductor Field Effect Transistors). IGBTs and MOSFETs generally have three electrodes: gate, source, and drain (or gate, emitter, and collector). Power semiconductor elements control the large current flowing between the source (emitter) and drain (collector) by changing the gate voltage, which is the voltage applied to the gate. In the diagram, the gate, which is the control electrode, and the control wiring connected to the gate are omitted.

[0022] In this embodiment, as shown in Figure 2, the shape of the power semiconductor element is rectangular when viewed perpendicular to one surface 101a, with one first power semiconductor element 3 arranged on the first heat spreader 2 and one second power semiconductor element 6 arranged on the second heat spreader 5. The type, shape, and number of power semiconductor elements are not limited to these, and the shape of the power semiconductor element may be other than rectangular, and multiple power semiconductor elements may be arranged on each heat spreader.

[0023] As shown in Figure 3, the first lead frame 1 has one end joined to the first surface 2a of the first heat spreader, with the other end exposed from the sealing resin 12. The second lead frame 4 has one end joined to the surface electrode of one surface of the first power semiconductor element 3, with the other end joined to the first surface 5a of the second heat spreader 5. The third lead frame 7 has one end joined to the surface electrode of one surface of the second power semiconductor element 6, with the other end exposed from the sealing resin 12. A metal with high electrical conductivity is used as the material for the lead frame. Suitable materials for the lead frame include, for example, copper (Cu), aluminum (Al), or an alloy containing at least one of these metals. However, the material for the lead frame is not limited to these.

[0024] The first insulating layer 8 is located on the second surface 2b of the first heat spreader 2, and the second insulating layer 10 is located on the second surface 5b of the second heat spreader 5. The shape of the insulating layer is rectangular when viewed perpendicular to one surface 101a. The shape of the insulating layer is not limited to this. As shown in Figure 3, the dimensions in the Z direction of the first insulating layer 8 and the second insulating layer 10 are equal, and one surface of each is in the same plane. The heights of the first insulating layer 8 and the second insulating layer 10 are not limited to these, and their heights may be different. The material of the insulating layer is, for example, epoxy resin, urethane resin, silicone resin, polyimide resin, polyamide resin, polyamide-imide resin, acrylic resin, rubber material, or ceramic. The material of the insulating layer is not limited to these, and other materials may be used.

[0025] The first conductive layer 9 is located on the other side of the first insulating layer 8, and the second conductive layer 11 is located on the other side of the second insulating layer 10. The shape of the conductive layer is rectangular when viewed perpendicular to one side 101a. The shape of the conductive layer is not limited to this. As shown in Figure 3, the dimensions of the first conductive layer 9 and the second conductive layer 11 in the Z direction are equal, and one side of each and the other side are in the same plane. The heights of the first conductive layer 9 and the second conductive layer 11 are not limited to this, and their heights may be different. A metal with high electrical conductivity is used as the material for the conductive layer. Suitable materials for the conductive layer include, for example, copper (Cu), aluminum (Al), or alloys containing at least one of these metals. However, the material for the conductive layer is not limited to these.

[0026] The bonding between the heat spreader and the power semiconductor element, the bonding between the heat spreader and the lead frame, and the bonding between the power semiconductor element and the lead frame are performed via a bonding material (the bonding material is not shown). Examples of bonding materials include high-temperature solder containing lead (Pb) and tin (Sn), silver (Ag) nanoparticle paste, or conductive adhesive containing silver particles and epoxy resin. However, the type of bonding material is not limited to these.

[0027] The sealing resin 12 seals the heat spreader, power semiconductor elements, insulating layer, conductive layer, and lead frames, leaving a portion of the first lead frame 1 and the third lead frame 7 exposed. The material of the sealing resin 12 is, for example, epoxy resin, urethane resin, silicone resin, polyimide resin, polyamide resin, polyamide-imide resin, acrylic resin, or rubber. The sealing resin 12 may also be composed of multiple resin materials, for example, by layering gel-like silicone resin and epoxy resin. However, the type of sealing resin 12 is not limited to these. The external shape of the semiconductor power module 101 sealed by the sealing resin 12 is rectangular parallelepiped, but the external shape of the semiconductor power module 101 is not limited to rectangular parallelepiped. In this embodiment, the other surface of the conductive layer is exposed from the sealing resin 12, and the other surface of the conductive layer is joined to a cooler (not shown). This configuration allows for efficient cooling of the semiconductor power module 101.

[0028] Next, the potentials of each component of the semiconductor power module 101 will be explained using Figure 3. The first heat spreader 2 has a different potential from the second heat spreader 5. For example, the first heat spreader 2 has a P-side potential, and the second heat spreader 5 has an AC-side potential.

[0029] Each lead frame has a potential equal to that of the heat spreader or the electrode of the power semiconductor element to which it is joined. When the first heat spreader 2 has a P-side potential and the second heat spreader 5 has an AC-side potential, the first lead frame 1 joined to the first heat spreader 2 has a P-side potential, and the second lead frame 4 joined to the second heat spreader 5 has an AC-side potential. In addition, the third lead frame 7 joined to the electrode on one side of the second power semiconductor element 6 has an N-side potential. The conductive layer has a ground potential.

[0030] In other words, there is a potential difference between the heat spreader and the conductive layer, and between the second lead frame 4 and the third lead frame 7. The heat spreader and the conductive layer, and the second lead frame 4 and the third lead frame 7 are two adjacent electrodes that overlap with a gap between them. In particular, due to the characteristics of the electric field, equipotential lines tend to concentrate at the corners of these electrodes, so the electric field becomes higher at the corners of these electrodes.

[0031] <Comparative Example> Prior to describing the essential features of the present invention, a comparative example will be described. Figure 4 is a plan view showing two electrodes arranged in overlapping positions in a semiconductor power module of the comparative example, Figure 5 is a diagram showing the normalized electric field value at the corner of the electrode of the comparative example, and Figure 6 is an equipotential diagram of the area around the corner of the electrode of the semiconductor power module of the comparative example, which is an equipotential diagram of the electrode corner when cut at the BB cross-section position in Figure 4. In Figure 4, the upper electrode shown by the solid line and the lower electrode shown by the dashed line are arranged in overlapping positions with a gap between them. The upper electrode and the lower electrode have different potentials, with the upper electrode on the high-voltage side and the lower electrode on the low-voltage side. When viewed perpendicular to the electrode surface, the lower electrode is positioned inward from the upper electrode, the corners of the upper electrode and the lower electrode do not overlap, and the corner of the upper electrode protrudes outward from the corner of the lower electrode.

[0032] The horizontal axis in Figure 5 represents the normalized distance between the lower and upper electrodes, where 0 indicates that they overlap. The vertical axis in Figure 5 represents the electric field at the corner of the lower electrode (indicated by the star in Figure 4), normalized to 1 when the corners of the lower and upper electrodes overlap. From Figure 5, it can be seen that when the lower electrode is moved inside the upper electrode, the electric field at the corner of the lower electrode is higher than when the corners are aligned. From Figure 6, it can be seen that when one electrode is moved inside the other electrode, equipotential lines concentrate at the end of the inner electrode, and the electric field at the end of the inner electrode becomes higher. In other words, when the corner of one electrode overlaps with the other electrode when viewed perpendicular to the electrode surface, the electric field around the corner of the one electrode becomes higher, and the insulation performance decreases at the point where the electric field is higher. Therefore, miniaturization and high voltage of semiconductor power modules become difficult.

[0033] <Electric field relaxation structure at the corner of the electrode> The configuration of the arrangement of the corners of two adjacent electrodes will be described as the electric field relaxation structure for the corners of the electrodes, which is the essential part of this application. Figure 7 is a plan view showing the corners of two adjacent electrodes of the semiconductor power module 101 according to Embodiment 1, and is an enlarged view of the area around the corner 2c of the first heat spreader 2 shown in Figure 1. Figure 8 is a diagram showing the normalized electric field value of the electrode corners of the semiconductor power module 101. In Figure 7, only the first heat spreader 2 and the first conductive layer 9 are shown. The corner of the first conductive layer 9 shown in Figure 7 will be called corner 9c.

[0034] The semiconductor power module 101 comprises multiple electrodes, each having a different potential and overlapping with gaps in between. When viewed in the direction in which the multiple electrodes overlap, each of the multiple electrodes has three or more corners. In this embodiment, the multiple electrodes are a heat spreader, a conductive layer, and another heat spreader. Two adjacent electrodes are a heat spreader and a conductive layer, and a second lead frame 4 and a third lead frame 7. These electrodes are formed in a rectangular shape when viewed in the direction in which the multiple electrodes overlap. Therefore, when viewed in the direction in which the multiple electrodes overlap, each of these electrodes has four corners. The number of corners that each of the multiple electrodes has is not limited to four; it may be three or five or more.

[0035] When viewed in the direction in which multiple electrodes overlap, at least a portion of the corner of one electrode is positioned outside that of the other electrode, and at least a portion of the corner of the other electrode is positioned outside that of the first electrode. With this configuration, the corner of one electrode does not overlap with the other electrode when viewed perpendicular to the electrode surface, and the corner of the other electrode does not overlap with the first electrode when viewed perpendicular to the electrode surface. As a result, the electric field at and around the corners of both electrodes can be suppressed. Since the electric field at and around the corners of the electrodes is suppressed, the semiconductor power module can be made to a higher voltage. Furthermore, as in Patent Document 3, no additional insulator is required, so the manufacturing process does not increase and the probability of insulation defects such as voids does not increase, thus the semiconductor power module can be made smaller and to a higher voltage. The details of the electric field relaxation structure in this embodiment will be described below.

[0036] Each of the multiple electrodes is formed in a plate shape, and as shown in Figure 7, when viewed in the direction in which the multiple electrodes overlap, the edges on both sides of the corners of each of two adjacent electrodes overlap. In each of two adjacent electrodes, the shape of one corner is different from the shape of the other corner, such that the first part of one corner is further out than the first part of the other corner, and the second part of the other corner is further out than the second part of the first corner. In Figure 7, the first part 2c1 of one corner, corner 2c, is further out than the first part 9c1 of the other corner, corner 9c, and the second part 9c2 of corner 9c is further out than the second part 2c2 of corner 2c.

[0037] With this configuration, when viewed in the direction in which multiple electrodes overlap, the edges on both sides of the corners of each of two adjacent electrodes overlap. Therefore, when molding the semiconductor power module 101 with the sealing resin 12, the electric field around the corners of the electrodes can be suppressed without using a complex jig to fix the electrodes. Since a complex jig is not required, the productivity of the semiconductor power module 101 can be improved.

[0038] In this embodiment shown in Figure 7, the shapes of the corners 2c and 9c of each of the two adjacent electrodes are elliptical arcs. In each of the corners 2c and 9c, the first distance from the curvature start point to the curvature center point and the second distance from the curvature end point to the curvature center point are different. Of the first and second distances, the longer distance is called the major axis, and the shorter distance is called the minor axis. In Figure 7, one of the overlapping sides is called the first side 13, and the other overlapping side is called the second side 14. The major axis AC of corner 2c is parallel to the second side 14, and the minor axis BC of corner 2c is parallel to the first side 13. The major axis B'C' of corner 9c is parallel to the first side 13, and the minor axis A'C' of corner 9c is parallel to the second side 14. The major axes of each of the two adjacent electrodes intersect when viewed in the direction in which the multiple electrodes overlap.

[0039] This configuration makes it easy to create a shape in which the first portion 2c1 of one corner 2c is outside the first portion 9c1 of the other corner 9c, and the second portion 9c2 of corner 9c is outside the second portion 2c2 of corner 2c. Since it is easy to create electrode shapes that suppress the electric field at the corners and around the corners of both electrodes, the productivity of the semiconductor power module 101 can be improved. Furthermore, when viewed in the direction in which multiple electrodes overlap, the edges on both sides of the corners of each of two adjacent electrodes overlap, so when molding the semiconductor power module 101 with the sealing resin 12, the electric field around the corners of the electrodes can be suppressed without using a jig with a complex structure to fix the electrodes.

[0040] The horizontal axis in Figure 8 represents the ratio of the semi-major axis to the semi-minor axis. When the horizontal axis is 1, the shape of corners 2c and 9c is an arc shape. The vertical axis in Figure 8 represents the electric field when corners 2c and 9c overlap, that is, the electric field of corner 2c normalized to 1 when the shape of corners 2c and 9c is an arc shape. The electric field of corner 9c shows a similar trend to Figure 8. From Figure 8, it can be seen that if the ratio of the semi-major axis to the semi-minor axis is greater than 1 and 3 or less, the electric field of corners 2c and 9c will be smaller than the electric field when corners 2c and 9c overlap. Therefore, it is desirable that the ratio of the semi-major axis to the semi-minor axis for each of the two adjacent electrodes is 3 or less. By configuring it in this way, it is possible to reliably make the electric field of corners 2c and 9c smaller than the electric field when corners 2c and 9c overlap.

[0041] Furthermore, in other corners not shown in Figure 7, the shapes of the corners of adjacent electrodes are similarly different. Also, regarding the corners of the electrode pairs of the second lead frame 4 and the third lead frame 7, the particularly sharp part of the corner of one electrode can be made to form a corner shape such that, when viewed perpendicular to the electrode surface, it does not overlap with the other electrode. In other words, the heat spreader and the conductive layer, and one or both of the second lead frame 4 and the third lead frame 7, are two adjacent electrodes. With this configuration, the electric field at the electrode corners and around the corners can be suppressed even between the second lead frame 4 and the third lead frame 7.

[0042] As described above, the semiconductor power module 101 according to Embodiment 1 is equipped with a plurality of electrodes, each having a different potential and overlapping with a gap in between. When viewed in the direction in which the plurality of electrodes overlap, each of the plurality of electrodes has three or more corners. In two adjacent electrodes, at least a portion of the corner of one electrode is located outside the other electrode, and at least a portion of the corner of the other electrode is located outside the first electrode. Therefore, when viewed in the direction in which the two adjacent electrodes overlap, the corner of one electrode does not overlap with the other electrode, and when viewed in the direction in which the two adjacent electrodes overlap, the corner of the other electrode does not overlap with the first electrode. As a result, the electric field at the corners and around the corners of the electrodes can be suppressed in both electrodes. Since the electric field at the corners and around the corners of the electrodes is suppressed, the semiconductor power module 101 can be made to operate at a higher voltage. Furthermore, as in Patent Document 3, no additional insulator is required, so the manufacturing process does not increase and the probability of insulation defects such as voids does not increase, thus the semiconductor power module 101 can be made smaller and to operate at a higher voltage. Furthermore, because an electric field mitigation effect can be expected, the deburring process, which is one of the causes of high electric fields, can be reduced, and the cycle time of the semiconductor power module 101 can be shortened. At the same time, impedance losses associated with the high-frequency operation voltage of the semiconductor power module 101 can be reduced.

[0043] Each of the multiple electrodes is formed in a plate shape, and when viewed in the direction in which the multiple electrodes overlap, the edges on both sides of the corners of each of two adjacent electrodes overlap, and the shapes of the corners of each of the two adjacent electrodes are different such that the first part of one corner is further out than the first part of the other corner, and the second part of the other corner is further out than the second part of the first corner. Because the edges on both sides of the corners of each of the two adjacent electrodes overlap when viewed in the direction in which the multiple electrodes overlap, the electric field around the corners of the electrodes can be suppressed during the molding of the semiconductor power module 101 with the sealing resin 12 without using a jig with a complex structure to fix the electrodes. Since a jig with a complex structure is not required, the productivity of the semiconductor power module 101 can be improved.

[0044] The shape of the corners of each of two adjacent electrodes is an elliptical arc, and the first distance from the curvature start point to the curvature center point of the elliptical arc shape and the second distance from the curvature end point to the curvature center point are different. The longer of the two distances is called the semi-major axis, and the shorter of the two distances is called the semi-minor axis. When the semi-major axes of the two adjacent electrodes intersect when viewed in the direction in which multiple electrodes overlap, the first portion 2c1 of one corner 2c is outside the first portion 9c1 of the other corner 9c, and the second portion 9c2 of corner 9c is outside the second portion 2c2 of corner 2c. This shape makes it easy to fabricate electrodes that suppress the electric field around the corners of both electrodes, thereby improving the productivity of the semiconductor power module 101. Furthermore, when viewed in the direction in which multiple electrodes overlap, the edges on both sides of the corners of each of two adjacent electrodes overlap. Therefore, when molding the semiconductor power module 101 with the sealing resin 12, the electric field around the corners of the electrodes can be suppressed without using a complex jig to fix the electrodes.

[0045] If the ratio of the major axis to the minor axis in each of two adjacent electrodes is 3 or less, the electric field at corners 2c and 9c can be reliably made smaller than the electric field when corners 2c and 9c overlap. Furthermore, in the semiconductor power module 101, if the heat spreader and the conductive layer, and one or both of the second lead frame 4 and the third lead frame 7 are two adjacent electrodes with corners arranged as described above, the electric field around the corners of the heat spreader and the conductive layer, and one or both of the second lead frame 4 and the third lead frame 7 can be suppressed.

[0046] Embodiment 2. A semiconductor power module 101 according to Embodiment 2 will be described. Figure 9 is a plan view showing two adjacent electrodes of the semiconductor power module 101 according to Embodiment 2, shown with the first heat spreader 2 and the first conductive layer 9 shown in Figure 1 transparent. Figure 10 is a plan view showing the corners of two adjacent electrodes of the semiconductor power module 101, shown with the corner 2c of the first heat spreader 2 and the corner 9c of the first conductive layer 9 transparent. Figure 11 is a diagram illustrating the movement of the corners of two adjacent electrodes of the semiconductor power module. Figure 12 is a diagram showing the normalized electric field value of the corner of the electrode of the semiconductor power module 101. Figure 13 is an equipotential diagram around the electrode at the CC cross-sectional position in Figure 11. Figure 14 is a diagram showing an approximate curve of the normalized electric field value of the corner of the electrode of the semiconductor power module 101. In the semiconductor power module 101 according to Embodiment 2, the corners of two adjacent electrodes are configured to have the same shape.

[0047] Each of the multiple electrodes is formed in a plate shape, and the shape of its corners is either a circular arc or an elliptical arc, and they are identical in shape. For each of two adjacent electrodes, the intersection of the extensions of the sides of the corner of one electrode is defined as the first intersection, and the intersection of the extensions of the sides of the corner of the other electrode is defined as the second intersection, and the distance between the adjacent first and second intersections is defined as the intersection distance. The intersection distance has a value greater than 0. The relationship between the electric field at the corner of each of the two adjacent electrodes and the intersection distance is expressed as an exponential or logarithmic function.

[0048] The specific configuration of this embodiment will be explained using Figures 9 and 10. In these figures, the multiple electrodes are the first heat spreader 2 and the first conductive layer 9. The corners 2c of the first heat spreader 2 and the corners 9c of the first conductive layer 9 are arc-shaped and identical in shape. As shown in Figure 9, two opposing sides 2d1 and 2d2 of the first heat spreader 2 are made longer than two opposing sides 9d1 and 9d2 of the first conductive layer 9, while two opposing sides 2d3 and 2d4 of the other heat spreader 2 are made shorter than two opposing sides 9d3 and 9d4 of the first conductive layer 9. With this configuration, when viewed in the direction in which the first heat spreader 2 and the first conductive layer 9 overlap, at least a portion of the corners of the first heat spreader 2 is located outside the first conductive layer 9, and at least a portion of the corners of the first conductive layer 9 is located outside the first heat spreader 2.

[0049] The first intersection point D1 is defined as the intersection of the extensions of the sides of corner 2c of the first heat spreader 2, and the second intersection point D2 is defined as the intersection of the extensions of the sides of corner 9c of the first conductive layer 9. The distance between the intersection point E, which is the distance between the first intersection point D1 and the second intersection point D2, is greater than 0. The relationship between the electric fields at corners 2c and 9c and the distance between the intersection point E can be expressed as an exponential or logarithmic function.

[0050] With this configuration, the distance between the intersection points is greater than 0, and the relationship between the electric field at the corners of each of the two adjacent electrodes and the distance E between the intersection points is expressed as an exponential or logarithmic function. As a result, the corner of one electrode does not overlap with the other electrode when viewed in the direction in which the two adjacent electrodes overlap, and the corner of the other electrode does not overlap with the first electrode when viewed in the direction in which the two adjacent electrodes overlap. Therefore, the electric field at the corners and around the corners of both electrodes can be suppressed. Furthermore, the relationship between the electric field at the corners of the electrodes and the distance E between the intersection points can be easily adjusted using the exponential or logarithmic function that expresses the relationship between the electric field at the corners of the electrodes and the distance E between the intersection points. In addition, since the shape of the corners is the same, either a circular arc or an elliptical arc, the manufacturing process of the electrodes is simplified, and the cycle time of the semiconductor power module 101 can be shortened. The details of the electric field relaxation structure in this embodiment will be described below.

[0051] The horizontal axis in Figure 12 represents the normalized value obtained by dividing the distance between the intersection points of the upper and lower electrodes in Figure 11 by the radius of curvature or semi-major axis of the corner shape. When the horizontal axis is 0, both the corners of the upper and lower electrodes overlap. In Figure 11, since the upper and lower electrodes are arc-shaped, the distance between the intersection points is normalized by dividing it by the radius of curvature. The vertical axis in Figure 12 represents the electric field of the corner portion of the upper electrode (the part indicated by the star in Figure 11), normalized to 1 when the electric field when the corners of the lower and upper electrodes overlap. The electric field of the lower electrode shows a similar trend to Figure 12. From Figure 12, it can be seen that as the distance between the intersection points increases, the electric field of each electrode decreases. This is because, as shown in Figure 13, the equipotential lines do not converge when the corner of one electrode does not overlap with the other electrode.

[0052] The exponential and logarithmic functions described above will now be explained. Let y be the value obtained by dividing the electric field at the corner of one electrode or the other electrode by the overlapping electric field, which is the electric field at the corner where the corners of both electrodes overlap, viewed in the direction in which the two electrodes overlap. If the shape of the corner is a circular arc, let x be the value obtained by dividing the distance between the intersection points by the radius of curvature, or if the shape of the corner is an elliptical arc, let x be the value obtained by dividing the distance between the intersection points by the semi-major axis. In this case, the exponential or logarithmic function described above is: y=(1-0.4)×exp(-x / 5)+0.4 ···(1) Or, x=5×ln{(1-0.4) / (y-0.4)} ···(2) That is the case.

[0053] Figure 14 shows the approximate curve and approximation formula for the normalized electric field value in Figure 12. In the approximation formula, x corresponds to the value on the horizontal axis in Figure 12. In the approximation formula, y corresponds to the value on the vertical axis in Figure 12. By expressing exponential and logarithmic functions with the approximation formulas shown in equation (1) or (2) in this way, the distance between intersections can be easily determined to satisfy the above approximation formula for a given desired electric field. Since the distance between intersections can be easily determined, the electric field generated at the corners can be easily made into the desired electric field.

[0054] Furthermore, in other corners not shown in Figure 10, the distance between intersections is similarly greater than 0. Also, the distance between intersections can be greater than 0 at the corners of the electrode pairs of the second lead frame 4 and the third lead frame 7. In other words, the heat spreader and the conductive layer, and one or both of the second lead frame 4 and the third lead frame 7, are two adjacent electrodes. With this configuration, the electric field generated at the corners between the second lead frame 4 and the third lead frame 7 can easily be set to the desired electric field.

[0055] The configuration of two adjacent electrodes in a semiconductor power module 101 having an intersection distance greater than 0 is not limited to the configuration shown in Figure 9, but may also be the configuration shown in Figure 15. Figure 15 is a plan view showing two adjacent electrodes in another semiconductor power module 101 according to Embodiment 2, and is shown with the first heat spreader 2 and the first conductive layer 9 transparent. The first heat spreader 2 and the first conductive layer 9 have the same shape when viewed in the direction in which they overlap. When viewed in the direction in which the first heat spreader 2 and the first conductive layer 9 overlap, Figure 15 shows a configuration in which one of the electrodes is rotated from a state in which the corners of the first heat spreader 2 and the first conductive layer 9 overlap so that the corners of both do not overlap. Even with this arrangement, it is possible to create a configuration in which the corner of one electrode does not overlap with the other electrode when viewed in the direction in which the two adjacent electrodes overlap, and the corner of the other electrode does not overlap with the first electrode when viewed in the direction in which the two adjacent electrodes overlap.

[0056] As described above, each of the multiple electrodes is formed in a plate shape, and the shape of the corners is the same, either a circular arc shape or an elliptical arc shape. For each of two adjacent electrodes, the intersection of the extensions of the sides on both sides of the corner of one electrode is defined as the first intersection, and the intersection of the extensions of the sides on both sides of the corner of the other electrode is defined as the second intersection. The distance between the adjacent first and second intersections is defined as the intersection distance, and the intersection distance is greater than 0. The relationship between the electric field at the corner of each of the two adjacent electrodes and the intersection distance is expressed as an exponential or logarithmic function. Therefore, the corner of one electrode does not overlap with the other electrode when viewed in the direction in which the two adjacent electrodes overlap, and the corner of the other electrode does not overlap with the first electrode when viewed in the direction in which the two adjacent electrodes overlap. Thus, the electric field at the corners and around the corners of both electrodes can be suppressed. Furthermore, the relationship between the electric field at the corner of the electrode and the distance E between the intersections can be easily adjusted using an exponential or logarithmic function that shows the relationship between the electric field at the corner of the electrode and the distance E between the intersections. In addition, since the shape of the corner is the same, either a circular arc or an elliptical arc, the manufacturing process of the electrode is simplified, and the cycle time of the semiconductor power module 101 can be shortened. Since the cycle time of the semiconductor power module 101 is shortened, the productivity of the semiconductor power module 101 can be improved.

[0057] If the electric field at the corner of one electrode or the other electrode is divided by the overlapping electric field (the electric field at the corner where the corners of both electrodes overlap, viewed in the direction in which the two electrodes overlap) and let y be that value, and x is the value obtained by dividing the distance between the intersection points by the radius of curvature if the shape of the corner is a circular arc, or by dividing the distance between the intersection points by the semi-major axis if the shape of the corner is an elliptical arc, and the exponential or logarithmic function described above is y = (1 - 0.4) × exp(-x / 5) + 0.4, or x = 5 × ln{(1 - 0.4) / (y - 0.4)}, then by expressing the exponential and logarithmic functions in this way as approximate formulas, the distance between the intersection points can be easily determined to satisfy the above approximate formula for a given desired electric field. Since the distance between the intersection points can be easily determined, the electric field generated at the corner can be easily made into the desired electric field.

[0058] Furthermore, although this application describes various exemplary embodiments and examples, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but can be applied individually or in various combinations to the embodiments. Accordingly, countless variations not illustrated are conceivable within the scope of the technology disclosed herein. These include, for example, modifications, additions, or omissions of at least one component, as well as the extraction of at least one component and its combination with components of other embodiments. [Explanation of Symbols]

[0059] 1 First lead frame, 2 First heat spreader, 2a First face, 2b Second face, 2c Corner, 2c1 First part, 2c2 Second part, 2d1, 2d2, 2d3, 2d4 Sides, 3 First power semiconductor element, 4 Second lead frame, 5 Second heat spreader, 5a First face, 5b Second face, 6 Second power semiconductor element, 7 Third lead frame, 8 First insulating layer, 9 First conductive layer, 9c Corner, 9c1 First part, 9c2 Second part, 9d1, 9d2, 9d3, 9d4 Sides, 10 Second insulating layer, 11 Second conductive layer, 12 Sealing resin, 13 First side, 14 Second side, 101 Semiconductor power module, 101a One face, 101b Other face, D1 First intersection, D2 Second intersection, E Intersection distance

Claims

1. It comprises multiple electrodes, each having a different potential and overlapping with gaps between them. When viewed in the direction in which the aforementioned multiple electrodes overlap, Each of the aforementioned plurality of electrodes has three or more corners, In the two adjacent electrodes, At least a portion of the corner of one of the electrodes is provided on the outside of the other electrode. At least a portion of the corner of the other electrode is provided on the outside of the first electrode. Each of the aforementioned plurality of electrodes is formed in a plate shape, When viewed in the direction in which the plurality of electrodes overlap, the edges on both sides of the corner of each of the two adjacent electrodes overlap. In each of the two adjacent electrodes, A semiconductor power module in which the shape of one corner and the shape of the other corner are different such that the first portion of one corner is located further out than the first portion of the other corner, and the second portion of the other corner is located further out than the second portion of the first corner.

2. The shape of the corner portion of each of the two adjacent electrodes is an elliptical arc shape, and the first distance from the curvature start point to the curvature center point of the elliptical arc shape and the second distance from the curvature end point to the curvature center point are different. Of the first distance and the second distance, the longer distance is defined as the semi-major axis, and the shorter distance is defined as the semi-minor axis. The semiconductor power module according to claim 1, wherein the semi-major radii of each of the two adjacent electrodes intersect when viewed in the direction in which the plurality of electrodes overlap.

3. The semiconductor power module according to claim 2, wherein the ratio of the major axis to the minor axis in each of the two adjacent electrodes is 3 or less.

4. A first heat spreader and a second heat spreader, which are plate-shaped and arranged side by side, A first power semiconductor element is formed in a plate shape and bonded to one side of the first heat spreader, and a second power semiconductor element is bonded to one side of the second heat spreader, A first conductive layer, which is formed in a plate shape and bonded to the other surface of the first heat spreader via an insulating layer, and a second conductive layer, which is bonded to the other surface of the second heat spreader via an insulating layer, The other side is joined to a first lead frame which is joined to one side of the first heat spreader, A second lead frame extends in the direction in which the heat spreaders are arranged, with one end of its other surface bonded to one surface of the first power semiconductor element and the other end of its other surface bonded to one surface of the second heat spreader. The third lead frame extends in the direction in which the heat spreaders are arranged, with one end of the other side bonded to one side of the second power semiconductor element, and at least a portion of the other side overlapping one side of the second lead frame with a gap between them, The semiconductor power module according to any one of claims 1 to 3, wherein the heat spreader and the conductive layer, and one or both of the second lead frame and the third lead frame are the two adjacent electrodes.

Images