Semiconductor module

Abstract

The semiconductor module of the present application includes: a semiconductor switching element; and a stress applying portion provided on one or both of a first surface and a second surface opposite to the first surface of the semiconductor switching element, having a linear expansion coefficient larger than that of a main material of the semiconductor switching element, and having a thickness thicker than that of the semiconductor switching element, the stress applying portion applying a compressive stress or a tensile stress to the semiconductor switching element by thermal contraction or thermal expansion of the stress applying portion accompanying a temperature change, and a threshold voltage at which the semiconductor switching element is turned on being lowered as the magnitude of the compressive stress or the tensile stress of the semiconductor switching element increases.

Description

Technical Field

[0001] This application relates to semiconductor modules. Background Technology

[0002] In the field of power electronics, power conversion devices such as AC / DC converters, DC / DC converters, and inverters are used. These power conversion devices consist of semiconductor modules that convert power using the switching action of semiconductor switching elements. Examples of semiconductor switching elements made of power semiconductors include MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors) and IGBTs (Insulated-Gate Bipolar Transistors). They typically have three electrodes: a gate, a source, and a drain (or a gate, emitter, and collector). Semiconductor switching elements control the large current flowing between the source (emitter) and drain (collector) by varying the voltage applied to the gate, i.e., the gate voltage.

[0003] In semiconductor switching devices, the gate voltage required to allow a predetermined current exceeding a certain value to flow between the source and drain (emitter and collector) is called the threshold voltage (Vth). Generally, the threshold voltage is negatively dependent on temperature. That is, as the temperature of the semiconductor switching device increases, the threshold voltage decreases. Conversely, if the temperature of the semiconductor switching device decreases, the threshold voltage increases. If the threshold voltage increases, not only does the on-resistance (channel resistance) of the semiconductor switching device increase, but the conduction speed during switching also increases, and the cutoff surge voltage also increases. If the steady-state loss increases due to the increased on-resistance, the heating of the semiconductor switching device cannot be suppressed. Furthermore, if the cutoff surge voltage increases, it may sometimes exceed the rated voltage of the semiconductor switching device. To suppress these situations, semiconductor switching device designs address these issues by increasing the current-carrying area of ​​the semiconductor switching device and setting a higher rated voltage for the semiconductor switching device.

[0004] Increasing the current-carrying area of ​​semiconductor switching elements increases the material cost per element, thus increasing the overall cost of the semiconductor module. Furthermore, the larger size of the encapsulated semiconductor switching elements leads to a larger overall semiconductor module size. Setting higher rated voltages for the semiconductor switching elements increases the manufacturing costs, such as forming thicker voltage-holding layers. Additionally, as the voltage-holding layer thickness increases, the on-resistance of the semiconductor switching element also increases, requiring a larger area to suppress heat generation. This further increases the cost of the semiconductor switching elements and the overall size of the semiconductor module. Therefore, the negative temperature characteristic of the threshold voltage of semiconductor switching elements contributes to both increased cost and larger size of the semiconductor module.

[0005] To address the aforementioned problems, a technique for configuring a switching circuit is disclosed, in which a resistor integrated into the gate drive circuit and affecting the gate drive speed is variable according to the temperature of each element (for example, see Patent Document 1). In the disclosed technique, by switching the resistor within the gate drive circuit, the increase in cutoff surge voltage is suppressed, thereby suppressing the increase in the rated voltage of the semiconductor switching element. Existing technical documents Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2002-199700 Summary of the Invention The technical problem that the invention aims to solve

[0007] In the structure of the semiconductor power conversion device in Patent Document 1, the resistance in the gate drive circuit is switched according to the temperature of the semiconductor switching element, thus suppressing the increase of the cutoff surge voltage. However, when switching the resistance in the gate drive circuit according to the temperature of the semiconductor switching element, it is necessary to monitor the temperature of the semiconductor switching element and control the resistance value of the gate drive circuit. Therefore, a large number of additional components are required, resulting in the following problems: the cost of the semiconductor module increases, and the semiconductor module becomes larger. Furthermore, the manufacturing cost of the semiconductor module increases in order to cope with the increased complexity of control and the increase in failure modes.

[0008] Therefore, the purpose of this application is to obtain a small and inexpensive semiconductor module that suppresses the increase in on-resistance and the increase in cutoff surge voltage at low temperatures. Technical means for solving technical problems

[0009] The semiconductor module disclosed in this application includes: a semiconductor switching element; and a stress application portion disposed on one or both of a first surface of the semiconductor switching element and a second surface opposite to the first surface. The stress application portion has a coefficient of linear expansion that is larger than that of the main material of the semiconductor switching element, and its thickness is greater than that of the semiconductor switching element. The stress application portion utilizes the thermal contraction or thermal expansion of the stress application portion accompanying temperature changes to generate compressive stress or tensile stress in the semiconductor switching element. As the magnitude of the compressive stress or tensile stress of the semiconductor switching element increases, the threshold voltage for the semiconductor switching element to conduct decreases. Invention Effects

[0010] According to the semiconductor module disclosed in this application, a stress-applying portion is provided on one or both of a first surface of a semiconductor switching element and a second surface opposite to the first surface. The stress-applying portion has a coefficient of linear expansion that is larger than that of the main material of the semiconductor switching element, and its thickness is greater than that of the semiconductor switching element. The stress-applying portion utilizes the thermal contraction or thermal expansion of the stress-applying portion with temperature changes to generate compressive or tensile stress in the semiconductor switching element. As the magnitude of the compressive or tensile stress of the semiconductor switching element increases, the threshold voltage for the semiconductor switching element to turn on decreases. Therefore, the increase in on-resistance and the increase in cutoff surge voltage can be suppressed. Thus, the area and rated voltage of the semiconductor switching element can be reduced, and a small and inexpensive semiconductor module can be obtained. Attached Figure Description

[0011] Figure 1 This is a top view showing an outline of the appearance of the semiconductor module involved in Embodiment 1. Figure 2 It is shown in Figure 1 A schematic cross-sectional view of the semiconductor module cut off at section AA. Figure 3 This is a top view showing the main parts of the semiconductor module involved in Embodiment 1. Figure 4 This is another top view showing the main parts of the semiconductor module involved in Embodiment 1. Figure 5 This is a diagram illustrating an example of the temperature characteristics of the threshold voltage of the semiconductor switching element of the semiconductor module according to Embodiment 1. Figure 6 This is a diagram illustrating an example of the voltage waveform applied to the semiconductor switching element of the semiconductor module involved in Embodiment 1. Figure 7 This is a diagram illustrating the compressive stress generated in the semiconductor switching element of the semiconductor module according to Embodiment 1. Figure 8 This is a top view showing the main parts of the other semiconductor modules involved in Embodiment 1. Figure 9 This is a top view showing the main parts of the other semiconductor modules involved in Embodiment 1. Figure 10 This is a cross-sectional view showing an outline of the semiconductor module involved in Embodiment 2. Figure 11 This is a diagram of the equivalent circuit obtained when the semiconductor switching element of the semiconductor module involved in the simulation embodiment 2 is turned off. Figure 12 This is a graph showing the correlation between the decrease in threshold voltage and the increase in off-time of the semiconductor switching element of the semiconductor module according to Embodiment 2. Figure 13 This is a graph showing the correlation between the deformation generated on the surface of the semiconductor switching element of the semiconductor module according to Embodiment 2 and the amount of change in the threshold voltage. Figure 14 This is a graph showing the correlation between the thickness of the stress-applying portion of the semiconductor module according to Embodiment 2 and the deformation generated on the surface of the semiconductor switching element. Figure 15 This is a side view showing an outline of other semiconductor modules involved in Embodiment 2. Figure 16 This is a top view showing the main parts of the other semiconductor modules involved in Embodiment 2. Figure 17 This is a top view showing the main parts of the other semiconductor modules involved in Embodiment 2. Figure 18 This is a cross-sectional view showing an outline of the semiconductor module involved in Embodiment 3. Detailed Implementation

[0012] Hereinafter, the semiconductor module involved in the embodiments of this application will be described based on the accompanying drawings. Furthermore, identical or equivalent components and parts will be labeled with the same reference numerals in the drawings.

[0013] Implementation method 1. Figure 1 This is a top view showing an outline of the appearance of the semiconductor module 100 according to Embodiment 1. Figure 2 It is shown in Figure 1 A schematic cross-sectional view of the semiconductor module 100 cut off at section AA. Figure 3 This is a top view showing the main parts of the semiconductor module 100. It is a diagram showing the semiconductor switching element 1, the stress application part 10a, and the busbar 5 (dashed line) on the first surface 1a side after the insulating resin material 6 has been removed. Figure 4 This is another top view showing the main parts of the semiconductor module 100, and a diagram showing the semiconductor switching element 1, stress application portion 10b, and stress application portion 10a (dashed line) on the second surface 1b side after the insulating resin material 6 has been removed. Figure 5 This is a diagram illustrating an example of the temperature characteristics of the threshold voltage of the semiconductor switching element 1 of the semiconductor module 100. Figure 6 This is a diagram illustrating an example of the voltage waveform applied to the semiconductor switching element 1 of the semiconductor module 100. Figure 7 This diagram schematically illustrates the compressive stress generated in the semiconductor switching element 1 of the semiconductor module 100. The semiconductor module 100 for power conversion is primarily mounted in power conversion devices that convert desired power into DC or AC voltage. The semiconductor module 100 internally includes semiconductor switching elements 1 such as MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) and IGBTs (Insulated Gate Bipolar Transistors). The semiconductor module 100 converts power through the switching action of the semiconductor switching elements 1.

[0014] <Semiconductor Module 100> like Figure 2 As shown, the semiconductor module 100 includes a semiconductor switching element 1, stress application parts 10a and 10b, a heat dissipation component 2, busbars 4 and 5, and an insulating resin material 6 covering them. Figure 2 In the diagram, the outline of the insulating resin material 6 is represented by dashed lines. A stress-applying part 10a is disposed between the semiconductor switching element 1 and the busbar 5, and a stress-applying part 10b is disposed between the semiconductor switching element 1 and the heat dissipation member 2. (As shown...) Figure 1 As shown, a portion of busbars 4 and 5 is exposed to the outside from the insulating resin material 6. The exposed portions of busbars 4 and 5 are connected to external devices. Busbars 4 and 5 are made, for example, of copper, which has high conductivity.

[0015] In this embodiment, in the semiconductor switching element 1 composed of power semiconductors such as MOSFETs or IGBTs, the side where the channel region is formed and the source electrode (emitter electrode) is located is referred to as the surface of the semiconductor switching element 1, i.e., the first surface 1a, and the surface where the drain electrode (collector electrode) is formed is referred to as the back surface, i.e., the second surface 1b. Hereinafter, unless otherwise stated, in the cross-sectional view of the semiconductor module 100, the first surface 1a of the semiconductor switching element 1 is shown facing upwards, and the second surface 1b is shown facing downwards. The main material of the semiconductor switching element 1 is a semiconductor material such as silicon, silicon carbide, or gallium oxide. In the semiconductor switching element 1, a wide-bandgap semiconductor, i.e., a compound semiconductor, with a wider bandgap than silicon, can be used. Due to its material properties, compound semiconductors are superior as devices for power conversion devices and are harder than silicon. Furthermore, the semiconductor switching element 1 composed of silicon carbide and gallium oxide is generally formed with a smaller thickness than the semiconductor switching element 1 composed of silicon.

[0016] The busbar 5 is connected to the portion of the stress application portion 10a disposed on the first surface 1a that is opposite to the side of the semiconductor switching element 1. Figure 3 In the diagram, the outline of busbar 5 is shown by dashed lines. Busbar 5 is connected to stress application portion 10a, for example, via solder. Heat dissipation member 2 is connected to the portion of stress application portion 10b provided on the second surface 1b opposite to the side of semiconductor switching element 1. Heat dissipation member 2 functions as an electrode connected to the second surface 1b, and busbar 4 is connected to heat dissipation member 2. Heat dissipation member 2 is formed of copper in a rectangular plate shape, for example, but is not limited to this and may also be in a block shape. Heat dissipation member 2 is thermally and electrically connected to semiconductor switching element 1 and releases the heat generated by semiconductor switching element 1 to the outside. In addition, heat dissipation member 2 releases the heat generated in busbar 5 via semiconductor switching element 1 to the outside. Heat dissipation member 2 is connected to stress application portion 10b, for example, via solder.

[0017] <Stress application parts 10a, 10b> The stress application portion is disposed on one or both of the first surface 1a and the second surface 1b opposite to the first surface 1a of the semiconductor switching element 1. In this embodiment, as shown... Figure 3 As shown, two stress-applying portions 10a are provided on the first surface 1a. Furthermore, as... Figure 4 As shown, one stress-applying part 10b is disposed on the second surface 1b. Two stress-applying parts 10a disposed on the first surface 1a are in a first direction parallel to the first surface 1a. Figure 3The stress-applying portion 10b, as shown by arrow B, is arranged at intervals. Viewed from a direction perpendicular to the first surface 1a, a stress-applying portion 10b disposed on the second surface 1b is disposed between the end 10a1 of the stress-applying portion 10a disposed on one side of the first direction on the first surface 1a and the end 10a2 of the stress-applying portion 10a disposed on the other side of the first direction on the first surface 1a.

[0018] The stress-applying portions 10a and 10b have a coefficient of linear expansion that is larger than that of the main material of the semiconductor switching element 1. The thickness of the stress-applying portions 10a and 10b is greater than that of the semiconductor switching element 1. The stress-applying portions 10a and 10b are made, for example, of copper with a thickness of 1 mm or more. The stress-applying portions 10a and 10b are not limited to copper and can also be made of other materials such as aluminum. When the stress-applying portions 10a and 10b are made of copper, they undergo thermal contraction as the temperature decreases. The stress-applying portion 10a is connected to the busbar 5, and the stress-applying portion 10b is connected to the heat dissipation member 2, which functions as an electrode. Therefore, the stress-applying portions 10a and 10b are preferably made of materials with excellent electrical and thermal conductivity. However, the stress-applying portions 10a and 10b are not limited to materials with excellent electrical and thermal conductivity; they can also be configured such that the busbar 5 and the heat dissipation member 2 are connected to the semiconductor switching element 1 without passing through the stress-applying portions 10a and 10b.

[0019] By configuring it in this way, due to the thermal contraction or expansion of the stress application parts 10a and 10b accompanying temperature changes, the stress application parts 10a and 10b induce compressive or tensile stress in the semiconductor switching element 1. Figure 7 As shown, when the stress application portions 10a and 10b undergo thermal contraction, compressive stress is generated on the first surface 1a side of the semiconductor switching element 1 in the directions indicated by the two arrows C. When the stress application portions 10a and 10b undergo thermal expansion, tensile stress is generated on the first surface 1a side of the semiconductor switching element 1.

[0020] Semiconductor switching element 1 is bonded to stress application portions 10a and 10b via a bonding layer (not shown) to prevent slippage. When the semiconductor switching element 1 is connected to the stress application portions 10a and 10b via the bonding layer, stress can be efficiently applied to the semiconductor switching element 1 from the stress application portions 10a and 10b. Furthermore, since peeling between the semiconductor switching element 1 and the stress application portions 10a and 10b is suppressed, the thermal reliability between the semiconductor switching element 1 and the stress application portions 10a and 10b can be improved. The bonding layer is formed using metal particles, such as silver or copper particles, and is made of a material harder than solder. When metal particles are used in the bonding layer, the semiconductor switching element 1 is bonded to the stress application portions 10a and 10b by heating and pressurizing, or both heating and pressurizing. Using a harder material like metal particles increases the stress applied to the semiconductor switching element 1 from the stress application portions 10a and 10b compared to not using a harder material.

[0021] The semiconductor switching element 1 and the stress application portions 10a and 10b are not limited to being joined by a bonding layer. The semiconductor switching element 1 and the stress application portions 10a and 10b may also be configured to abut against each other. For example, when the semiconductor switching element 1 and the stress application portions 10a and 10b abut against each other by compression, the stress application portions 10a and 10b have the function of applying stress to the semiconductor switching element 1 and joining the semiconductor switching element 1 to the heat dissipation member 2. When the semiconductor switching element 1 and the stress application portions 10a and 10b abut against each other, a bonding layer is not required, thus improving the productivity of the semiconductor module 100.

[0022] The effects of compressive or tensile stress on the semiconductor switching element 1 are explained. In the semiconductor switching element 1, the threshold voltage (Vth), which requires a predetermined current of a certain value or higher to flow between the source and drain (emitter and collector), exhibits a negative dependence on temperature. For example... Figure 5 As shown by the solid line, the threshold voltage of semiconductor switching element 1 increases as the temperature decreases. If the threshold voltage increases, the on-resistance (channel resistance) of semiconductor switching element 1 increases, and the cutoff surge voltage also increases. Therefore, the voltage waveform applied to semiconductor switching element 1 as the temperature decreases becomes... Figure 6 The waveform is as shown by the solid line.

[0023] The semiconductor switching element 1 has the following characteristic: as the magnitude of the compressive or tensile stress applied to the semiconductor switching element 1 increases, the threshold voltage at which the semiconductor switching element 1 turns on decreases. When using stress application portions 10a and 10b made of copper, as the temperature of the semiconductor module 100 decreases, the thermal contraction of the stress application portions 10a and 10b becomes more significant, and the compressive stress generated in the semiconductor switching element 1 also becomes stronger. Therefore, as the temperature of the semiconductor module 100 decreases, the effect of reducing the threshold voltage of the semiconductor switching element 1 becomes greater.

[0024] Figure 5 In the middle, it is indicated by a dashed line. Figure 5 The solid line shows an example of the temperature characteristic of the threshold voltage of semiconductor switching element 1, after adding the characteristic of threshold voltage reduction under stress. The dotted-dash line shows a flatter slope for the temperature characteristic of the threshold voltage compared to the existing solid line without stress. Figure 5 One point to note is that when stress is applied... Figure 5 The solid line does not move parallel; instead, it starts from the threshold voltage at temperature T2, and the slope of the temperature characteristic decreases. Therefore, it can maintain the threshold voltage required at high temperatures and suppress the increase of the threshold voltage at low temperatures.

[0025] exist Figure 6 In addition to being applied to Figure 6 In addition to the solid line showing the voltage waveform of semiconductor switching element 1 as the temperature decreases, a dashed line also shows an example of the voltage waveform at a stress-applied temperature T1. The increase in the threshold voltage at low temperatures is suppressed, therefore, as... Figure 6 As shown by the dotted line, at time t1, the increase in cutoff surge voltage and cutoff surge velocity caused by the increase in the threshold voltage of semiconductor switching element 1 at low temperature can be suppressed. Furthermore, since the increase in threshold voltage can be suppressed, the increase in on-resistance can also be suppressed. Therefore, in the design of semiconductor switching element 1, it is not necessary to increase the current-carrying area of ​​semiconductor switching element 1 or set a higher rated voltage to avoid the increase in on-resistance and cutoff surge voltage at low temperature; the area of ​​semiconductor switching element 1 can be reduced, and the rated voltage of semiconductor switching element 1 can be lowered. By mitigating the temperature characteristics of the threshold voltage of semiconductor switching element 1, the area and rated voltage of semiconductor switching element 1 can be reduced. Therefore, a small and inexpensive semiconductor module that suppresses the increase in on-resistance and cutoff surge voltage at low temperature can be obtained.

[0026] Considering that the temperature characteristics of the threshold voltage are affected by the stress applied to the first surface 1a where the channel region is formed, but not by the stress applied to the second surface 1b of the semiconductor switching element 1, stress application portions 10a and 10b are preferably configured to generate stress on the first surface 1a. The configuration, size, and number of stress application portions 10a and 10b that generate stress on the first surface 1a are not limited to the structure of this embodiment, but with the structure of this embodiment, stress can be efficiently applied to the first surface 1a with a smaller number of stress application portions 10a and 10b. Furthermore, the smaller number of stress application portions 10a and 10b improves the productivity of the semiconductor module 100. Additionally, stress application portions can be provided only on either the first surface 1a or the second surface 1b. When stress application portions are provided only on either the first surface 1a or the second surface 1b, the number of components is reduced, thus enabling miniaturization of the semiconductor module 100 and improving its productivity.

[0027] In addition, such as Figure 8 As shown, multiple stress application sections 10a can also be further provided. Figure 8 This is a top view showing the main parts of the other semiconductor module 100 according to Embodiment 1. It shows the semiconductor switching element 1, stress application portion 10a, and busbar 5 (dashed line) on the first surface 1a side after the insulating resin material 6 has been removed. This configuration allows for variation in the magnitude and area of ​​stress applied to the first surface 1a, thus easily obtaining the desired temperature characteristics of the threshold voltage. In this embodiment, only an example of increasing the number of stress application portions 10a is shown, but it is not limited to this; the number of stress application portions 10b can also be increased. Furthermore, not only the number of stress application portions can be varied, but also the arrangement and size of the stress application portions can be changed to alter the magnitude and area of ​​stress applied to the first surface 1a.

[0028] In addition, such as Figure 9 As shown, the stress application part 10a and the busbar 5 disposed on the first surface 1a can be made of the same material, and the stress application part 10a and the busbar 5 disposed on the first surface 1a can be integrated. Figure 9 This is a top view showing the main parts of the other semiconductor module 100 according to Embodiment 1, and is a view taken from the semiconductor switching element 1 side showing the integrated stress application portion 10a and busbar 5. The material forming the stress application portion 10a and busbar 5 is, for example, copper. Figure 9In the structure shown, the stress application portion 10a and the busbar 5 can be formed with the same width. By forming the stress application portion 10a and the busbar 5 with the same width, the productivity of the integrated stress application portion 10a and the busbar 5 can be improved. When the stress application portion 10a and the busbar 5 are integrated using the same material, the number of components constituting the semiconductor module 100 is reduced, and the process of joining the stress application portion 10a and the busbar 5 can be eliminated, thus improving the productivity of the semiconductor module 100.

[0029] In the semiconductor switching element 1, the main material can be a compound semiconductor, such as silicon carbide or gallium oxide, which is harder than silicon. Because compound semiconductors are harder than silicon, the stress applied to the semiconductor switching element 1 from the stress application portions 10a and 10b can be increased. This increased stress from the stress application portions 10a and 10b results in a more significant effect in mitigating the temperature characteristic of the threshold voltage. Furthermore, compound semiconductors such as silicon carbide and gallium oxide are formed with a thinner thickness than silicon; therefore, when the semiconductor switching element 1 is a compound semiconductor, the stress applied to the semiconductor switching element 1 from the stress application portions 10a and 10b can be increased.

[0030] As described above, the semiconductor module 100 according to Embodiment 1 includes a stress application portion disposed on one or both of the first surface 1a and the second surface 1b opposite to the first surface 1a of the semiconductor switching element 1. The stress application portion has a coefficient of linear expansion that is larger than that of the main material of the semiconductor switching element 1 and is thicker than the semiconductor switching element 1. The stress application portion generates compressive or tensile stress in the semiconductor switching element 1 by means of thermal contraction or thermal expansion of the stress application portion with temperature change. As the magnitude of the compressive or tensile stress of the semiconductor switching element 1 increases, the threshold voltage for the semiconductor switching element 1 to conduct decreases. Therefore, the increase in on-resistance and the increase in cutoff surge voltage can be suppressed. Thus, the area and rated voltage of the semiconductor switching element 1 can be reduced, and a small and inexpensive semiconductor module 100 can be obtained.

[0031] The first surface 1a has two or more stress-applying portions 10a, and the second surface 1b has one or more stress-applying portions 10b. The two or more stress-applying portions 10a on the first surface 1a are arranged side by side in a first direction parallel to the first surface. When viewed in a direction perpendicular to the first surface 1a, the one or more stress-applying portions 10b on the second surface 1b are positioned between the end 10a1 on one side of the first direction and the end 10a2 on the other side of the first direction among the two or more stress-applying portions 10a on the first surface 1a. In this case, the stress-applying portions 10a and 10b can be arranged to generate stress on the first surface 1a. Furthermore, by varying the number, arrangement, and size of the stress-applying portions, the magnitude and area of ​​stress applied to the first surface 1a can be easily changed, thus easily obtaining the desired temperature characteristics of the threshold voltage.

[0032] The first surface 1a has two stress-applying portions 10a, and the second surface 1b has one stress-applying portion 10b. The two stress-applying portions 10a on the first surface 1a are spaced apart in a first direction parallel to the first surface 1a. When viewed in a direction perpendicular to the first surface 1a, the stress-applying portion 10b on the second surface 1b is positioned between the end portion 10a1 on one side of the first direction of the stress-applying portion 10a on the first surface 1a and the end portion 10a2 on the other side of the first direction of the stress-applying portion 10a on the other side of the first surface 1a. In this configuration, stress can be applied to the first surface 1a efficiently with a smaller number of stress-applying portions 10a and 10b. Furthermore, the smaller number of stress-applying portions 10a and 10b improves the productivity of the semiconductor module 100.

[0033] The stress-applying portion is provided on both the first surface 1a and the second surface 1b. The semiconductor module 100 includes a busbar 5 provided in the stress-applying portion 10a on the first surface 1a and connected to the portion opposite to the semiconductor switching element 1, and a heat dissipation member 2 provided in the stress-applying portion 10b on the second surface 1b and connected to the portion opposite to the semiconductor switching element 1. In this case, the heat dissipation member 2 can easily release the heat generated in the busbar 5 and the semiconductor switching element 1 to the outside. Furthermore, the stress-applying portion 10a and the busbar 5 provided on the first surface 1a are made of the same material and are integrated. In this case, the number of components constituting the semiconductor module 100 is reduced, and the process of joining the stress-applying portion 10a and the busbar 5 can be reduced, thus improving the productivity of the semiconductor module 100.

[0034] When the semiconductor switching element 1 and the stress application portion are bonded together through a bonding layer to prevent slippage, stress can be efficiently applied to the semiconductor switching element 1 from the stress application portions 10a and 10b, and the thermal reliability between the semiconductor switching element 1 and the stress application portions 10a and 10b can be improved. Furthermore, when the bonding layer is formed using metal particles harder than solder, the stress applied to the semiconductor switching element 1 from the stress application portions 10a and 10b can be increased.

[0035] Since a bonding layer is not required when the semiconductor switching element 1 is in contact with the stress application portion, the productivity of the semiconductor module 100 can be improved. Furthermore, when the main material of the semiconductor switching element 1 is a compound semiconductor that is harder than silicon, the stress applied to the semiconductor switching element 1 from the stress application portions 10a and 10b can be increased, thus making the effect of slowing down the temperature characteristics of the threshold voltage more significant.

[0036] Implementation method 2. The semiconductor module 100 according to Embodiment 2 will be described. Figure 10 This is a cross-sectional view showing an outline of the semiconductor module 100 according to Embodiment 2, in conjunction with... Figure 1 The image shows the result of cutting at the same location as the AA section. Figure 11 This is a diagram of the equivalent circuit obtained when the semiconductor switching element 1 of the simulating semiconductor module 100 is turned off. Figure 12 This is a graph showing the correlation between the decrease in the threshold voltage of the semiconductor switching element 1 of the semiconductor module 100 and the increase in the off-time. Figure 13 This is a graph showing the correlation between the deformation of the surface of the semiconductor switching element 1 of the semiconductor module 100 and the amount of change in the threshold voltage. Figure 14 This is a graph showing the correlation between the thickness of the stress-applying portion of the semiconductor module 100 and the deformation generated on the surface of the semiconductor switching element 1. The semiconductor module 100 according to Embodiment 2 differs from that of Embodiment 1 in that the portion that is joined to the second surface 1b of the heat dissipation member, i.e., the heat sink 2a, is configured as the stress-applying portion 10b provided on the second surface 1b.

[0037] <Semiconductor Module 100> Semiconductor module 100 Figure 10 As shown, it includes a semiconductor switching element 1, a heat sink 2a as a heat dissipation component, busbars 4 and 5, a bonding layer 3 that bonds the semiconductor switching element 1 to the heat sink 2a, and an insulating resin material 6 covering them. Figure 10In the diagram, the outline of the insulating resin material 6 is represented by dashed lines. The semiconductor switching element 1 is a power semiconductor formed in a plate shape and mainly made of silicon carbide. The busbar 5 is connected to one side of the first surface 1a of the semiconductor switching element 1, for example, by solder. The heat sink 2a is formed in a plate shape and is bonded to the second surface 1b of the semiconductor switching element 1 via a bonding layer 3. It is made of copper, and the outline of the plate surface is larger than that of the semiconductor switching element 1. The portion bonded to the second surface 1b of the heat sink 2a is the stress application portion 10b provided on the second surface 1b.

[0038] The heat sink 2a functions as an electrode connected to the second surface 1b, and the busbar 4 is connected to the heat sink 2a, for example, by solder. The heat sink 2a is thermally and electrically connected to the semiconductor switching element 1, and releases the heat generated by the semiconductor switching element 1 to the outside. Furthermore, the heat sink 2a releases the heat generated in the busbar 5 via the semiconductor switching element 1 to the outside. A portion of the busbars 4 and 5 is exposed to the outside from the insulating resin material 6. The exposed portions of the busbars 4 and 5 are connected to external devices. The busbars 4 and 5 are made, for example, of copper, which has high conductivity. The thickness of the semiconductor switching element 1 is less than 200 μm, and the thickness of the heat sink 2a is more than 1 mm. The reasons for the above-described size and structure will be explained later.

[0039] With this configuration, similar to Embodiment 1, compressive stress can be applied to the semiconductor switching element 1 by utilizing the thermal contraction of the stress-applying portion 10b accompanied by a decrease in temperature. Figure 10 (The dashed arrow indicates this). By applying compressive stress, the threshold voltage of the semiconductor switching element 1 can be reduced. By reducing the threshold voltage, the increase in on-resistance and the increase in cutoff surge voltage can be suppressed. Therefore, the area and rated voltage of the semiconductor switching element 1 can be reduced, resulting in a small and inexpensive semiconductor module. Furthermore, compared with Embodiment 1, the number of components can be reduced, thus further miniaturizing the semiconductor module 100. In addition, since manufacturing steps are reduced, the manufacturing costs are lowered, thereby improving the productivity of the semiconductor module 100.

[0040] <Thickness of semiconductor switching element 1 and heat sink 2a> The reasons for setting the thickness of the semiconductor switching element 1 to 200 μm or less and the thickness of the heat sink 2a to 1 mm or more are explained. In recent years, there has been a strong interest in replacing silicon with silicon carbide as the main material for the semiconductor switching element 1. Compared to silicon, silicon carbide is superior in physical properties for power semiconductor materials, such as thermal conductivity and dielectric breakdown electric field. Therefore, silicon carbide is a promising semiconductor material for achieving lower power consumption and smaller power conversion devices. Furthermore, the thinner the semiconductor switching element 1, the lower its resistance. Therefore, regardless of whether it is silicon carbide, there is a tendency in the design of power conversion devices to make the thickness of the semiconductor switching element 1 as thin as possible. In particular, silicon carbide has a higher dielectric breakdown electric field than silicon, thus allowing the layer required to maintain voltage withstand to be thinner than that of silicon. Therefore, in most cases, the thickness of the semiconductor switching element 1 made of silicon carbide is generally 200 μm or less. In this embodiment, the case where the semiconductor switching element 1, whose main material is silicon carbide, is constructed with a thickness of 200 μm is described.

[0041] If the temperature characteristic of the threshold voltage of semiconductor switching element 1 is moderated and the increase in threshold voltage at low temperatures is suppressed, then the switching time when semiconductor switching element 1 is turned off increases. Therefore, the time change of current at turn-off decreases, and the parasitic inductance of the circuit and the surge voltage dependent on the time change of current are reduced. Figure 11 In the equivalent circuit shown, the switching off time of semiconductor switching element 1, varying with the threshold voltage, can be considered as a discharge process originating from the gate capacitance Cg. Figure 11 In this case, the gate resistance is set to Rg, the gate capacitance is set to Cg, the voltage applied to the capacitor at the start of discharge is set to Vgp, and the gate negative bias is set to Vgn. In this case, the time change V(t) of the capacitor voltage in the equivalent circuit is represented by equation (1). [Mathematical Expression 1] The cutoff time during switching is the discharge time during which the capacitor voltage drops from the voltage Vm at the end of the mirror period to the threshold voltage Vth, and is therefore expressed by equation (2). [Mathematical Expression 2]

[0042] As can be seen from equation (2), the cutoff time during switching varies non-linearly with respect to the threshold voltage. Figure 12An example illustrating the relationship between the rate of change of the time required for switch cutoff and the rate of change of the threshold voltage of a semiconductor switching element 1 made of silicon carbide is shown. If the threshold voltage of semiconductor switching element 1 decreases by more than 10%, the cutoff time increases sharply. That is, reducing the threshold voltage by more than 10% is effective in order to significantly suppress the increase of the cutoff surge voltage.

[0043] Figure 13 This diagram illustrates the relationship between the amount of deformation occurring on the surface of a silicon carbide semiconductor switching element 1, specifically on its first surface 1a, and the rate of change in the threshold voltage of the semiconductor switching element 1. As described above, the first surface 1a of the semiconductor switching element 1 is the side where the source electrode of the MOSFET is formed, and the channel region, which is the primary factor determining the threshold voltage, is located. Figure 13 As shown, the authors confirmed that the threshold voltage of semiconductor switching element 1 decreases with increasing deformation in the first surface 1a of the semiconductor switching element. Figure 13 It is known that when silicon carbide is used in the semiconductor switching element 1, in order to reduce the threshold voltage by 10%, the first surface 1a of the semiconductor switching element 1 needs to have a density of 1000 × 10⁻⁶ mm. -6 The deformation.

[0044] Figure 14 This involves bonding a 200μm thick plate of silicon carbide to a copper plate via a bonding layer at high temperature, and varying the thickness of the copper plate to calculate the deformation caused by stress applied to the surface of the silicon carbide on the opposite side of the copper plate. The greater the thickness of the copper plate, the greater the stress applied to the silicon carbide surface, and the greater the surface deformation. It is known that to achieve a 10% threshold voltage variation and generate a 1000×10⁻⁶ deformation on the surface of semiconductor switching element 1... -6 The deformation can be achieved by using a copper plate with a thickness of 1 mm or more to form a heat sink 2a with a stress application portion 10b.

[0045] Based on the above considerations, it can be said that in the semiconductor module 100, which includes a semiconductor switching element 1 made of silicon carbide with a thickness of 200 μm and a heat sink 2a having a stress application portion 10b made of copper, in order to effectively suppress the increase of switching surge voltage at low temperatures, it is preferable to form the thickness of the stress application portion 10b to be 1 mm or more. Furthermore, in the heat sink 2a, as long as the stress application portion 10b is made of copper, it is acceptable for other parts of the heat sink 2a to be made of other materials.

[0046] Furthermore, the case where the thickness of the semiconductor switching element 1 made of silicon carbide is set to 200 μm is described, but the thickness of the semiconductor switching element 1 can also be made thinner than 200 μm. When the thickness of the semiconductor switching element 1 is made thinner than 200 μm, the rigidity of the semiconductor switching element 1 decreases. Therefore, when the semiconductor switching element 1 receives the same stress from the stress application portion 10b as when the thickness is 200 μm, the amount of deformation generated on the surface of the semiconductor switching element 1 increases. As a result, the effect of lowering the threshold voltage at low temperatures increases, and therefore, the effect of suppressing the increase of the cutoff surge voltage during switching and the effect of suppressing the increase of the on-resistance can be further amplified. Alternatively, the thickness of the portion of the stress application portion 10b can be made larger than 1 mm without changing the thickness of the semiconductor switching element 1. When the thickness of the portion of the stress application portion 10b is larger than 1 mm, the stress applied to the semiconductor switching element 1 increases, and therefore, the same effect as when the thickness of the semiconductor switching element 1 is thinned can be obtained.

[0047] <Joint Layer 3> The bonding layer 3 is formed using metal particles, such as silver or copper particles, and is made of a material harder than solder. When a harder material, such as metal particles, is used in the bonding layer 3, the stress applied from the stress application portion 10b to the semiconductor switching element 1 can be increased compared to when a harder material is not used. The thickness of the bonding layer 3 is approximately 10 μm to 50 μm.

[0048] Preferably, the thickness of the bonding layer 3 is as thin as possible, and its coefficient of linear expansion is close to that of the semiconductor switching element 1. If the bonding layer 3 is thicker, the stress transmitted from the stress application portion 10b to the semiconductor switching element 1 is mitigated, thus reducing the effect of mitigating the temperature characteristics of the threshold voltage of the semiconductor switching element 1. Similarly, if there is a difference in the coefficient of linear expansion between the semiconductor switching element 1 and the bonding layer 3, the stress transmitted from the stress application portion 10b to the semiconductor switching element 1 is mitigated, thus reducing the effect of mitigating the temperature characteristics of the threshold voltage of the semiconductor switching element 1. In the case where the stress applied to the semiconductor switching element 1 from the stress application portion 10b is mitigated by the bonding layer 3, a thicker heat sink 2a can be used. By using a thicker heat sink 2a, even when the stress is mitigated by the bonding layer 3, the required amount of stress can still be applied to the semiconductor switching element 1, and the desired mitigation effect of the temperature characteristics of the threshold voltage can be obtained.

[0049] Preferably, when viewed in a direction perpendicular to the first surface 1a of the semiconductor switching element 1, the shape of the bonding layer 3 is larger than or the same size as the shape of the semiconductor switching element 1. Even when the bonding layer 3 is formed in an area smaller than the shape of the semiconductor switching element 1 when viewed in a direction perpendicular to the first surface 1a of the semiconductor switching element 1, the effects of this application can be achieved. However, by forming the bonding layer 3 to a size larger than the shape of the semiconductor switching element 1, stress can be transferred from the stress application portion 10b to the entire semiconductor switching element 1. Therefore, by forming the bonding layer 3 to a size larger than the shape of the semiconductor switching element 1, the effects of this application can be fully utilized.

[0050] <Number of semiconductor switching elements 1> Multiple semiconductor switching elements 1 can be configured to be bonded to the same surface of the heat sink 2a via a bonding layer 3. In this embodiment, as... Figure 15 As shown, an example is illustrated where two semiconductor switching elements 1 are bonded to the same surface of a heat sink 2a. Figure 15 This is a side view showing an outline of another semiconductor module 100 according to Embodiment 2, and is a diagram showing the process after removing the insulating resin material 6. In power conversion devices that handle high currents, considering the manufacturing yield and cost of the semiconductor switching element 1, a method of arranging multiple semiconductor switching elements 1 in parallel is often used to ensure the current output. If the semiconductor switching element 1 becomes too large as the current increases, the yield of the semiconductor switching element 1 decreases. Therefore, arranging and connecting multiple semiconductor switching elements 1 of a certain size in parallel can better suppress the overall cost. The size of the semiconductor switching element 1 in this embodiment, which is formed into a rectangular shape, is, for example, 4mm × 4mm.

[0051] This configuration allows multiple semiconductor switching elements 1 to be arranged side-by-side on a heat sink 2a, reducing the cost of the semiconductor switching elements 1 while ensuring the required current. Furthermore, compared to individually providing a stress application portion for each semiconductor switching element 1, simultaneously bonding multiple semiconductor switching elements 1 side-by-side onto a heat sink 2a reduces manufacturing steps, thus improving the productivity of the semiconductor module 100. Moreover, since each semiconductor switching element 1 bonded to the heat sink 2a is a stress application portion 10b, a temperature characteristic reduction effect on the threshold voltage of each of the multiple semiconductor switching elements 1 is achieved. Additionally, a bonding layer 3 is provided directly below each of the multiple semiconductor switching elements 1. Furthermore, a busbar 5 is provided connected to electrodes located on the first surface 1a of each of the multiple semiconductor switching elements 1.

[0052] Figure 16This is a top view showing the main parts of the other semiconductor module 100 according to Embodiment 2. It is a diagram showing the semiconductor switching element 1, heat sink 2a, and bonding layer 3 on the first surface 1a side after the insulating resin material 6 has been removed. When two semiconductor switching elements 1 are provided, the point where the stress generated on the surface of the heat sink 2a is the greatest can be achieved. Figure 16 Using the center O of the heat sink 2a as the origin, two semiconductor switching elements 1 are arranged symmetrically at the origin. This configuration uniformizes the stress applied to each semiconductor switching element 1, thus homogenizing the threshold voltage in each semiconductor switching element 1. Consequently, current deviation during switching is reduced, thus decreasing the maximum current flowing through the semiconductor switching element 1 and further improving the suppression of surge voltage.

[0053] Figure 17 This is a top view showing the main parts of the other semiconductor module 100 involved in Embodiment 2, and a diagram showing the semiconductor switching element 1, heat sink 2a and bonding layer 3 on the first surface 1a side after the insulating resin material 6 has been removed. Figure 17 and Figure 16 The difference in the configuration of the semiconductor switching element 1 shown is that, Figure 17 Consider the following example: Semiconductor switching element 1 is positioned at a non-origin symmetrical location on heat sink 2a. It is known that the current flowing through each semiconductor switching element 1 may deviate due to the influence of the wiring length and arrangement of busbars 4 and 5 in semiconductor module 100. In this case, by changing the stress applied to each semiconductor switching element 1, adjustments can be made to significantly reduce the threshold voltage of a specific semiconductor switching element 1, thereby balancing the current flow of each semiconductor switching element 1. This further suppresses the increase in cutoff surge voltage during switching at low temperatures.

[0054] As described above, the semiconductor module 100 in Embodiment 2 includes: a semiconductor switching element 1 made of silicon carbide as the main material; a busbar 5; and a heat sink 2a. The heat sink 2a is bonded to the second surface 1b of the semiconductor switching element 1 via a bonding layer 3. The heat sink 2a is made of copper and its surface area is larger than that of the semiconductor switching element 1. The portion bonded to the second surface 1b of the heat sink 2a is a stress application portion 10b provided on the second surface 1b. The thickness of the semiconductor switching element 1 is less than 200 μm, and the thickness of the heat sink 2a is more than 1 mm. Therefore, the increase of the switching surge voltage of the semiconductor switching element 1 at low temperature can be effectively suppressed.

[0055] The bonding layer 3 is formed using metal particles that are harder than solder. When viewed in a direction perpendicular to the first surface 1a of the semiconductor switching element 1, the shape of the bonding layer 3 is larger than or the same size as the shape of the semiconductor switching element 1. In this case, stress can be transferred from the stress application portion 10b to the entire semiconductor switching element 1. Therefore, the increase in the switching surge voltage of the semiconductor switching element 1 at low temperatures can be further effectively suppressed.

[0056] When multiple semiconductor switching elements 1 are bonded to the same surface of the heat sink 2a via the bonding layer 3, the process of bonding multiple semiconductor switching elements 1 side by side to a single heat sink 2a reduces manufacturing steps compared to the case where each semiconductor switching element 1 has an independently provided stress application portion. Therefore, the productivity of the semiconductor module 100 can be improved. Furthermore, since each portion of the semiconductor switching element 1 bonded to the heat sink 2a is a stress application portion 10b, the temperature characteristics of the threshold voltage of each of the multiple semiconductor switching elements 1 are softened.

[0057] Implementation method 3. The semiconductor module 100 involved in Embodiment 3 will be described. Figure 18 This is a cross-sectional view showing an outline of the semiconductor module 100 according to Embodiment 3, in conjunction with... Figure 1 The image shows the semiconductor module 100 cut at the same position as the AA section. The difference between Embodiment 3 and Embodiment 2 is that the semiconductor module 100 includes a stress application section 10b. Furthermore, unlike Embodiment 1, only the stress application section 10b is provided, and the stress application section 10a is not provided.

[0058] One or more stress-applying portions 10b are provided on the second surface 1b of the semiconductor switching element 1. In this embodiment, one stress-applying portion 10b is provided with a size larger than the outer diameter of the semiconductor switching element 1, and the stress-applying portion 10b is made of copper with a thickness of 1 mm or more. The shape, material, and number of stress-applying portions 10b are not limited to these; other shapes, materials, and numbers are also possible as long as stress can be applied to the desired location of the semiconductor switching element 1. The stress-applying portion 10b and the semiconductor switching element 1 are bonded, for example, via a bonding layer (not shown).

[0059] The semiconductor module 100 includes a heat dissipation member 2, which is connected to a portion of the stress application portion 10b disposed on the second surface 1b opposite to the side of the semiconductor switching element 1. The heat dissipation member 2 and the stress application portion 10b are connected, for example, by solder. The heat dissipation member 2 functions as an electrode connected to the second surface 1b, and a busbar 4 is connected to the heat dissipation member 2. The busbar 5 is connected, for example, to the first surface 1a side of the semiconductor switching element 1 by solder. In this embodiment, with Figure 2 Compared to the structure of Embodiment 1 shown, the stress application part 10a is not provided, and the number of parts is less, thus enabling the semiconductor module 100 to be miniaturized.

[0060] In Embodiment 2, the portion bonded to the second surface 1b of the heat sink 2a is the stress application portion 10b provided on the second surface 1b. However, by adopting the structure shown in Embodiment 3, the stress application portion 10b and the heat dissipation member 2 can be provided as separate components. Therefore, the adjustment of the threshold voltage by applying stress to the semiconductor switching element 1 in the stress application portion 10b and the heat dissipation performance of the semiconductor switching element 1 in the heat dissipation member 2 can be considered separately. Thus, compared with Embodiment 2, the construction design of the semiconductor module 100 can be easily performed. For example, if the threshold voltage of the semiconductor switching element 1 is to be further reduced when the heat dissipation performance is sufficient, the material or thickness of the stress application portion 10b can be changed. In this case, no design change of the heat dissipation member 2 is required, and the thickness of the heat dissipation member 2 can be suppressed to the required minimum. Therefore, the cost of the semiconductor module 100 can be reduced.

[0061] In this embodiment, when the semiconductor module 100 includes multiple semiconductor switching elements 1, a stress application portion 10b can be provided for each semiconductor switching element 1. Therefore, by changing the thickness and material of the stress application portion 10b for each semiconductor switching element 1, the amount of variation in the threshold voltage in each semiconductor switching element 1 can be adjusted. There is a situation where it is known in advance that the current flowing through each semiconductor switching element 1 will deviate due to the influence of the wiring length and arrangement of the busbars 4 and 5 in the semiconductor module 100. In this case, by changing the stress applied to each semiconductor switching element 1, an adjustment can be made to reduce the threshold voltage of a specific semiconductor switching element 1 to a greater extent, thereby balancing the current of each semiconductor switching element 1. This further suppresses the increase in cutoff surge voltage during switching at low temperatures.

[0062] As described above, in the semiconductor module 100 according to Embodiment 3, one or more stress application portions 10b are provided on the second surface 1b, and a heat sink 2a is provided. The heat sink 2a is connected to the portion of the stress application portion 10b provided on the second surface 1b that is opposite to the side of the semiconductor switching element 1. Therefore, since the stress application portion 10a is not provided and the number of components is small, the semiconductor module 100 can be miniaturized. In addition, the stress application portion 10b and the heat sink 2 can be provided as independent components. Therefore, the adjustment of the threshold voltage based on the stress application portion 10b and the heat dissipation performance based on the heat sink 2 can be considered separately. Therefore, compared with Embodiment 2, the structural design of the semiconductor module 100 can be easily carried out.

[0063] Furthermore, although this application describes various exemplary implementation methods and embodiments, the various features, methods and functions described in one or more implementation methods are not limited to specific implementation methods, but can also be applied to implementation methods individually, or in various combinations to be applied to implementation methods. Therefore, it can be assumed that numerous variations not illustrated are also included within the scope of the technology disclosed in this application. For example, this includes cases where at least one constituent element is modified, added to, or omitted, and cases where at least one constituent element is extracted and combined with constituent elements of other embodiments. Label Explanation

[0064] 1 Semiconductor switching element 1a First surface 1b Second Surface 2. Heat dissipation components 2a Heat sink 3 Bonding layer 4. Busbar 5. Busbar 6. Insulating resin materials 10a Stress application section 10a1 end 10a2 end 10b Stress application section 100 Semiconductor Modules.

Claims

1. A semiconductor module, characterized in that, include: Semiconductor switching elements; as well as A stress-applying portion is disposed on both a first surface and a second surface opposite to the first surface of the semiconductor switching element. This stress-applying portion has a coefficient of linear expansion greater than that of the main material of the semiconductor switching element, and its thickness is greater than that of the semiconductor switching element. The stress-applying part utilizes the thermal contraction or expansion of the stress-applying part in response to temperature changes to generate compressive or tensile stress in the semiconductor switching element. As the magnitude of the compressive stress or tensile stress in the semiconductor switching element increases, the threshold voltage at which the semiconductor switching element turns on decreases. The first surface is provided with two or more stress application portions. The second surface is provided with one or more of the aforementioned stress application portions. Two or more stress-applying portions disposed on the first surface are arranged side by side in a first direction parallel to the first surface. When viewed in a direction perpendicular to the first surface, one or more stress-applying portions provided on the second surface are disposed between the end of one side of the first direction and the end of the other side of the first direction among two or more stress-applying portions provided on the first surface.

2. The semiconductor module as described in claim 1, characterized in that, The stress-applying portion is disposed on both the first surface and the second surface. The semiconductor module includes: a busbar connected to a portion of the stress-applying portion disposed on the first surface opposite to the side of the semiconductor switching element; and A heat dissipation component is connected to the portion of the stress application portion disposed on the second surface that is opposite to the side of the semiconductor switching element.

3. The semiconductor module as described in claim 2, characterized in that, The stress-applying portion disposed on the first surface and the busbar are made of the same material, and the stress-applying portion disposed on the first surface and the busbar are integrated.

4. The semiconductor module as described in claim 1, characterized in that, One or more of the stress-applying portions are disposed on the second surface. The semiconductor module includes a heat dissipation member connected to a portion of the stress application portion disposed on the second surface opposite to the side of the semiconductor switching element.

5. The semiconductor module as described in claim 1, characterized in that, The semiconductor switching element and the stress application portion are bonded together through a bonding layer to prevent them from slipping.

6. The semiconductor module as described in claim 5, characterized in that, The bonding layer is formed using metal particles that are harder than solder.

7. The semiconductor module as described in claim 1, characterized in that, The semiconductor switching element abuts against the stress application portion.

8. The semiconductor module as described in claim 1, characterized in that, The main material of the semiconductor switching element is a compound semiconductor, which is harder than silicon.

9. A semiconductor module, characterized in that, include: Semiconductor switching elements; as well as A stress-applying portion is disposed on both a first surface and a second surface opposite to the first surface of the semiconductor switching element. This stress-applying portion has a coefficient of linear expansion greater than that of the main material of the semiconductor switching element, and its thickness is greater than that of the semiconductor switching element. The stress-applying part utilizes the thermal contraction or expansion of the stress-applying part in response to temperature changes to generate compressive or tensile stress in the semiconductor switching element. As the magnitude of the compressive stress or tensile stress in the semiconductor switching element increases, the threshold voltage at which the semiconductor switching element turns on decreases. Two stress-applying portions are provided on the first surface. One stress-applying portion is provided on the second surface. The two stress-applying portions disposed on the first surface are spaced apart in a first direction parallel to the first surface. When viewed in a direction perpendicular to the first surface, one of the stress-applying portions disposed on the second surface is positioned between the end of the stress-applying portion disposed on the first direction side of the first surface and the end of the stress-applying portion disposed on the other direction side of the first surface.

10. The semiconductor module as described in claim 9, characterized in that, The stress-applying portion is disposed on both the first surface and the second surface. The semiconductor module includes: a busbar connected to a portion of the stress-applying portion disposed on the first surface opposite to the side of the semiconductor switching element; and A heat dissipation component is connected to the portion of the stress application portion disposed on the second surface that is opposite to the side of the semiconductor switching element.

11. The semiconductor module as claimed in claim 10, characterized in that, The stress-applying portion disposed on the first surface and the busbar are made of the same material, and the stress-applying portion disposed on the first surface and the busbar are integrated.

12. The semiconductor module as described in claim 9, characterized in that, The semiconductor switching element and the stress application portion are bonded together through a bonding layer to prevent them from slipping.

13. The semiconductor module as described in claim 12, characterized in that, The bonding layer is formed using metal particles that are harder than solder.

14. The semiconductor module as claimed in claim 9, characterized in that, The semiconductor switching element abuts against the stress application portion.

15. The semiconductor module as described in claim 9, characterized in that, The main material of the semiconductor switching element is a compound semiconductor, which is harder than silicon.

16. The semiconductor module as claimed in claim 9, characterized in that, include: The semiconductor switching element is formed in a plate shape and is mainly made of silicon carbide; A busbar connected to one side of the first surface of the semiconductor switching element; as well as A heat dissipation component, formed in the shape of a plate, is bonded to the second surface of the semiconductor switching element via a bonding layer. The plate is made of copper, and its external dimensions are larger than those of the semiconductor switching element. The portion of the heat dissipation component that is joined to the second surface is the stress application portion disposed on the second surface. The thickness of the semiconductor switching element is less than 200 μm, and the thickness of the heat dissipation component is more than 1 mm.

17. The semiconductor module as claimed in claim 16, characterized in that, The bonding layer is formed using metal particles that are harder than solder. When viewed in a direction perpendicular to the first surface of the semiconductor switching element, The outer shape of the bonding layer is larger than or the same size as the outer shape of the semiconductor switching element.

18. The semiconductor module as described in claim 16 or 17, characterized in that, Multiple semiconductor switching elements are bonded to the same surface of the heat dissipation member via the bonding layer.

Images