WO2017208941A1 - Semiconductor device and method for manufacturing same - Google Patents

Abstract

A semiconductor device provided with an insulating substrate (5) having conductor patterns (52a, 52b) on both sides of an insulating base material (51), a semiconductor element (1) mounted on one of the conductor patterns, and an electrode terminal (3) bonded to the one conductor pattern by means of a hard brazing material (10), wherein the electrode terminal has a larger heat capacity per unit area than the heat capacity per unit area in the one conductor pattern.

Description

Semiconductor device and manufacturing method thereof

The present invention relates to a semiconductor device and a manufacturing method thereof, and more particularly to a power semiconductor device and a manufacturing method thereof.

In conventional semiconductor devices, particularly power semiconductor devices that handle large currents, it is necessary to bond electrode terminals in a large area in order to efficiently flow large currents, so solder bonding is used for bonding electrode terminals. (See Patent Document 1). However, as the temperature environment in which the power semiconductor device is used becomes severe, there is a possibility that the required reliability cannot be satisfied even in the case of using a high heat-resistant solder in the conventional solder joint.

As a method of solving this problem, a method of joining the electrode terminals using a high heat resistance solder or a brazing filler metal having a melting point higher than that of the solder can be considered. However, since the melting point of the brazing filler metal is generally as high as 450 ° C. or higher, brazing using an ordinary torch or in-furnace brazing, a member around the semiconductor device, such as an adhesive, a resin case, etc. There is a problem that the solder that joins the insulating substrate to the base plate or the solder that die-bonds the semiconductor element melts. Therefore, in order to braze the electrode terminal of the semiconductor device using a hard brazing material, it is necessary to locally heat the tip of the electrode terminal, which is the part to be brazed, in a short time.

In Patent Document 2, as a method for locally heating the tip of an electrode terminal in a short time, the melting point of the electrode terminal and the surface electrode is higher than at least one of the surfaces facing each other in the electrode terminal and the surface electrode of the semiconductor element. A low melting point metal layer is formed in advance. And the technique which fuse | melts the surface of an electrode terminal with a laser beam, fuses a low melting metal layer with the heat | fever, and joins a surface electrode and an electrode terminal is proposed.

JP 2006-253516 A JP 2009-105266 A

は ん だ Soldering or brazing using laser light does not require a heating process for the entire semiconductor device. Therefore, at the time of joining the electrode terminals, for example, the electrode terminals can be joined without melting the entire solder joining the base plate and the insulating substrate, the surrounding adhesive, the case, or the like.

However, in Patent Document 2, as described above, the object to be melted by heating with laser light is a low melting point metal layer, not a high heat resistance solder or a brazing filler metal having a higher melting point than solder. Further, in Patent Document 2, the object to which the electrode terminal is bonded is the surface electrode in the semiconductor element, and is not in the form of bonding the electrode terminal to the conductor pattern on the insulating substrate, for example.

The present invention has been made to solve the above-described problems. In a semiconductor device in which electrode terminals are hard-brazed to a conductor pattern on an insulating substrate, the electrode terminal and the conductor pattern are higher than conventional ones. An object of the present invention is to provide a semiconductor device having the following junction reliability and a method for manufacturing the same.

In order to achieve the above object, the present invention is configured as follows.
That is, the semiconductor device according to one embodiment of the present invention includes an insulating substrate having a conductor pattern on both surfaces of an insulating base, a semiconductor element mounted on one conductor pattern on the insulating substrate, and the one conductor pattern. In the semiconductor device including the electrode terminal bonded via the bonding material, the bonding material is a hard brazing material, and the electrode terminal has a unit area larger than a heat capacity per unit area in the one conductor pattern. It has a heat capacity per unit.

According to the semiconductor device of one embodiment of the present invention, since the electrode terminal has a larger heat capacity per unit area than the conductor pattern, the difference in heat capacity between the electrode terminal and the conductor pattern is reduced, and the electrode terminal is used for local heating. And the temperature difference between the conductor pattern is reduced. Therefore, it is possible to prevent the electrode terminal from being melted before the hard brazing material is melted. Therefore, higher bonding reliability between the electrode terminal and the conductor pattern can be obtained as compared with the conventional case.

FIG. 3 is a cross-sectional view showing a part of the configuration of the power semiconductor device in the first embodiment. 1B is a plan view of the power semiconductor device shown in FIG. 1A. FIG. FIG. 6 is a cross-sectional view showing a part of the configuration of a power semiconductor device in a second embodiment. FIG. 2B is a plan view of the power semiconductor device shown in FIG. 2A. It is sectional drawing for demonstrating an example of the manufacturing method of the power semiconductor device shown to FIG. 2A. It is a top view of the power semiconductor device shown in FIG. 2C. It is sectional drawing for demonstrating the other example of the manufacturing method of the semiconductor device for electric power shown to FIG. 2A. It is a top view of the power semiconductor device shown to FIG. 2E. It is sectional drawing for demonstrating another example of the manufacturing method of the power semiconductor device shown to FIG. 2A. FIG. 2B is a plan view of the power semiconductor device shown in FIG. 2G. It is sectional drawing for demonstrating the manufacturing method (Embodiment 3) of the semiconductor device for electric power shown to FIG. 1A. It is sectional drawing for demonstrating the manufacturing method of the power semiconductor device shown to FIG. 1A. It is sectional drawing for demonstrating the manufacturing method of the power semiconductor device shown to FIG. 1A. It is sectional drawing of the semiconductor device for electric power for demonstrating the method to join the conductor pattern and electrode terminal in an insulated substrate using solder. It is sectional drawing of the semiconductor device for electric power for demonstrating the method to join the conductor pattern and electrode terminal in an insulated substrate using solder. It is sectional drawing of the semiconductor device for electric power for demonstrating the method to join the conductor pattern and electrode terminal in an insulated substrate using solder. It is sectional drawing of the semiconductor device for electric power for demonstrating the method to join the conductor pattern and electrode terminal in an insulated substrate using a brazing material. It is sectional drawing of the semiconductor device for electric power for demonstrating the method to join the conductor pattern and electrode terminal in an insulated substrate using a brazing material. FIG. 9 is a cross-sectional view showing a part of the configuration of a power semiconductor device in a fourth embodiment. FIG. 6B is a plan view of the power semiconductor device shown in FIG. 6A. FIG. 10 is a cross-sectional view showing a positional relationship between a through hole and a light beam in a fourth embodiment. It is sectional drawing which shows the state of the heating by a laser beam, and is a figure which shows the case where there is no through-hole. It is sectional drawing which shows the state of the heating by a laser beam, and is a figure which shows the case where there exists a through-hole. It is a top view which shows the other example of the irradiation method by a laser beam. It is sectional drawing which shows a part of structure of the semiconductor device for electric power concerned about the case where Embodiment 4 is not used. FIG. 6B is a plan view illustrating another example of the configuration of the power semiconductor device illustrated in FIG. 6A. FIG. 6B is a plan view illustrating another example of the configuration of the power semiconductor device illustrated in FIG. 6A. It is a figure which shows the modification of a through-hole. It is a figure which shows the modification of a through-hole. It is sectional drawing which shows the usage number of a terminal deformation | transformation at the time of pressurizing an electrode terminal with a jig | tool. In the semiconductor device for electric power in Embodiment 1, it is the figure which showed the state of the electrode terminal by the irradiation of a laser beam and a brazing material, and the joining property of an electrode terminal and a conductor pattern with the change of the thickness of an electrode terminal. In Embodiment 4, it is sectional drawing which shows a part of structure of the semiconductor device for electric power for demonstrating the contact angle of the electrode terminal and brazing material in the lower part of an electrode terminal. FIG. 15B is a plan view of the power semiconductor device shown in FIG. 15A. FIG. 10 is a cross-sectional view showing a part of the configuration of a power semiconductor device in a fifth embodiment. It is a figure which shows the modification of the press structure shown in FIG. It is a figure which shows the other modification of the press structure shown in FIG. It is a figure which shows another modification of the press structure shown in FIG.

The semiconductor device according to the embodiment and the manufacturing method thereof will be described below with reference to the drawings. In each figure, the same or similar components are denoted by the same reference numerals. In addition, in order to avoid the following description from becoming unnecessarily redundant and to facilitate understanding by those skilled in the art, a detailed description of already well-known matters and a duplicate description of substantially the same configuration may be omitted. . Further, the contents of the following description and the accompanying drawings are not intended to limit the subject matter described in the claims.

Moreover, between each figure, the size or scale of each corresponding component is independent. For example, in the drawing in which a part of the configuration is changed and the illustration in which the configuration is not changed, the size or scale of the same component may be different. In addition, the configuration of the power semiconductor device is actually provided with a plurality of members, but for the sake of simplicity, only the portions necessary for the description are described, and the description of the other portions is omitted. ing.
In the following description, a power semiconductor device is taken as an example, but each embodiment can also be applied to a semiconductor device that handles a normal current instead of power.

Embodiment 1 FIG.
FIG. 1 is a partial cross-sectional view (FIG. 1A) and a plan view (FIG. 1B) showing a schematic configuration of a power semiconductor device 101 according to the first embodiment. The power semiconductor device 101 includes a power semiconductor element 1, an electrode terminal 3, and an insulating substrate 5 as basic components, and can further include a heat dissipation member 6 and a case 12 as illustrated. The power semiconductor element 1 and the electrode terminal 3 are disposed on one side of the insulating substrate 5, and the heat dissipating member 6 is disposed on the other surface side of the insulating substrate 5, and the power semiconductor element is disposed around the heat dissipating member 6. A case 12 is disposed so as to surround 1 and the electrode terminal 3. Further details will be described below.

As the power semiconductor element 1, for example, an IGBT (Insulated Gate Bipolar Transistor) or the like corresponds.

The insulating substrate 5 will be described.
The insulating substrate 5 includes an insulating base material 51, a conductor pattern 52a formed on one surface of both surfaces facing each other in the thickness direction of the insulating base material 51, and a conductor pattern 52b formed on the other surface. The insulating base 51 is an electrical insulator and is preferably a material having a high thermal conductivity in order to efficiently cool the power semiconductor element 1, and generally has a thickness of, for example, 0.635 mm or 0.32 mm. A ceramic plate of AlN, Si ₃ N ₄ , Al ₂ O ₃ or the like is used.

In general, the same material is used for the conductor pattern 52a and the conductor pattern 52b. The main electrode of the power semiconductor element 1 is soldered to one conductor pattern 52a corresponding to the circuit surface side, and the conductor pattern 52a is an electrode terminal with a hard brazing material 10 corresponding to an example of a bonding material interposed therebetween. 3 and a joint part are formed. Since such a conductor pattern 52a is a wiring member for electrically connecting the power semiconductor element 1 and an external circuit, a metal having a small electric resistance is preferable. Therefore, the conductor patterns 52a and 52b are generally made of Cu or Al having a thickness of about 1.0 mm or less, for example.
The heat radiating member 6 is joined by solder 8 to the other conductor pattern 52b corresponding to the heat radiating surface.

Next, the electrode terminal 3 will be described.
The electrode terminal 3 is a wiring member for electrically connecting the power semiconductor element 1 and an external circuit. Therefore, the material of the electrode terminal 3 is preferably a metal having a low electric resistance, and is generally Cu or Al. Such an electrode terminal 3 is obtained by cutting a sheet metal or by pressing, and one end of the electrode terminal 3 forms a joint with the conductor pattern 52a with the brazing filler metal 10 interposed therebetween, and the other end is a case. 12 extends to the outer surface and is electrically connected to another circuit member or an external circuit.

Moreover, in order to braze the electrode terminal 3 to the conductor pattern 52a using the brazing filler metal 10, heat is indirectly applied from the surface 3a side of the electrode terminal 3 to heat the brazing filler metal 10 and the conductor pattern 52a. There is a need to. Therefore, it is preferable that the heat capacity of the electrode terminal 3 is larger than the heat capacity of the conductor pattern 52a so that the temperature difference between the electrode terminal 3 and the conductor pattern 52a does not increase during heating.

Here, the heat capacity is the amount of heat necessary to raise the temperature of the object by a unit temperature of 1K. That is, when the same amount of heat is applied, it is preferable that the electrode terminal 3 is less likely to rise in temperature than the conductor pattern 52a. For example, when the electrode terminal 3 and the conductor pattern 52a are made of the same material, for example, Cu, the volume of the electrode terminal 3 is preferably larger than the volume of the conductor pattern 52a because the heat capacity is proportional to the volume. On the other hand, when the materials of the electrode terminal 3 and the conductor pattern 52a are different, the amount of heat required to raise the temperature of the substance with constant conditions such as weight and size is different by 1K. It is necessary to consider.

However, as described above, since the heat dissipation member 6 is joined to the other conductor pattern 52b, the insulating substrate 5 serves as a heat dissipation path. Therefore, when it is considered as one object including the heat radiation member 6 connected to the insulating substrate 5, generally, the heat capacity of the electrode terminal 3 is smaller than that of the insulating substrate 5, and moreover, for brazing. Since heating is performed from the electrode terminal 3 side, the temperature of the electrode terminal 3 is likely to rise during brazing.

However, when the electrode terminal 3 is heated with laser light, only the portion irradiated with the laser light is instantaneously heated. Therefore, even if the heat capacity of the electrode terminal 3 as a whole is not necessarily larger than the heat capacity of the conductor pattern 52a, the heat capacity of the electrode terminal 3 heated by the laser light and the heat capacity in the vicinity thereof are electrodes viewed from the irradiation direction of the laser light. What is necessary is just to become larger than the heat capacity of the conductor pattern 52a in the same plane as the terminal 3.

From the above viewpoint, it is preferable that the electrode terminal 3 has a relatively large cross-sectional area in order to make the heat capacity of the electrode terminal 3 larger than the heat capacity of the conductor pattern 52a and further increase the current that can be energized. Here, the cross-sectional area of the electrode terminal 3 is an area of the cross section of the electrode terminal 3 in a plane (that is, a yz plane) perpendicular to the x direction shown in FIGS. 1A and 1B. The cross-sectional area of the electrode terminal 3 is determined by the width and thickness of the electrode terminal 3, but in order to reduce the size of the power semiconductor device 101, it is preferable that the width of the electrode terminal 3 is small. Here, the width of the electrode terminal 3 is a dimension of the electrode terminal 3 along the y direction shown in FIGS. 1A and 1B.

Therefore, in order to increase the cross-sectional area of the electrode terminal 3, it is preferable to increase the thickness rather than the width of the electrode terminal 3. For example, when the electrode terminal 3 is heated with laser light, the relationship between the width dimension and thickness of the electrode terminal 3 is preferably as follows. That is, in the electrode terminal 3, when the surface to be bonded to the brazing filler metal 10 is the bonding surface 3j, in order to spread the heat applied to the surface 3a of the electrode terminal 3 by the laser beam over the entire bonding surface 3j, the electrode In the width direction of the terminal 3, the length from the end of the laser light irradiation region of the electrode terminal 3 to the end of the electrode terminal 3 is preferably about the thickness dimension of the electrode terminal 3 or less. More specifically, for example, when a region having a radius of 1.0 mm from the center of the bonding surface 3j is heated with a laser beam with respect to the electrode terminal 3 having a thickness of 1.5 mm, the width dimension of the electrode terminal 3 is 5. It is preferably 0 mm (= 1.0 mm × 2 + 1.5 mm × 2) or less.
The heating operation of the electrode terminal 3 will be described in detail in a third embodiment to be described later, including the case of using laser light.

On the other hand, if the thickness of the electrode terminal 3 is excessively increased, the temperature difference between the bonding surface 3j and the surface 3a of the electrode terminal 3 that is opposed to the bonding surface 3j and is heated by irradiation with laser light becomes too large. Therefore, the thickness of the electrode terminal 3 is preferably in the range of about 0.5 mm to 2.0 mm as an example, but more preferably 0.8 mm to 2.0 mm. As one means for making the heat capacity of the electrode terminal 3 larger than the heat capacity of the conductor pattern 52a, the electrode terminal 3 has a thickness exceeding the thickness of the conductor pattern 52a.

By configuring the electrode terminal 3 as described above, the heat capacity per unit area (1 mm ² ) of the electrode terminal 3 heated by the laser beam is set to be a conductor pattern facing the electrode terminal 3 heated by the laser beam. It can be made larger than the heat capacity per unit area (1 mm ² ) in 52a.

Moreover, even if the diameter of the laser beam is reduced so that the surface of the electrode terminal 3 is focused on the surface 3a, the effect of spreading the heat obtained by increasing the thickness of the electrode terminal 3 can be increased. Therefore, the temperature distribution on the joint surface 3j can be reduced.

The heat radiating member 6 will be described.
The heat dissipating member 6 is joined to a single or a plurality of insulating substrates 5 with solder 8 and plays a role as a heat dissipating plate, and a heat sink is provided on the connection surface 6b of the heat dissipating member 6 with heat conduction grease or the like. Is a member for efficiently radiating heat generated in the power semiconductor device 101 to the outside. Therefore, the material of the heat radiating member 6 is preferably a metal having a high thermal conductivity, and generally a metal plate of Cu, Al, AlSiC or the like having a thickness of about 1.0 to 5.0 mm is used. The connection surface 6 b of the heat radiating member 6 is a surface facing the joint surface 6 a with the solder 8.

Next, the case 12 will be described.
The case 12 has a frame shape. In the first embodiment, the case 12 is erected on the joint surface 6 a around the heat dissipation member 6 and is fixed to the heat dissipation member 6 with an adhesive 11. Such a case 12 covers the entire power semiconductor device 101, supports the other end of the electrode terminal 3, and also plays a role of connection to an external circuit outside the case 12. The material of the case 12 is generally a thermoplastic plastic material such as PBT (polybutylene terephthalate) or PPS (polyphenylene sulfide), and its melting point is about 300 ° C. or less.

Next, the solder 8, the brazing filler metal 10, and the adhesive 11 will be described.
The solder 8 joins the conductive pattern 52b located on the heat radiation surface side of the insulating substrate 5 and the heat radiation member 6. Therefore, the material of the solder 8 is preferably a metal having a relatively low melting point and a high thermal conductivity. Generally, an alloy containing Sn, Pb, Ag, Cu or the like and having a melting point of less than 450 ° C. is used. Further, the thickness after joining of the solder 8 is preferably about 0.1 to 0.3 mm from the viewpoint of connection reliability and heat dissipation.

The brazing filler metal 10 joins the joint surface 3 j of the electrode terminal 3 and the conductor pattern 52 a of the insulating substrate 5. The brazing filler metal 10 is a heat dissipation path for cooling the electrode terminal 3 that has generated heat by energization, and is also an energization path for electrically connecting the electrode terminal 3 and the conductor pattern 52a. Therefore, the material of the brazing filler metal 10 is preferably a metal having a relatively high melting point and a large thermal conductivity and electrical conductivity. Unlike the solder 8, it contains Au, Ag, Cu, Zn, Ni, etc. An alloy having a temperature of 450 ° C. or higher is used. In the first embodiment, since the electrode terminal 3 and the conductor pattern 52a both use Cu, the brazing filler metal 10 uses a phosphor copper brazing material (Cu—Ag—P) having a melting point of about 800 ° C. ing. Further, the thickness after the joining is generally preferably thinner from the viewpoint of connection reliability, and is preferably 0.25 mm or less. Especially 0.1 mm or less is preferable. Within this range, sufficient results can be obtained in all of the strength of the joint, the electrical resistance of the brazed part, and the thermal resistance of the brazed part.

The adhesive 11 bonds the heat radiating member 6 and the case 12 together. Therefore, generally, an epoxy thermosetting resin is used.

The effect of the power semiconductor device 101 configured as described above will be described. Here, in order to melt the brazing filler metal 10, the surface 3 a facing the joining surface 3 j of the electrode terminal 3 joined to the brazing filler metal 10 is heated by irradiating laser light.
The heat capacity per unit area (1 mm ² ) of the electrode terminal 3 heated by the laser light is larger than the heat capacity per unit area (1 mm ² ) of the conductor pattern 52a facing the electrode terminal 3 heated by the laser light. By enlarging, the temperature difference of the electrode terminal 3 and the conductor pattern 52a by the heating of a laser beam can be made small, and it can prevent that the electrode terminal 3 fuse | melts before brazing. Furthermore, by increasing the thickness of the electrode terminal 3, the heat applied to the surface 3a of the electrode terminal 3 by the laser beam can be spread to the bonding surface 3j on the hard soldering material 10 side. Therefore, the whole joining surface 3j with the brazing filler metal 10 and the conductor pattern 52a can be heated uniformly, and an unjoined portion in the brazing filler metal 10 can be eliminated. Therefore, a brazing area necessary for energization can be secured.

In addition, by increasing the thickness of the electrode terminal 3, the cross-sectional area of the electrode terminal 3 is increased, and a larger current can be passed through one electrode terminal 3. Therefore, the power semiconductor device 101 can be reduced by reducing the number of the electrode terminals 3 for energizing the power semiconductor element or by reducing the width of the electrode terminals 3 by increasing the thickness of the electrode terminals 3. Can be miniaturized.

In the conventional power semiconductor device, when the thickness of the electrode terminal is increased, the stress generated at the junction between the insulating substrate and the electrode terminal increases due to the difference in linear expansion coefficient between the insulating substrate and the electrode terminal. There was a problem that the reliability of the system deteriorated.
However, in the first embodiment, by using the brazing filler metal 10 for joining the electrode terminal 3 and the conductor pattern 52a, the reliability of the joint can be greatly improved as compared with the conventional power semiconductor device. Therefore, in power semiconductor device 101 in the first embodiment, there is no problem with reliability in the range of life generally required for power semiconductor devices.

That is, when the brazing filler metal 10 is used for joining the electrode terminal 3 and the conductor pattern 52a, the mechanical strength of the brazing filler metal 10 is generally the main material of the electrode terminal 3 and the conductor pattern 52a. It is larger than Cu, which is also used in the first embodiment. Therefore, the reliability with respect to the thermal stress caused by the heat generated by the power semiconductor device 101 at the joint between the electrode terminal 3, the brazing filler metal 10 and the conductor pattern 52 a is the conventional power semiconductor device using solder for joining the electrode terminals. Compared with, it can be greatly improved.

In addition, in solder, Sn is the main material, whereas the general brazing filler metal 10 is mainly Cu or Ag. Therefore, since heat conductivity and electrical conductivity are high compared with solder, combined with the above-described effects, the minimum joint area required for the product can be reduced. As a result, the power semiconductor device can be reduced in size because the area necessary for joining the electrode terminals 3 is reduced.

In addition, since the power semiconductor device using SiC for the power semiconductor element 1 operates at a higher temperature, the temperature change of the semiconductor device becomes particularly large on the high temperature side. Therefore, the thermal stress and tensile stress generated in the joint portion are increased, and the decrease in material strength due to high temperature tends to be increased. Therefore, in the power semiconductor device using SiC, the merit by this embodiment becomes more effective.
At this time, it is preferable that the material of the brazing filler metal 10 has a higher melting point in order to improve reliability, and the material of the electrode terminal 3 and the conductor pattern 52a is preferably higher than the melting point of the brazing filler metal 10. Furthermore, in order not to remelt the solder 8 soldered to the insulating substrate 5 as much as possible, the heating time for the electrode terminal 3 is preferably short. However, since the temperature control becomes more difficult as the heating time becomes shorter, it is preferable that the melting point of the brazing filler metal 10, the electrode terminal 3 and the conductor pattern 52a is 250 ° C. or more.

Further, it is necessary to select a material that spreads wet on both the electrode terminal 3 and the conductor pattern 52a for the brazing filler metal 10. Therefore, it is preferable that the material of the electrode terminal 3 and the conductor pattern 52a is the same, and Cu is more preferable as mentioned above.

On the other hand, when different materials are used for the electrode terminal 3 and the conductor pattern 52a, both materials are materials having a melting point higher than the melting point of the hard soldering material 10, and insulation when the conductor pattern 52a is melted. In consideration of breakage of the insulating base material 51 due to local expansion of the base material 51, the conductor pattern 52 a is preferably a material whose melting point is higher than that of the electrode terminal 3. In addition, in order to reduce the temperature difference between the electrode terminal 3 and the conductor pattern 52 a due to heating of the laser light, the conductor pattern 52 a is preferably a material having a lower thermal conductivity than the electrode terminal 3. For example, the conductor pattern 52a is made of Ni plated on the surface of Cu, the material of the electrode terminal 3 is Cu without plating, and the hard soldering material 10 is silver solder, phosphor bronze, or the like.

As described above, according to the power semiconductor device 101 of the first embodiment, the heat capacity per unit area (1 mm ² ) of the electrode terminal 3 heated by the laser light is heated by the laser light. By making it larger than the heat capacity per unit area (1 mm ² ) of the conductor pattern 52a facing the electrode terminal 3, the temperature difference between the electrode terminal 3 and the conductor pattern 52a due to heating of the laser light is reduced, and before brazing. It is possible to prevent the electrode terminal 3 from melting.

Furthermore, by making the thickness of the electrode terminal 3 larger than before, the heat applied to the surface 3a of the electrode terminal 3 by the laser beam can be spread to the bonding surface 3j on the brazing filler metal 10 side. Therefore, the whole joining surface of the brazing filler metal 10 and the conductor pattern 52a can be heated uniformly, and the unjoined part of the brazing filler metal 10 can be eliminated. Therefore, a brazing area necessary for energization can be secured between the electrode terminal 3 and the conductor pattern 52a. As a result, it is possible to obtain a power semiconductor device 101 having a high bonding reliability even in a high temperature operation that secures a brazing area necessary for energizing a large current and prevents the conductor pattern 52a and the insulating base 51 from being damaged. it can.

The applicant uses an electrode terminal 3 whose heat capacity is changed by changing the thickness of the electrode terminal 3, and actually conducts an experiment to compare the wettability of the brazing filler metal 10 and whether or not the electrode terminal 3 is melted. did. In the experiment, three types of electrode terminals 3 having a length of 12 mm, a width of 4 mm, and a thickness of 0.4 mm, 0.8 mm, and 1.2 mm were prepared. As the brazing filler metal 10, a length of 12 mm, a width of 4 mm, A phosphor copper braze having a thickness of 0.13 mm was used. The laser beam is irradiated from the surface 3a side of the electrode terminal 3 sandwiching the sheet-like hard solder material 10, and the focal position of the laser beam is adjusted so that the center of the surface 3a and the center of the laser beam overlap, and the maximum is 4 kW. Irradiation with a fiber laser. After irradiating the laser beam, with respect to the three types of electrode terminals 3, by visual inspection, the presence or absence of melting of the electrode terminals 3, the presence or absence of melting of the brazing filler metal 10, and the presence or absence of bonding of the electrode terminals 3 to the conductor pattern 52a Were compared and examined. The experimental results are shown in FIG.

As shown in FIG. 14, in the case of the electrode terminal 3 having a thickness of 0.4 mm, the electrode terminal 3 was melted before the brazing filler metal 10 was spread over the conductor pattern 52 a, and bonding was not possible. On the other hand, in the case of the electrode terminal 3 having a thickness of 0.8 mm, the surface 3a of the electrode terminal 3 is melted by heating with laser light, but before the entire electrode terminal 3 is melted, the brazing filler metal 10 is melted. Wetting and spreading on the conductor pattern 52a was completed. Furthermore, in the electrode terminal 3 having a thickness of 1.2 mm, the surface 3a of the electrode terminal 3 was not melted by heating with laser light, and the joining was completed.

For the electrode terminal 3 used in this experiment, the ratio of the cross-sectional area of the electrode terminal 3 at the laser light irradiation location to the cross-sectional area of the conductor pattern 52a facing the electrode terminal 3 at this irradiation location is 0 mm in thickness. In the case of 4 mm electrode terminal 3 and conductor pattern 52a, 4: 3, in the case of 0.8 mm thickness electrode terminal 3 and conductor pattern 52a, 8: 3, electrode terminal 3 and conductor 1.2 mm in thickness. In the case of the pattern 52a, it is 4: 1. Therefore, it is preferable that the electrode terminal 3 heated by the laser beam is designed to have a cross-sectional area of at least twice or more, more preferably three times or more with respect to the conductor pattern 52a facing the electrode terminal 3. it is conceivable that.

Embodiment 2. FIG.
FIG. 2 shows a schematic configuration of the power semiconductor device 102 according to the second embodiment in a partial cross-sectional view and a plan view. The power semiconductor device 102 according to the second embodiment also basically has the same configuration as that of the power semiconductor device 101 according to the first embodiment, but differs in the following points. Therefore, here, mainly the differences will be described, and the description of the same components will be omitted. Note that FIG. 2 includes FIGS. 2A to 2H, and shows only the joint portion of the electrode terminal 3 in the power semiconductor device 102, and the other components are not shown.

In Embodiment 1, in order to make the heat capacity of the electrode terminal 3 larger than that of the conductor pattern 52a, the thickness of the electrode terminal 3 is made larger than the thickness of the conductor pattern 52a.
On the other hand, in the second embodiment, as shown in FIGS. 2A and 2B, the electrode terminal 3 is joined to the conductor pattern 52a with the brazing filler metal 10 and has a joint portion 31 having a first cross-sectional area, The second embodiment is different from the first embodiment in that it is configured to have a non-joining portion 32 having a second sectional area larger than one sectional area. This will be described in detail below.

Note that, as shown in FIG. 2A, the joint portion 31 and the non-joint portion 32 can be bent and oriented, but the directions of the surfaces representing the first cross-sectional area and the second cross-sectional area are different. That is, the first cross-sectional area in the joint portion 31 is the area in the cross section of the electrode terminal 3 in the plane perpendicular to the x direction (that is, the yz plane) as in the case of the first embodiment. The second cross-sectional area in the portion 32 is an area in the cross section of the electrode terminal 3 in a plane perpendicular to the z direction (that is, the xy plane).

As described in the first embodiment, in order to reduce the temperature difference between the electrode terminal 3 and the conductor pattern 52a due to the heating of the laser beam, that is, in order to make the heat capacity of the electrode terminal 3 larger than the heat capacity of the conductor pattern 52a. It has been stated that the sectional area of the electrode terminal 3 is increased by increasing the thickness of the electrode terminal 3 than the width of the electrode terminal 3.
On the other hand, when the thickness of the electrode terminal 3 is excessively increased, a temperature difference is generated between the surface 3a of the electrode terminal 3 heated by the laser beam and the bonding surface 3j, and before the hard soldering material 10 is melted. Further, there is a possibility that the surface 3a of the electrode terminal 3 is melted.
Therefore, a portion of the electrode terminal 3 that has the joint surface 3j and is joined to the conductor pattern 52a is a joint 31 as shown in FIG. 2A, and a cross-sectional area at the joint 31 is a first cross-sectional area. Further, a portion that is located close to the joint surface 3j and that is not joined to the conductor pattern 52a is defined as a non-joint portion 32, and the non-joint portion 32 is configured to have a second cross-sectional area larger than the first cross-sectional area. Thus, the heat capacity of the part adjacent to the part heated by the laser beam of the electrode terminal 3 can be made larger than before.

With this configuration, the cross-sectional area of the heat applied by the laser beam is increased without increasing the width of the electrode terminal 3 (the dimension of the electrode terminal 3 in the y direction shown in FIGS. 2A and 2B). It is possible to efficiently escape to the non-joining portion 32. Therefore, it is more preferable that the power capacity of the electrode terminal 3 can be made larger than the heat capacity of the conductor pattern 52a while the power semiconductor device 102 is downsized.
Further, the same effect can be obtained even if the width of the electrode terminal 3 is increased in the non-joint portion 32 instead of the thickness. If the width of the electrode terminal 3 is increased as long as it is in the vicinity of the joint surface 3j of the electrode terminal 3, the power semiconductor device 102 is not hindered in size reduction.

Further, as shown in FIGS. 2C and 2D, a hard brazing material 10 may be brazed in advance in a brazing furnace or the like to the joint surface 3j of the electrode terminal 3.
By comprising in this way, the heat capacity of the electrode terminal 3 can be increased by the amount brazed previously. In addition, the temperature rise of the electrode terminal 3 due to the heating of the laser light can be suppressed by the latent heat of melting when the brazing filler metal 10 is melted by the heating of the laser light.
Furthermore, the interface where the contact thermal resistance between the electrode terminal 3 and the conductor pattern 52a occurs can be reduced by one, and the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be reduced, which is more preferable. .

Similarly, as shown in FIGS. 2E and 2F, contact thermal resistance between the electrode terminal 3 and the conductor pattern 52a is also generated by brazing the hard soldering material 10 to the conductor pattern 52a in advance. The effect of reducing the interface can be obtained.

Furthermore, as shown in FIGS. 2G and 2H, a material having a melting point lower than that of the electrode terminal 3 may be preliminarily applied in the vicinity of the joint surface 3j of the electrode terminal 3 by plating 3m or the like.
With this configuration, the heat capacity of the electrode terminal 3 can be increased by the amount of the applied material, and when the temperature of the electrode terminal 3 rises due to heating with laser light, the applied material becomes the electrode terminal. Since the melting point reaches the melting point before 3, the temperature rise of the electrode terminal 3 can be suppressed by the latent heat of fusion.

Embodiment 3 FIG.
In the third embodiment, the method for manufacturing the power semiconductor device 101 in the first embodiment will be described with reference to FIGS. 3 (FIGS. 3A to 3C), 4 (FIGS. 4A to 4C), and 5 (FIGS. 5A and 5B). ) To explain. Here, FIG. 3 shows a method for manufacturing the power semiconductor device 101 in the first embodiment, and FIGS. 4 and 5 are diagrams for supplementing the description of the method for manufacturing the power semiconductor device 101.

First, a method for manufacturing the power semiconductor devices 100A and 100B will be described with reference to FIGS.
As explained at the beginning of this specification, in order to satisfy the required reliability as the temperature environment in which the power semiconductor device is used has become severe in recent years, a hard solder having a melting point higher than that of solder is used. A method of joining electrode terminals using a material is conceivable. In describing the bonding method using the brazing filler metal, first, referring to FIG. 4, power semiconductor device 100 </ b> A having a configuration similar to that of power semiconductor device 101 in the first embodiment is used in insulating substrate 5. The case where the electrode terminal 3 is joined to the conductor pattern 52a using solder will be described. Next, with reference to FIG. 5, the case where the electrode terminal 3 is joined to the conductor pattern 52a on the insulating substrate 5 using the brazing filler metal 10 will be described.

In power semiconductor device 100A, as shown in FIG. 4A, solder 9 of the same alloy as solder 8 is used as a material for joining electrode terminal 3 and conductor pattern 52a.
As shown in FIG. 4A, solder 9 is placed on or supplied to the conductor pattern 52a to start soldering. At this time, as shown in FIG. 4B, the entire power semiconductor device 100A being soldered is heated from the bottom surface of the heat radiating member 6 by the hot plate 40 or the like, and from the surface 3a of the electrode terminal 3 by the hot tool 41 or the like. Heated and pressurized. As shown in FIG. 4C, the solder 9 is melted to join the electrode terminal 3 and the conductor pattern 52a.

As the solder 9, a solder having a lower melting point than that of the solder 8 is generally selected. Note that the melting points of the solder 8 and the solder 9 are both about 250 ° C. or less. Therefore, the heating temperature of the hot plate 40 is set to be equal to or lower than the melting point of the solder 8, and the solder 8 and the case 12 are not melted by the heating of the hot plate 40. Further, when the entire power semiconductor device 100A is heated from the bottom surface side of the heat dissipation member 6 to the vicinity of the melting point of the solder 9 by a hot plate, and is locally heated from the surface 3a of the electrode terminal 3 by the hot tool 41. The solder 9 easily reaches the melting point and melts. At this time, since the conductor pattern 52a side also reaches a temperature equal to or higher than the melting point of the solder 9 due to heating by the hot plate 40, the molten solder 9 spreads wet to the conductor pattern 52a side and forms a fillet.

However, due to the configuration as described above, the melting point of the solder 9 must be lowered, and it is difficult to cope with the severe temperature environment in which the power semiconductor device is used. Therefore, it is conceivable that the power semiconductor device 100A configured using the solder 9 cannot satisfy the reliability required for the product.

On the other hand, in the power semiconductor device 100B, as shown in FIG. 5A, a hard brazing material 10 is used instead of the solder 9. In this case, in order to melt the brazing filler metal 10 by the heating method described with reference to FIG. 4, (i) the temperature of the hot plate 40 is set near the melting point of the brazing filler metal 10 (800 in the third embodiment). (Ii) Use a hot tool 41 that can be heated to the melting point of the brazing filler metal 10 or higher, or (iii) Heat locally from the surface 3a of the electrode terminal 3 to the melting point of the brazing filler metal 10 or higher. It is necessary to use possible heating means.

However, in the method (i), since the entire power semiconductor device 100B is heated to, for example, about 800 ° C., a phenomenon such as melting of the case 12 and the solder 8 occurs, and functions necessary for the power semiconductor device are achieved. The problem of losing arises. In the method (ii), although melting of the case 12 can be prevented, the heating by the hot tool 41 is a heat transfer by point contact between the surface 3a of the electrode terminal 3 and the hot tool 41, and thus heating takes a long time. . As a result, the heating proceeds to the whole of the solder 8, and the solder 8 may be melted to cause displacement of the insulating substrate 5.

Therefore, in order to join the electrode terminal 3 to the conductor pattern 52a on the insulating substrate 5 with the brazing filler metal 10, the method (iii), that is, the surface of the electrode terminal 3 is used so as not to melt the case 12 and the solder 8. It is necessary to use a heating method capable of locally heating to a temperature equal to or higher than the melting point of the brazing filler metal 10 in a short time of about several seconds with respect to 3a. As such a heating method, there is a method using friction stirring or ultrasonic waves, but a heating method using laser light can be employed.

This heating by laser light does not cause mechanical damage such as pressurization or vibration to the insulating substrate of the semiconductor device, does not impair the insulation and conductivity required for the power semiconductor device, and can be handled in a small space. In addition, since it can be used in air, it is preferable. As a result, the brazing filler metal 10 is melted and brazed, and a material harder than the electrode terminal 3 is used for the brazing filler metal 10 as in the phosphor copper brazing material used in the first embodiment, whereby the electrode terminal 3 A more reliable joint can be obtained.

However, the conductor pattern 52a to be joined to the electrode terminal 3 that is heated by laser light is formed on the insulating substrate 5. In a general power semiconductor device, the insulating substrate 5 has a heat dissipation structure for efficiently cooling the semiconductor elements. In addition, the area of the insulating substrate 5 is compared with the area of the bonding surface 3j of the electrode terminal 3. In addition, the heat dissipation member 6 is connected. Therefore, the heat capacity of the insulating substrate 5 is much larger than that of the electrode terminal 3, and even when the surface 3a of the electrode terminal 3 is locally heated using laser light, the electrode terminal 3, Due to the contact thermal resistance between the brazing filler metal 10 and the conductor pattern 52a, a temperature difference occurs between them. Therefore, even if the brazing filler metal 10 is melted by heating, the temperature of the conductor pattern 52a does not reach the melting point of the brazing filler metal 10, and the brazing filler metal 10 is bonded to the electrode terminal 3 as shown in FIG. 5B. It becomes wet only on the 3j side and does not get wet on the conductor pattern 52a side. Furthermore, even if heating from the bottom surface side of the heat radiating member 6 is simultaneously performed using the hot plate 40, the above temperature difference is compensated by heating below the melting point of the solder 8 and the case 12 (about 200 ° C. or less). Is difficult. In FIG. 5B, reference numeral “42” denotes a laser device that generates and emits laser light.

Further, oxidation of the surface of the conductor pattern 52a proceeds due to heating, and there is a concern that the brazing property is lowered. If the heating by the laser beam is continued as it is, the temperature of the electrode terminal 3 rises too much before the conductor pattern 52a reaches the temperature at which the brazing filler metal 10 gets wet, and the electrode terminal 3 can reach its melting point and melt. There is sex. As the electrode terminal 3 melts, the absorption rate of the laser light is significantly increased, and the temperature rise of the electrode terminal 3 is further increased. As a result, in the laser light irradiation region, the entire electrode terminal 3 in the thickness direction is melted, and the laser light may be in a state similar to general laser light welding in which the surface of the conductor pattern 52a is directly melted. There is.

When the laser beam directly melts the conductor pattern 52a as in general laser beam welding, the laser beam penetrates the conductor pattern 52a and reaches the insulating base material 51 of the insulating substrate 5, and the insulating base material 51 May cause damage. Even when the laser beam does not penetrate the conductor pattern 52a, the melted portion of the conductor pattern 52a or the region in the insulating substrate 5 heated by the laser beam locally expands in a short time. Therefore, local stress is generated in the insulating base material 51, which may damage the insulating base material 51.
In general, in a power semiconductor device, a current flowing through a semiconductor element during operation flows through a conductor pattern 52a from an electrode terminal 3 through a junction. Since the conductor pattern 52a is generally formed integrally with the insulating base 51, damage to the insulating base 51 makes it difficult to ensure insulation of the power semiconductor device.

In addition, since the joint portion obtained by general laser beam welding is limited to the laser light irradiation region and its periphery, the joint area is smaller than the joint portion obtained by melting the brazing filler metal 10. Become. The junction between the electrode terminal 3 of the power semiconductor device 100B and the conductor pattern 52a is an energization path for energizing the semiconductor element of the power semiconductor device 100B, particularly in a power semiconductor device that handles a large current. In order to efficiently flow a large current, it is necessary to join the electrode terminals 3 in a large area. Therefore, when the bonding area between the electrode terminal 3 and the conductor pattern 52a is small, the bonding area required for energization is not ensured, and stable power supply is hindered. Furthermore, since the electric resistance is locally increased and the heat dissipation path of the electrode terminal 3 is also reduced, the temperature of the electrode terminal 3 is increased, and the reliability of the joint portion is lowered.

Therefore, when the electrode terminal 3 is locally heated and brazed using a laser beam, the cross-sectional area of the electrode terminal 3 is preferably larger as already described in the first embodiment. It is preferable to make the thickness thicker than the width of the electrode terminal 3.

Based on the above description, particularly with reference to FIG. 5, referring to FIG. 3, the power semiconductor devices 101 and 102 having the electrode terminals 3 having a larger heat capacity than the conductor pattern 52 a in the third embodiment. The manufacturing method, specifically, the brazing process using laser light will be described below. The power semiconductor device 101 is taken as an example. 3 is a control device that controls the operation of the laser device 42.

Here, the control device 43 is actually realized using a computer, and includes software corresponding to the operation control function of the laser device 42 and hardware such as a CPU (Central Processing Unit) and a memory for executing the software. It is configured.

First, as shown in FIG. 3A, the electrode terminal 3 is placed with the brazing filler metal 10 sandwiched between the conductor pattern 52 a so that the joint surface 3 j of the electrode terminal 3 is in contact with the brazing filler metal 10. Then, the surface 3a of the electrode terminal 3 and the conductor pattern 52a are irradiated with laser light to heat them.
At this time, the laser beam is irradiated to the entire surface 3a facing the bonding surface 3j of the electrode terminal 3 by defocusing or the like so that the entire bonding surface 3j can be heated uniformly by the laser beam. Is preferred. This operation is executed by controlling the laser device 42 by the control device 43.
Each surface of the electrode terminal 3, the conductor pattern 52a, and the brazing filler metal 10 is covered with an oxide film. At this time, the temperature of the brazing filler metal 10 has not risen to its melting point, and the electrode terminal 3 and the conductor pattern 52a are not joined by the hard soldering material 10.

Next, as shown in FIG. 3B, the electrode terminal 3 is heated by further irradiating a laser beam in a range including at least the surface 3 a of the electrode terminal 3. The brazing filler metal 10 is indirectly heated by the heating of the electrode terminal 3, and the brazing filler metal 10 reaches the melting point and melts. At this time, the temperature of the surface of the conductor pattern 52a approaches the temperature at which the brazing filler metal 10 spreads out, but has not yet reached the temperature at which the brazing filler metal 10 can spread out. Therefore, at this time, the molten brazing filler metal 10 wets and spreads only on the joint surface 3j of the electrode terminal 3. Further, since the latent heat of fusion exists when the brazing filler metal 10 is melted, the temperature rise of the electrode terminal 3 due to the heating of the laser beam is slowed during this time.

In the power semiconductor device 101, Cu is used for the electrode terminal 3 and the conductor pattern 52 a, and phosphorous copper solder is used for the hard solder material 10. Therefore, when the brazing filler metal 10 is melted, the bonding surface 3j of the electrode terminal 3 is reduced by the reducing action of phosphorus contained in the phosphor copper brazing, and the brazing filler metal 10 wets and spreads on the electrode terminal 3 even without flux. On the other hand, when a brazing material that requires flux is used, it is necessary to apply the flux to the portions to be joined at the time of the step shown in FIG. 3A. Further, the melting point of Cu is higher by about 300 ° C. than the melting point of the brazing filler metal 10, and a large difference in melting point from the brazing filler metal 10 can be obtained. Therefore, when Cu is used for the electrode terminal 3, the merit in this embodiment becomes more effective.

Furthermore, by continuing the heating with the laser beam, as shown in FIG. 3C, the temperature of the surface of the conductor pattern 52a is increased to the temperature at which the brazing filler metal 10 spreads out before the electrode terminal 3 is melted. Then, the brazing filler metal 10 can be spread and spread on the conductor pattern 52a side.

The temperature of the conductor pattern 52a rises most in the portion immediately below the range heated by the laser beam. Therefore, the brazing filler metal 10 is wetted from the portion immediately below to the conductor pattern 52a side, and as the heating is continued, the wetted area spreads outside the surface of the conductor pattern 52a that faces the joint surface 3j. Eventually, after the brazing filler metal 10 wets and spreads over the entire surface facing the joint surface 3j, a wet spread and fillet is formed further outside.

In the third embodiment, similarly to the electrode terminal 3, when the brazing filler metal 10 is wet and spreads, the conductor pattern 52a is reduced by the reducing action of phosphorus contained in the phosphor copper brazing, so that the hard pattern is hard even without flux. The brazing material 10 spreads over the conductor pattern 52a. At this time, the laser beam is irradiated on the entire surface 3a of the electrode terminal 3 by defocusing or the like so that the entire bonding surface 3j can be uniformly heated by the laser beam. Thereby, the temperature of the surface which opposes the joint surface 3j in the conductor pattern 52a rises equally. Therefore, when the brazing filler metal 10 gets wet on the conductor pattern 52a side, the brazing filler metal 10 spreads almost simultaneously on the entire facing surface of the bonding surface 3j in the conductor pattern 52a. Therefore, the time required to wet and spread the brazing filler metal 10 on the conductor pattern 52a side can be shortened, and the electrode terminal 3 is not locally heated. Therefore, the temperature difference between the electrode terminal 3 and the conductor pattern 52a is reduced. Can be small.

As a heating method for the surface 3 a of the electrode terminal 3, electromagnetic waves other than laser light can be used. Even when other electromagnetic waves are used, the effect of reducing the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be obtained.
In particular, since the electron beam can be finely distributed in the heated portion, the temperature control for heating the bonding surface 3j of the electrode terminal 3 and the surface facing the bonding surface 3j in the conductor pattern 52a to the same level is possible. Since it becomes easy and can be stably heated regardless of the surface state of the electrode terminal 3 and the conductor pattern 52a, it is preferable in the same manner as the laser beam.

Further, since the thickness of the electrode terminal 3 is increased in order to increase the cross-sectional area, the thickness of the electrode terminal 3 is obtained even when the entire surface 3a of the electrode terminal 3 is not irradiated with laser light. Heat spreads in the direction. Therefore, compared with the case where the thickness of the electrode terminal 3 is relatively thin, the effect of reducing the temperature difference of the conductor pattern 52a can be obtained.

Embodiment 4 FIG.
FIG. 6 shows a schematic configuration of the power semiconductor device 104 according to the fourth embodiment in a partial cross-sectional view and a plan view. The power semiconductor device 104 in the fourth embodiment also basically has the same configuration as that of the power semiconductor device 101 in the first embodiment, but differs in the following points. Therefore, here, mainly the differences will be described, and the description of the same components will be omitted. Note that FIG. 6 includes FIGS. 6A to 6B, and illustrates only the joint portion of the electrode terminal 3 in the power semiconductor device 104, and the other components are not illustrated.

In the first embodiment, in order to reduce the temperature difference between the electrode terminal 3 and the conductor pattern 52a due to heating of the laser beam, the electrode terminal 3 is configured to have a thickness larger than the thickness of the conductor pattern 52a. The heat capacity per unit area was made larger than that of the conductor pattern 52a.
On the other hand, in the fourth embodiment, the electrode terminal 3 is directly heated by a laser beam and includes a joining surface 3j joined to the conductor pattern 52a by the brazing filler metal 10 as shown in FIGS. 6A and 6B. The present embodiment is different from the first embodiment in that it is configured to have a through hole 33 in a part of the surface. This will be described in detail below. In addition, the through hole 33 corresponds to an example of a defective portion in which the electrode terminal 3 is lost in the thickness direction of the electrode terminal 3.

As described in the first embodiment, when the electrode terminal 3 is heated by laser light, in order to prevent the electrode terminal 3 from melting before brazing, the temperature difference between the electrode terminal 3 and the conductor pattern 52a is set. It is necessary to make it smaller. Therefore, in the first embodiment, the cross-sectional area of the electrode terminal 3 is increased by making the thickness larger than the width of the electrode terminal 3.
On the other hand, as described in the second embodiment, when the thickness of the electrode terminal 3 is excessively increased, a temperature difference is generated between the surface 3a of the electrode terminal 3 heated by the laser beam and the bonding surface 3j. Therefore, there is a possibility that the surface 3a of the electrode terminal 3 is melted before the brazing filler metal 10 is melted.

Therefore, as shown in FIG. 7, the electrode terminal 3 is configured to have a through-hole 33 in a part of the portion directly heated by the laser beam. By providing the through hole 33, it is possible to directly heat the conductor pattern 52a together with the electrode terminal 3 by laser light, so that the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be reduced.

Therefore, the electrode terminal 3 and the conductor pattern 52a can be evenly heated by the heat given by the laser beam. Therefore, the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be further reduced as compared with the case of the above-described embodiment while the power semiconductor device 104 is downsized. This is because, as described in the first embodiment, when considering the reliability of the power semiconductor device, the minimum required junction area can be reduced, so that the effect of reducing the size of the power semiconductor device 104 is maintained. This means that the effect of reducing the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be obtained.

Also, as described in the third embodiment, even when the laser beam is defocused, the temperature is likely to increase as the laser beam is focused. Therefore, it is more preferable that the conductor pattern 52a can be efficiently heated by causing the laser beam focal point and the through-hole 33 to coincide with each other or close to each other.

This will be described in detail below.
As shown in FIG. 8A, even in the case of defocused laser light, the vicinity of the optical axis of the laser light has the highest energy density and is easily heated. Further, in the case of the electrode terminal 3 having no through-hole 33, the surface of the conductor pattern 52a is not directly irradiated with laser light. Therefore, the temperature rise of the conductor pattern 52a is caused by the heat of the surface 3a of the electrode terminal 3 heated by the laser beam between the electrode terminal 3 and the brazing filler metal 10 where the thermal contact resistance exists. It is necessary to wait for conduction at two places between the patterns 52a. Therefore, the temperature rise of the conductor pattern 52 a is smaller than that of the electrode terminal 3.

On the other hand, when the through hole 33 is provided in the electrode terminal 3, as shown in FIG. 8B, the surface of the conductor pattern 52a can be directly heated by the laser light passing through the through hole 33. Both the surface 3a and the surface of the conductor pattern 52a are directly heated. Therefore, the surface temperature of the conductor pattern 52a tends to rise. In addition, when the through hole 33 is provided in the central portion of the electrode terminal 3 (the central portion of the bonding surface 3j), the surface of the conductor pattern 52a can be directly heated at the portion where the energy density of the laser beam is highest, Furthermore, the surface temperature of the conductor pattern 52a can be increased. In addition, since the surface 3a of the electrode terminal 3 deviates from the focal position of the laser beam, the temperature rise of the electrode terminal 3 is delayed compared to the electrode terminal 3 that does not have the through hole 33. Therefore, the temperature difference between the conductor pattern 52a and the electrode terminal 3 can be reduced.

Accordingly, (i) the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be reduced. (Ii) Since the heat heated from the center of the joint surface 3j spreads over the entire joint surface 3j, the surface of the conductor pattern 52a. The effect that the temperature difference in the bonding surface 3j can be suppressed is obtained.

Further, as shown in FIG. 9, for example, by irradiating laser light in a spiral shape from the center of the joint surface 3 j to the periphery, the temperature does not deviate from the center of the joint surface 3 j toward the outside of the electrode terminal 3. It is preferable because it can be heated.

Further, as described in the third embodiment, in order to join the electrode terminal 3 to the conductor pattern 52a with the brazing filler metal 10, the surface 3a of the electrode terminal 3 is prevented from melting the case 12 and the solder 8. On the other hand, it is necessary to locally heat to a temperature equal to or higher than the melting point of the brazing filler metal 10 in a short time of about several seconds.
For this reason, in the joining using the brazing filler metal 10, unlike the manufacturing method using the solder 9 as shown in FIG. 4, the void below the joining surface 3 j is obtained by scrubbing the electrode terminal 3 in the joining process. Reduction measures cannot be implemented. Therefore, there is a concern that the void 34 remains below the bonding surface 3j after bonding, as shown in FIG.

Since the brazing filler metal 10 is used, even if the void 34 remains, the reliability of the power semiconductor device 104 does not decrease. However, since the brazing filler metal 10 is not present in the portion where the void 34 exists, the bonding area is small. Become. For this reason, the energization path and the heat dissipation path between the electrode terminal 3 and the conductor pattern 52a are reduced, and therefore it is preferable that there is no void 34.

Therefore, by providing a through hole 33 having a certain size through which gas can pass in a plane including the bonding surface 3j of the electrode terminal 3, the vicinity of the center of the bonding surface 3j of the electrode terminal 3 when the brazing filler metal 10 is melted. Thus, the gas remaining in the brazing filler metal 10 can be discharged to the outside through the through hole 33. Therefore, the presence of the void 34 below the joint surface 3j can be reduced.

The applicant conducted an experiment in which the electrode terminal 3 not provided with the through-hole 33 and the electrode terminal 3 provided with the through-hole 33 were respectively irradiated with laser light for brazing. Each experimental state is shown in FIG. 10 and FIG. 6A. In the experiment, the electrode terminal 3 having a length of 12 mm, a width of 4 mm, and a thickness of 0.8 mm was used, and the through hole 33 was provided in the center of the electrode terminal 3 with a size of φ2 mm. As a laser beam, a laser was irradiated with a fiber laser of a maximum of 4 kW so that the center of the through hole 33 of the electrode terminal 3 overlapped with the center of the laser beam.

As a result of the experiment, a void 34 was present at the center of the electrode terminal 3 at the brazed portion of the electrode terminal 3 where the through-hole 33 was not provided, as shown in FIG. On the other hand, in the electrode terminal 3 provided with the through hole 33, as shown in FIG. 6A, the hard brazing filler metal 10 spreads in the through hole 33, and no void was observed. From the experimental results, it was found that by providing the through holes 33 in the electrode terminals 3, the voids 34 below the joint surface 3j can actually be reduced. Furthermore, by providing the through hole 33 at the focal position of the laser beam, the phenomenon that the temperature of the surface 3a of the electrode terminal 3 is excessively increased and the surface 3a is melted was not observed. Thus, it was found that the temperature difference between the electrode terminal 3 and the conductor pattern 52a was reduced by providing the through hole 33 in the electrode terminal 3.

Furthermore, when the brazing filler metal 10 is supplied in the same manner as in the case of the electrode terminal 3 having no through hole 33, the electrode terminal 3 having the through hole 33 enters the through hole 33 as shown in FIG. 6A. The amount of the brazing filler metal 10 to be reduced is reduced. Accordingly, the cross-sectional area of the electrode terminal 3 is reduced by the amount of the through hole 33, and the electrical resistance of the electrode terminal 3 is increased. Furthermore, since the heat of the electrode terminal 3 does not spread in the through-hole 33 part, the state where it does not become a heat dissipation path of the electrode terminal 3 occurs.
In order to prevent such a joined state that the through-hole 33 portion becomes an unjoined portion and the energization path and the heat dissipation path become small, the brazing filler metal 10 may be supplied in excess. In this case, the extra brazing filler metal 10 can enter the through hole 33 by having the through hole 33. As a result, it is possible to eliminate the joint state that is caused by providing the through hole 33 in the electrode terminal 3 and locally increases in electrical resistance and decreases in the heat dissipation path.

Moreover, generally, the brazing filler metal 10 becomes more advantageous as the bonding reliability becomes thinner. By providing the through-hole 33, the molten brazing filler metal 10 wets and spreads inside the through-hole 33, and the thickness of the brazing filler metal 10 below the electrode terminal 3 is set so that the brazing filler metal 10 wetted and spreads inside the through-hole 33. Thinner than the thickness. Therefore, it is possible to obtain an effect of improving the bonding reliability.

Further, the brazing filler metal 10 may not be supplied into the through hole 33. As a technique for that purpose, a hole having the same size as or larger than the through hole 33 is provided in the brazing filler metal 10, and the brazing filler metal 10 is arranged so that the center of the provided hole is aligned with the center of the through hole 33. That is, the hard brazing material 10 is arranged so as not to be seen when viewed from the laser beam irradiation side during assembly. By doing so, the surface of the conductor pattern 52a can be reliably heated with laser light, so that the temperature difference between the electrode terminal 3 and the conductor pattern 52a can be reliably reduced.
By comprising in this way, the electrode terminal 3 and the conductor pattern 52a can be heated more uniformly with a laser beam.

On the other hand, when the brazing filler metal 10 is supplied into the through-hole 33 in advance, it operates as follows. That is, the brazing filler metal 10 is first melted by the heating of the laser beam. Until the conductive pattern 52a reaches the melting point of the brazing filler metal 10, the molten brazing filler metal 10 does not wet and spread on the conductive pattern 52a, but gathers around the through holes 33 due to the surface tension of the brazing filler metal 10. Therefore, heating of the conductor pattern 52a by laser light is not hindered. In addition, when the surface temperature of the conductor pattern 52a reaches the melting point of the brazing filler metal 10, the brazing filler metal 10 can be prevented from spreading and getting into the through-hole 33, and reducing the energization path and the heat dissipation path. . Further, the brazing filler metal 10 is melted during heating, and it is more preferable because it does not prevent the voiding gas from coming out of the through hole 33.

Further, a plurality of through holes 33 may be provided in the electrode terminal 3 as shown in FIG. 11A. By providing a plurality, it is more preferable that the through holes 33 can be evenly arranged in the joint surface 3j, and the voids 34 in the joint portion 31 can be more efficiently reduced. At this time, if the focal position of the laser beam is deviated, the temperature is also deviated, so that the focal position of the laser beam is preferably at the center of the bonding surface 3j from the viewpoint of uniform temperature in the bonding surface 3j. Therefore, when a plurality of through holes 33 are provided in the electrode terminal 3, it is preferable to provide the through holes 33 evenly in a concentric manner with the focal point as the center.

Furthermore, as shown in FIG. 11B, the through hole 33 may be formed in a groove shape instead of a hole. In addition, this groove | channel is corresponded to an example of the defect | deletion part which lost the electrode terminal 3 in the thickness direction of the electrode terminal 3. FIG. The groove extends from the tip 3c of the electrode terminal 3 (FIG. 13) toward the bent portion of the electrode terminal 3. Therefore, the tip 3c of the electrode terminal 3 is divided by the groove.
In this case, if the distance (groove width) between both walls along the extending direction of the groove is not more than twice the distance of the fillet formed by the brazing filler metal 10, the molten brazing filler metal 10 gets wet in the groove. Can spread. Therefore, both the effect of reducing the temperature difference between the electrode terminal 3 and the conductor pattern 52a and the effect of reducing the void 34 in the joint surface 3j can be obtained without reducing the energization path and the heat dissipation path.

Further, by reducing the groove width, even when the supply of the brazing filler metal 10 is biased and the brazing filler metal 10 is not supplied to the tip of the groove, the brazing filler metal 10 is wetted to the tip of the groove by capillary action. spread. Therefore, the brazing filler metal 10 can be uniformly spread in the joining surface regardless of the supply position of the brazing filler metal 10 or the variation in the supply amount.

Furthermore, as shown in FIG. 12A, if the side wall of the through-hole 33 of the electrode terminal 3 is a square shape, the temperature at the root 33a side (the side in contact with the brazing filler metal 10) of the square shape is This is considered to be lower than the tip 33b side of the C shape (the surface 3a side of the electrode terminal 3). Therefore, it is possible to obtain an effect of suppressing creeping of the brazing filler metal 10 (a state in which the molten brazing filler metal 10 is wet and spreads to the surface 3a side facing the bonding surface 3j of the electrode terminal 3 through the wall surface of the electrode terminal 3). Therefore, it is preferable. This is because, when the temperature on the electrode terminal 3 side rises faster than the conductor pattern 52a side, the direction in which the brazing filler metal 10 spreads out is only toward the electrode terminal 3, so that it tends to creep up. This is because the temperature on the side in contact with the brazing filler metal 10 is the surface of the electrode terminal 3 even if the reverse staircase shape shown in FIG. 12B (the side of the electrode terminal 3 in contact with the brazing filler metal 10 has a counterbore shape). Since it is thought that it becomes low compared with 3a, the effect which suppresses the creeping to the electrode terminal 3 side of the same brazing filler metal 10 can be acquired.
Moreover, the effect which becomes easy to remove the void in the hard soldering | brazing | wax material 10 can be acquired by any of a square shape and a reverse staircase shape. In addition, since the wall surface of the through-hole 33 is an inclined surface and there is no level | step difference, the effect of void removal is larger in the C shape.

In the conventional method using soldering, soldering is possible even without applying pressure to the extent that the electrode terminal 3 is deformed. On the other hand, when the brazing filler metal 10 is used as in the present embodiment, the conductor pattern is reduced in order to reduce the contact thermal resistance between the electrode terminal 3 and the brazing filler metal 10 and between the brazing filler metal 10 and the conductor pattern 52a. It is necessary to pressurize the electrode terminal 3 against 52a. At this time, when the electrode terminal 3 is irradiated with laser light with the tip 3c and the root 3d of the electrode terminal 3 shown in FIG. The electrode terminal 3 is deformed, and the deformation remains after brazing.
Since the deformation is larger as the tip 3c of the electrode terminal 3 is thinner, the deformation can be suppressed by making the electrode terminal 3 thicker. Moreover, by providing the electrode terminal 3 with the through hole 33 and setting the focal position of the laser beam to the position of the through hole 33, the heating temperature of the electrode terminal 3 can be suppressed as described above, and the electrode terminal by heating The deformation of 3 can be similarly suppressed.

In the embodiment in which the surface 3a of the electrode terminal 3 and the conductor pattern 52a are simultaneously heated with laser light as described in the embodiment 1-4, the arrangement of the electrode terminal 3 and the conductor pattern 52a is as follows. It is preferable that the relationship is as follows.
That is, for example, as shown in FIG. 2A, in a form in which the non-joining portion 32 is bent with respect to the joining portion 31 of the electrode terminal 3 and extends along the z direction, as shown in FIG. When the laser beam is irradiated along the z direction with respect to the surface 3a, a part of the irradiated laser beam is also irradiated to the non-joint portion 32. Therefore, in the electrode terminal 3, the lower portion 3 e (FIG. 15A) of the non-joining portion 32 is blocked by the non-joining portion 32, so that the lower portion 3 e is not heated when viewed very strictly. It may be enough. On the other hand, the joint portions 31 other than the lower portion 3e in the electrode terminal 3 are heated to a temperature equal to or higher than the melting point of the brazing filler metal 10 because they are heated by the laser beam. Since the molten brazing filler metal 10 wets and spreads in the portion heated to the melting point of the brazing filler metal 10 or higher, the brazing filler metal 10 wets and spreads in most parts of the joint portion 31, but in the conductor pattern 52a, the lower portion 3e. In the part corresponding to, the brazing filler metal 10 does not spread and fillets may not be formed. In such a case, as shown in FIG. 15A, in the lower portion 3e, the brazing filler metal 10 does not spread to the outside of the electrode terminal 3, and the contact angle (wetting angle) between the brazing filler metal 10 and the electrode terminal 3 is 90. A fillet shape opposite to the tip 3c of the electrode terminal 3 is formed. That is, in the lower part 3 e of the electrode terminal 3 corresponding to the bent part of the electrode terminal 3, the contact angle between the brazing filler metal 10 and the electrode terminal 3 is the contact angle in the part other than the lower part 3 e of the electrode terminal 3. Smaller than

Using such a situation, as shown in FIG. 15A, the lower portion 3e is oriented toward the end 52aE side of the conductor pattern 52a, thereby making it possible to prevent the insulating base material 51 from being damaged. This will be described in detail below.

For example, the end surface 3cE of the tip 3c of the electrode terminal 3 and the end 52aE of the conductor pattern 52a are arranged at the same position in the x direction in the direction opposite to that shown in FIG. When the laser beam is irradiated, the tip 3c is irradiated with the laser beam and heated. Therefore, the brazing filler metal 10 wets and spreads to the end surface 3cE of the tip 3c and the end 52aE of the conductor pattern 52a. As a result, the conductor pattern 52a is pulled by the difference in linear expansion coefficient between the insulating substrate 5 and the brazing filler metal 10, and the insulating base material 51 may be damaged from the end 52aE of the conductor pattern 52a.

On the other hand, as shown in FIG. 15A, by orienting the lower portion 3e of the electrode terminal 3 on the end 52aE side of the conductor pattern 52a, the heating of the lower portion 3e is suppressed as described above. Therefore, the temperature on the end 52aE side of the conductor pattern 52a does not rise to the melting point of the brazing filler metal 10, and the brazing filler metal 10 does not spread to the end of the conductor pattern 52a. Therefore, it is possible to prevent the insulating base material 51 from being damaged due to a difference in linear expansion coefficient between the insulating substrate 5 and the brazing filler metal 10.

As a result, it is possible to place the electrode terminal 3 closer to the end side 52aE side of the conductor pattern 52a by using the brazing filler metal 10 as compared with the conventional joining using solder. Therefore, the size of the conductor pattern 52a required for joining the electrode terminals 3 can be reduced as compared with the conventional case, and as a result, the entire power semiconductor device can be further reduced.

Embodiment 5 FIG.
FIG. 16 is a partial cross-sectional view showing a schematic configuration of power semiconductor device 105 in the fifth embodiment. The power semiconductor device 105 according to the fifth embodiment also basically has the same configuration as that of the power semiconductor device 101 according to the first embodiment, but differs in the following points. Therefore, here, mainly the differences will be described, and the description of the same components will be omitted. Note that FIG. 16 illustrates only the joint portion of the electrode terminal 3 in the power semiconductor device 105, and the other components are not illustrated.

In the first embodiment, in order to reduce the temperature difference between the electrode terminal 3 and the conductor pattern 52a due to heating of the laser beam, the electrode terminal 3 is configured to have a thickness larger than the thickness of the conductor pattern 52a. It has been described that the heat capacity per unit area is made larger than that of the conductor pattern 52a. In the fourth embodiment, when a brazing filler metal is used, in order to reduce contact thermal resistance at two locations between the electrode terminal 3 and the brazing filler metal 10 and between the brazing filler metal 10 and the conductor pattern 52a, the conductor pattern 52a is used. On the other hand, it is necessary to pressurize the electrode terminal 3, and when the tip 3 c (FIG. 13) and the root 3 d (FIG. 13) of the electrode terminal 3 are pressed with a jig, laser light is applied to the electrode terminal 3. It has been described that the electrode terminal 3 is deformed by thermal expansion of the electrode terminal 3 and remains deformed after brazing.

The power semiconductor device 105 in the fifth embodiment has a pressing structure that enables the pressing described in the fourth embodiment to be performed without using a jig. With this pressing structure, the electrode terminal 3 is forcibly pressed against the conductor pattern 52a.
For example, as shown in FIG. 16, a pressing structure is formed on the rising portion 36 that is not joined to the conductor pattern 52 a in the electrode terminal 3, and the electrode terminal 3 is given elasticity (spring property) to force In particular, the electrode terminal 3 is installed while being pressed against the conductor pattern 52a. In FIG. 16, as an example of the pressing structure, the rising portion 36 has, for example, a V-shaped curved shape.

By adopting such a structure, the electrode terminal 3 can be pressed against the conductor pattern 52a by the elastic force of the electrode terminal 3 itself without using a separate jig. Therefore, it is possible to reduce the contact thermal resistance at two locations between the electrode terminal 3 and the hard solder material 10 and between the hard solder material 10 and the conductor pattern 52a.

Further, by adopting the above-described pressing structure, the arrangement position of the electrode terminal 3 is not restricted by the jig for pressurizing the electrode terminal 3, so that the degree of freedom of arrangement of the electrode terminal 3 is improved. be able to. The pressing structure at the rising portion 36 of the electrode terminal 3 needs to be provided at a position that does not hinder laser light irradiation on the surface 3 a of the electrode terminal 3.
Moreover, the pressing structure in the rising part 36 of the electrode terminal 3 is not limited to the above-mentioned curved shape, For example, as shown to FIG. 17A and FIG. 17B, a U-shaped curved shape, an S-shaped curved shape, etc. may be sufficient. .

Furthermore, the pressing structure is not limited to that formed on the electrode terminal 3 itself. For example, when the electrode terminal 3 is supported by the case 12, as shown in FIG. 18, the pressing member 12 </ b> A that enables the tip 3 c of the electrode terminal 3 to be pressed against the conductor pattern 52 a is used as a part of the case 12. It can also be produced integrally with the case 12. Thus, the pressing structure can be provided on at least one of the electrode terminal 3 and the case 12. For example, the case 12 can be manufactured by injection molding or the like, and can be manufactured in the same process as the conventional case even if it has a complicated shape. Therefore, the pressing member 12A can also be formed integrally with the case 12.
In such a structure, it is possible to press the electrode terminal 3 without using a jig without forming the above-described pressing structure on the electrode terminal 3 itself. Therefore, the manufacturing cost of the power semiconductor device 105 can be reduced.

At this time, since the case 12 is made of a plastic material having a low heat-resistant temperature such as a thermoplastic resin, the pressing member 12A has the electrode terminal 3 placed at a position where the laser beam is not melted or deformed. It is necessary to press.
Further, in order to prevent the entire case 12, particularly the pressing member 12 </ b> A, from being melted and deformed by the heat generation of the electrode terminal 3 due to the laser beam, the pressing surface of the pressing member 12 </ b> A with respect to the electrode terminal 3 is preferably small, particularly one point like a protrusion If it is the shape which can pressurize with, it is more preferable.

It is also possible to adopt a configuration in which the above-described embodiments are combined, and it is also possible to combine components shown in different embodiments. Thereby, it can comprise so that there may exist each effect.

Although the present invention has been fully described in connection with preferred embodiments with reference to the accompanying drawings, various variations and modifications will be apparent to those skilled in the art. Such changes and modifications are to be understood as included within the scope of the present invention as long as they do not depart from the scope of the present invention.
In addition, Japanese Patent Application No. Japanese Patent Application No. 2016-110084 and Japanese Patent Application No. 1 filed on Nov. 24, 2016. The entire disclosure of each specification, drawings, claims, and abstract in Japanese Patent Application No. 2016-227966 is incorporated herein by reference.

1 power semiconductor element, 3 electrode terminal, 3j bonding surface, 5 insulating substrate,
6 Heat dissipation member, 8, 9 Solder, 10 Hard brazing material, 12 Case,
31 joints, 32 non-joints, 33 through holes, 34 voids,
42 laser equipment, 43 control equipment,
51 Insulating base material, 52a, 52b Conductor pattern,
101, 102, 104, 105 Power semiconductor devices.

Claims (17)

1. An insulating substrate having a conductor pattern on both sides of the insulating base;
   A semiconductor element mounted on one conductor pattern on the insulating substrate;
   In a semiconductor device comprising: an electrode terminal bonded to the one conductor pattern via a bonding material;
   The bonding material is a hard brazing material,
   The electrode terminal has a heat capacity per unit area larger than the heat capacity per unit area in the one conductor pattern.
   Semiconductor device.
2. An insulating substrate having a conductor pattern on both sides of the insulating base;
   A semiconductor element mounted on one conductor pattern on the insulating substrate;
   In a semiconductor device comprising: an electrode terminal bonded to the one conductor pattern via a bonding material;
   The bonding material is a hard brazing material,
   The electrode terminal has a thickness exceeding the thickness of the one conductor pattern,
   Semiconductor device.
3. An insulating substrate having a conductor pattern on both sides of the insulating base;
   A semiconductor element mounted on one conductor pattern on the insulating substrate;
   In a semiconductor device comprising: an electrode terminal bonded to the one conductor pattern via a bonding material;
   The bonding material is a hard brazing material,
   The electrode terminal has a joint portion having a first cross-sectional area and a non-joint portion having a second cross-sectional area larger than the first cross-sectional area, which are joined to the one conductor pattern by the brazing material.
   Semiconductor device.
4. 4. The semiconductor device according to claim 1, wherein a material of the electrode terminal is the same as a material of the conductor pattern.
5. The semiconductor device according to any one of claims 1 to 4, wherein a material of the electrode terminal is copper.
6. 4. The semiconductor device according to claim 1, wherein the electrode terminal joined to the conductor pattern via the brazing filler metal has a defect portion that is missing in a thickness direction. 5.
7. The semiconductor device according to claim 6, wherein the defect portion is a through hole.
8. The semiconductor device according to claim 6, wherein the defect portion is a groove extending from a tip of the electrode terminal toward a bent portion of the electrode terminal.
9. 9. The semiconductor device according to claim 6, wherein the defect portion is filled with the hard soldering material.
10. An insulating substrate having a conductor pattern on both surfaces of an insulating base, a semiconductor element mounted on one conductor pattern in the insulating substrate, and an electrode terminal, and a method for manufacturing a semiconductor device,
    A hard brazing material is provided for the one conductor pattern,
    The electrode terminal is arranged with the brazing material sandwiched between the one conductor pattern,
    Irradiating the surface of the electrode terminal corresponding to the brazing filler metal with an electromagnetic wave to heat the electrode terminal;
    A method for manufacturing a semiconductor device.
11. The manufacturing method according to claim 10, wherein the electromagnetic wave is a laser beam.
12. The manufacturing method according to claim 10, wherein the electromagnetic wave is an electron beam.
13. A part of the electrode terminal corresponding to the brazing filler metal is deficient, and the surface of the electrode terminal including the deficient part is irradiated with the electromagnetic wave to heat the electrode terminal. The production method according to claim 1.
14. 3. The semiconductor device according to claim 2, wherein the thickness of the electrode terminal is at least twice the thickness of the conductor pattern.
15. 3. The semiconductor device according to claim 2, wherein a cross-sectional area of the electrode terminal is twice or more a cross-sectional area of the conductor pattern in a portion facing the electrode terminal.
16. In the lower part of the electrode terminal corresponding to the bent part of the electrode terminal, the contact angle between the brazing filler metal and the electrode terminal is smaller than the contact angle in the part other than the lower part of the electrode terminal. The semiconductor device according to claim 1.
17. A case for supporting the electrode terminal;
    4. The semiconductor device according to claim 1, further comprising: a pressing structure that is provided on at least one of the electrode terminal and the case and presses the electrode terminal against the conductor pattern. 5.

Images